mirror/freebsd
1
0
Fork 0
mirror of https://git.FreeBSD.org/src.git synced 2024-12-24 11:29:10 +00:00
Code Issues Releases Activity
0562034333
freebsd/sys/dev/aic7xxx/aic7xxx.c

7431 lines
196 KiB
C
Raw Normal View History

Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Generic driver for the aic7xxx based adaptec SCSI controllers
* Product specific probe and attach routines can be found in:
* i386/eisa/ahc_eisa.c 27/284X and aic7770 motherboard controllers
* pci/ahc_pci.c 3985, 3980, 3940, 2940, aic7895, aic7890,
* aic7880, aic7870, aic7860, and aic7850 controllers
*
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
* Copyright (c) 1994, 1995, 1996, 1997, 1998, 1999, 2000 Justin T. Gibbs.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions, and the following disclaimer,
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
* without modification.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
* 2. The name of the author may not be used to endorse or promote products
* derived from this software without specific prior written permission.
*
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
* Alternatively, this software may be distributed under the terms of the
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
* GNU Public License ("GPL").
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE FOR
* ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
$Id$ -> $FreeBSD$
1999-08-28 01:08:13 +00:00
* $FreeBSD$
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
/*
* A few notes on features of the driver.
*
* SCB paging takes advantage of the fact that devices stay disconnected
* from the bus a relatively long time and that while they're disconnected,
* having the SCBs for these transactions down on the host adapter is of
* little use. Instead of leaving this idle SCB down on the card we copy
* it back up into kernel memory and reuse the SCB slot on the card to
* schedule another transaction. This can be a real payoff when doing random
* I/O to tagged queueing devices since there are more transactions active at
* once for the device to sort for optimal seek reduction. The algorithm goes
* like this...
*
* The sequencer maintains two lists of its hardware SCBs. The first is the
* singly linked free list which tracks all SCBs that are not currently in
* use. The second is the doubly linked disconnected list which holds the
* SCBs of transactions that are in the disconnected state sorted most
* recently disconnected first. When the kernel queues a transaction to
* the card, a hardware SCB to "house" this transaction is retrieved from
* either of these two lists. If the SCB came from the disconnected list,
* a check is made to see if any data transfer or SCB linking (more on linking
* in a bit) information has been changed since it was copied from the host
* and if so, DMAs the SCB back up before it can be used. Once a hardware
* SCB has been obtained, the SCB is DMAed from the host. Before any work
* can begin on this SCB, the sequencer must ensure that either the SCB is
* for a tagged transaction or the target is not already working on another
* non-tagged transaction. If a conflict arises in the non-tagged case, the
* sequencer finds the SCB for the active transactions and sets the SCB_LINKED
* field in that SCB to this next SCB to execute. To facilitate finding
* active non-tagged SCBs, the last four bytes of up to the first four hardware
* SCBs serve as a storage area for the currently active SCB ID for each
* target.
*
* When a device reconnects, a search is made of the hardware SCBs to find
* the SCB for this transaction. If the search fails, a hardware SCB is
* pulled from either the free or disconnected SCB list and the proper
* SCB is DMAed from the host. If the MK_MESSAGE control bit is set
* in the control byte of the SCB while it was disconnected, the sequencer
* will assert ATN and attempt to issue a message to the host.
*
* When a command completes, a check for non-zero status and residuals is
* made. If either of these conditions exists, the SCB is DMAed back up to
* the host so that it can interpret this information. Additionally, in the
* case of bad status, the sequencer generates a special interrupt and pauses
* itself. This allows the host to setup a request sense command if it
* chooses for this target synchronously with the error so that sense
* information isn't lost.
*
*/
#include <opt_aic7xxx.h>
#include <pci.h>
#include <stddef.h> /* For offsetof */
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/malloc.h>
Implement a new generic mechanism for attaching handler functions to events, in order to pave the way for removing a number of the ad-hoc implementations currently in use. Retire the at_shutdown family of functions and replace them with new event handler lists. Rework kern_shutdown.c to take greater advantage of the use of event handlers. Reviewed by: green
1999-08-21 06:24:40 +00:00
#include <sys/eventhandler.h>
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
#include <sys/proc.h>
#include <cam/cam.h>
#include <cam/cam_ccb.h>
#include <cam/cam_sim.h>
#include <cam/cam_xpt_sim.h>
#include <cam/cam_debug.h>
#include <cam/scsi/scsi_all.h>
#include <cam/scsi/scsi_message.h>
#if NPCI > 0
#include <machine/bus_memio.h>
#endif
#include <machine/bus_pio.h>
#include <machine/bus.h>
#include <machine/clock.h>
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
#include <machine/endian.h>
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
#include <sys/rman.h>
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
#include <vm/vm.h>
#include <vm/vm_param.h>
#include <vm/pmap.h>
#include <dev/aic7xxx/aic7xxx.h>
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
#include <dev/aic7xxx/aicasm_insformat.h>
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
#include <aic7xxx_reg.h>
#include <aic7xxx_seq.h>
#include <sys/kernel.h>
#ifndef AHC_TMODE_ENABLE
#define AHC_TMODE_ENABLE 0
#endif
#define MAX(a,b) (((a) > (b)) ? (a) : (b))
#define MIN(a,b) (((a) < (b)) ? (a) : (b))
#define ALL_CHANNELS '\0'
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
#define ALL_TARGETS_MASK 0xFFFF
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
#define INITIATOR_WILDCARD (~0)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
#define SIM_IS_SCSIBUS_B(ahc, sim) \
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
((sim) == ahc->sim_b)
#define SIM_CHANNEL(ahc, sim) \
(((sim) == ahc->sim_b) ? 'B' : 'A')
#define SIM_SCSI_ID(ahc, sim) \
(((sim) == ahc->sim_b) ? ahc->our_id_b : ahc->our_id)
#define SIM_PATH(ahc, sim) \
(((sim) == ahc->sim_b) ? ahc->path_b : ahc->path)
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
#define SCSIID_TARGET(ahc, scsiid) \
(((scsiid) & ((((ahc)->features & AHC_TWIN) != 0) ? TWIN_TID : TID)) \
>> TID_SHIFT)
#define SCSIID_OUR_ID(scsiid) \
((scsiid) & OID)
#define SCSIID_CHANNEL(ahc, scsiid) \
((((ahc)->features & AHC_TWIN) != 0) \
? ((((scsiid) & TWIN_CHNLB) != 0) ? 'B' : 'A') \
: 'A')
#define SCB_IS_SCSIBUS_B(ahc, scb) \
(SCSIID_CHANNEL(ahc, (scb)->hscb->scsiid) == 'B')
#define SCB_GET_OUR_ID(scb) \
SCSIID_OUR_ID((scb)->hscb->scsiid)
#define SCB_GET_TARGET(ahc, scb) \
SCSIID_TARGET((ahc), (scb)->hscb->scsiid)
#define SCB_GET_CHANNEL(ahc, scb) \
SCSIID_CHANNEL(ahc, (scb)->hscb->scsiid)
#define SCB_GET_LUN(scb) \
((scb)->hscb->lun)
#define SCB_GET_TARGET_OFFSET(ahc, scb) \
(SCB_GET_TARGET(ahc, scb) + (SCB_IS_SCSIBUS_B(ahc, scb) ? 8 : 0))
#define SCB_GET_TARGET_MASK(ahc, scb) \
(0x01 << (SCB_GET_TARGET_OFFSET(ahc, scb)))
#define TCL_TARGET_OFFSET(tcl) \
((((tcl) >> 4) & TID) >> 4)
aic7xxx.c: Correct the BUILD_TCL macro. It was placing the target id in the wrong bits. This was only an issue for adapters that do not perform SCB paging (aha-3940AUW for instance). Don't bother inlining ahc_index_busy_tcl. It is never used in a performance critical path and is a bit chunky. Correct ahc_index_busy_tcl to deal with "busy target tables" embedded in the latter half of 64byte SCBs. Don't initialize the busy target table to its empty state until after we have finished extracting configuration information from chip SRAM. In the common case of using 16 bytes of chip SRAM to do untagged target lookups, we were trashing the last 8 targets configuration data. (actually only target 8 because of the bug in the BUILD_TCL macro). Cram the "bus reset delivered" message back under bootverbose. Fix the cleanup of the SCB busy target table when aborting commands. If the lun is wildcarded, we must loop through all possible luns. aic7xxx.h: Only bother supporting 64 luns right now. It doesn't seem like either this driver or any peripherals will be doing information unit transfers (where the lun number is a 32 bit integer) any time soon. aic7xxx.seq: Fix support for the aic7895. We must flush the data FIFO if performing a manual transfer that is not a multiple of 8 bytes. We were doing this quite regularly for embedded cdbs. Manaually flush the fifo on earlier adapters when dealing with embedded cdbs too. We were stuffing the FIFO with 16 bytes instead, but triggering the flush is more efficient and allows us to remove two instructions from the "copy_to_fifo" routine.
2000-07-24 22:27:40 +00:00
#define TCL_LUN(tcl) \
(tcl & (AHC_NUM_LUNS - 1))
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
#define BUILD_TCL(scsiid, lun) \
aic7xxx.c: Correct the BUILD_TCL macro. It was placing the target id in the wrong bits. This was only an issue for adapters that do not perform SCB paging (aha-3940AUW for instance). Don't bother inlining ahc_index_busy_tcl. It is never used in a performance critical path and is a bit chunky. Correct ahc_index_busy_tcl to deal with "busy target tables" embedded in the latter half of 64byte SCBs. Don't initialize the busy target table to its empty state until after we have finished extracting configuration information from chip SRAM. In the common case of using 16 bytes of chip SRAM to do untagged target lookups, we were trashing the last 8 targets configuration data. (actually only target 8 because of the bug in the BUILD_TCL macro). Cram the "bus reset delivered" message back under bootverbose. Fix the cleanup of the SCB busy target table when aborting commands. If the lun is wildcarded, we must loop through all possible luns. aic7xxx.h: Only bother supporting 64 luns right now. It doesn't seem like either this driver or any peripherals will be doing information unit transfers (where the lun number is a 32 bit integer) any time soon. aic7xxx.seq: Fix support for the aic7895. We must flush the data FIFO if performing a manual transfer that is not a multiple of 8 bytes. We were doing this quite regularly for embedded cdbs. Manaually flush the fifo on earlier adapters when dealing with embedded cdbs too. We were stuffing the FIFO with 16 bytes instead, but triggering the flush is more efficient and allows us to remove two instructions from the "copy_to_fifo" routine.
2000-07-24 22:27:40 +00:00
((lun) | (((scsiid) & TID) << 4))
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
#define BUILD_SCSIID(ahc, sim, target_id, our_id) \
((((target_id) << TID_SHIFT) & TID) | (our_id) \
| (SIM_IS_SCSIBUS_B(ahc, sim) ? TWIN_CHNLB : 0))
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
#define ccb_scb_ptr spriv_ptr0
#define ccb_ahc_ptr spriv_ptr1
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
char *ahc_chip_names[] =
{
"NONE",
"aic7770",
"aic7850",
"aic7855",
"aic7859",
"aic7860",
"aic7870",
"aic7880",
"aic7895",
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
"aic7890/91",
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
"aic7896/97",
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
"aic7892",
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
"aic7899"
};
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
typedef enum {
ROLE_UNKNOWN,
ROLE_INITIATOR,
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
ROLE_TARGET
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
} role_t;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
struct ahc_devinfo {
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
int our_scsiid;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
int target_offset;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
uint16_t target_mask;
uint8_t target;
uint8_t lun;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
char channel;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
role_t role; /*
* Only guaranteed to be correct if not
* in the busfree state.
*/
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
};
typedef enum {
SEARCH_COMPLETE,
SEARCH_COUNT,
SEARCH_REMOVE
} ahc_search_action;
#ifdef AHC_DEBUG
static int ahc_debug = AHC_DEBUG;
#endif
#if NPCI > 0
void ahc_pci_intr(struct ahc_softc *ahc);
#endif
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
static int ahcinitscbdata(struct ahc_softc *ahc);
static void ahcfiniscbdata(struct ahc_softc *ahc);
static bus_dmamap_callback_t ahcdmamapcb;
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
#if UNUSED
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static void ahc_dump_targcmd(struct target_cmd *cmd);
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
#endif
Implement a new generic mechanism for attaching handler functions to events, in order to pave the way for removing a number of the ad-hoc implementations currently in use. Retire the at_shutdown family of functions and replace them with new event handler lists. Rework kern_shutdown.c to take greater advantage of the use of event handlers. Reviewed by: green
1999-08-21 06:24:40 +00:00
static void ahc_shutdown(void *arg, int howto);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static cam_status
ahc_find_tmode_devs(struct ahc_softc *ahc,
struct cam_sim *sim, union ccb *ccb,
struct tmode_tstate **tstate,
struct tmode_lstate **lstate,
int notfound_failure);
static void ahc_action(struct cam_sim *sim, union ccb *ccb);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
static void ahc_async(void *callback_arg, uint32_t code,
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
struct cam_path *path, void *arg);
static void ahc_execute_scb(void *arg, bus_dma_segment_t *dm_segs,
int nsegments, int error);
static void ahc_poll(struct cam_sim *sim);
static void ahc_setup_data(struct ahc_softc *ahc,
struct ccb_scsiio *csio, struct scb *scb);
static void ahc_freeze_devq(struct ahc_softc *ahc, struct cam_path *path);
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
static void ahcallocscbs(struct ahc_softc *ahc);
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
#if UNUSED
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
static void ahc_scb_devinfo(struct ahc_softc *ahc,
struct ahc_devinfo *devinfo,
struct scb *scb);
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
#endif
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static void ahc_fetch_devinfo(struct ahc_softc *ahc,
struct ahc_devinfo *devinfo);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
static void ahc_compile_devinfo(struct ahc_devinfo *devinfo, u_int our_id,
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
u_int target, u_int lun, char channel,
role_t role);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
static u_int ahc_rem_wscb(struct ahc_softc *ahc, u_int scbpos, u_int prev);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static void ahc_done(struct ahc_softc *ahc, struct scb *scbp);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
static struct tmode_tstate *
ahc_alloc_tstate(struct ahc_softc *ahc,
u_int scsi_id, char channel);
static void ahc_free_tstate(struct ahc_softc *ahc,
u_int scsi_id, char channel, int force);
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
static void ahc_handle_en_lun(struct ahc_softc *ahc, struct cam_sim *sim,
union ccb *ccb);
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
static void ahc_update_scsiid(struct ahc_softc *ahc, u_int targid_mask);
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
static int ahc_handle_target_cmd(struct ahc_softc *ahc,
Change the delivery mechanism for incoming target commands. We now use a 256 entry ring buffer of descriptersfor this purpose. This allows the use of a simple 8bit counter in the sequencer code for tracking start location. Entries in the ring buffer now contain a "cmd_valid" byte at their tail. As an entry is serviced, this byte is cleared by the kernel and set by the sequencer during its dma of a new entry. Since this byte is the last portion of the command touched during a dma, the kernel can use this byte to ensure the command it processes is completely valid. The new command format requires a fixed sized DMA from the controller to deliver which allowed for additional simplification of the sequencer code. The hack that required 1 SCB slot to be stolen for incoming command delivery notification is also gone.
1998-11-23 01:33:47 +00:00
struct target_cmd *cmd);
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
static void ahc_handle_seqint(struct ahc_softc *ahc, u_int intstat);
static void ahc_handle_scsiint(struct ahc_softc *ahc, u_int intstat);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
static void ahc_build_transfer_msg(struct ahc_softc *ahc,
struct ahc_devinfo *devinfo);
static void ahc_setup_initiator_msgout(struct ahc_softc *ahc,
struct ahc_devinfo *devinfo,
struct scb *scb);
static void ahc_setup_target_msgin(struct ahc_softc *ahc,
struct ahc_devinfo *devinfo);
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
static int ahc_handle_msg_reject(struct ahc_softc *ahc,
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
struct ahc_devinfo *devinfo);
static void ahc_clear_msg_state(struct ahc_softc *ahc);
static void ahc_handle_message_phase(struct ahc_softc *ahc,
struct cam_path *path);
static int ahc_sent_msg(struct ahc_softc *ahc, u_int msgtype, int full);
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
typedef enum {
MSGLOOP_IN_PROG,
MSGLOOP_MSGCOMPLETE,
MSGLOOP_TERMINATED
} msg_loop_stat;
static int ahc_parse_msg(struct ahc_softc *ahc, struct cam_path *path,
struct ahc_devinfo *devinfo);
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
static void ahc_handle_ign_wide_residue(struct ahc_softc *ahc,
struct ahc_devinfo *devinfo);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
static void ahc_handle_devreset(struct ahc_softc *ahc,
struct ahc_devinfo *devinfo,
cam_status status, ac_code acode,
char *message,
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
int verbose_level);
#ifdef AHC_DUMP_SEQ
static void ahc_dumpseq(struct ahc_softc *ahc);
#endif
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static void ahc_loadseq(struct ahc_softc *ahc);
static int ahc_check_patch(struct ahc_softc *ahc,
struct patch **start_patch,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int start_instr, u_int *skip_addr);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static void ahc_download_instr(struct ahc_softc *ahc,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int instrptr, uint8_t *dconsts);
static int ahc_match_scb(struct ahc_softc *ahc, struct scb *scb,
int target, char channel, int lun, u_int tag,
role_t role);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
#ifdef AHC_DEBUG
static void ahc_print_scb(struct scb *scb);
#endif
static int ahc_search_qinfifo(struct ahc_softc *ahc, int target,
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
char channel, int lun, u_int tag,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
role_t role, uint32_t status,
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
ahc_search_action action);
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
static void ahc_abort_ccb(struct ahc_softc *ahc, struct cam_sim *sim,
union ccb *ccb);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static int ahc_reset_channel(struct ahc_softc *ahc, char channel,
int initiate_reset);
static int ahc_abort_scbs(struct ahc_softc *ahc, int target,
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
char channel, int lun, u_int tag, role_t role,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
uint32_t status);
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
static int ahc_search_disc_list(struct ahc_softc *ahc, int target,
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
char channel, int lun, u_int tag,
int stop_on_first, int remove,
int save_state);
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
static u_int ahc_rem_scb_from_disc_list(struct ahc_softc *ahc,
u_int prev, u_int scbptr);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static void ahc_add_curscb_to_free_list(struct ahc_softc *ahc);
static void ahc_clear_intstat(struct ahc_softc *ahc);
static void ahc_reset_current_bus(struct ahc_softc *ahc);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
static struct ahc_syncrate *
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahc_devlimited_syncrate(struct ahc_softc *ahc, u_int *period,
u_int *ppr_options);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static struct ahc_syncrate *
ahc_find_syncrate(struct ahc_softc *ahc, u_int *period,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int *ppr_options, u_int maxsync);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static u_int ahc_find_period(struct ahc_softc *ahc, u_int scsirate,
u_int maxsync);
static void ahc_validate_offset(struct ahc_softc *ahc,
struct ahc_syncrate *syncrate,
u_int *offset, int wide);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
static void ahc_validate_width(struct ahc_softc *ahc, u_int *bus_width);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
static void ahc_update_target_msg_request(struct ahc_softc *ahc,
struct ahc_devinfo *devinfo,
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
struct ahc_initiator_tinfo *tinfo,
Use TCL_CHANNEL/SCSI_ID/etc. macros in more locations. Don't mess with the IRQMS bit in the host control register unless we are an aic7770 chip. Use calling context to determine if the card is already paused when we update the target message request bit field in controller scratch ram. Looking at the paused bit in the HCNTRL register opened up a race condition. Insert delays in the target message request update routine as a temporary work around for what looks like a chip bug. I'm still investigating this one. Fix the Abort/Abort Tag/BDR handler to pull its message from the message buffer in our softc instead of attempting to get it from a register on the controller. The message is never recorded by the controller in the new message scheme. Don't rely on having an SCB when a BDR occurs. We can issue these during invalid reconnects to. Fix a few cases where we were restarting the sequencer but then still falling out of a switch statement to unpause the sequencer again. This could cause us to mess up sequencer state if it generated another pausing interrupt between the time of the restart and unpause. Kill the 'transceiver settle' loop during card initialization. I failed to realize that a controller that is not connected to any cables will never settle or enable the SCSI transceivers at all. The correct solution is to monitor the IOERR interrupt which indicates that the transceiver state has changed (UW<->LVD). Modify the aic7xxx assembler to properly echo input when stdin is not a tty.
1999-03-23 07:24:29 +00:00
int force, int paused);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
static int ahc_create_path(struct ahc_softc *ahc,
struct ahc_devinfo *devinfo,
struct cam_path **path);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static void ahc_set_syncrate(struct ahc_softc *ahc,
struct ahc_devinfo *devinfo,
struct cam_path *path,
struct ahc_syncrate *syncrate,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int period, u_int offset,
u_int ppr_options, u_int type,
Correct a bug where the chip could be unpaused in the middle of a bus or device reset error recovery operation.
1999-05-18 03:58:49 +00:00
int paused);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static void ahc_set_width(struct ahc_softc *ahc,
struct ahc_devinfo *devinfo,
Correct a bug where the chip could be unpaused in the middle of a bus or device reset error recovery operation.
1999-05-18 03:58:49 +00:00
struct cam_path *path, u_int width, u_int type,
int paused);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
static void ahc_set_tags(struct ahc_softc *ahc,
struct ahc_devinfo *devinfo,
int enable);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static void ahc_construct_sdtr(struct ahc_softc *ahc,
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
u_int period, u_int offset);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
static void ahc_construct_wdtr(struct ahc_softc *ahc, u_int bus_width);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
static void ahc_construct_ppr(struct ahc_softc *ahc, u_int period,
u_int offset, u_int bus_width,
u_int ppr_options);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
static __inline int ahc_check_residual(struct scb *scb);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static void ahc_calc_residual(struct scb *scb);
static void ahc_update_pending_syncrates(struct ahc_softc *ahc);
static void ahc_set_recoveryscb(struct ahc_softc *ahc, struct scb *scb);
static timeout_t
ahc_timeout;
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
static void ahc_queue_lstate_event(struct ahc_softc *ahc,
struct tmode_lstate *lstate,
u_int initiator_id, u_int event_type,
u_int event_arg);
static void ahc_send_lstate_events(struct ahc_softc *ahc,
struct tmode_lstate *lstate);
aic7xxx.c: Correct the BUILD_TCL macro. It was placing the target id in the wrong bits. This was only an issue for adapters that do not perform SCB paging (aha-3940AUW for instance). Don't bother inlining ahc_index_busy_tcl. It is never used in a performance critical path and is a bit chunky. Correct ahc_index_busy_tcl to deal with "busy target tables" embedded in the latter half of 64byte SCBs. Don't initialize the busy target table to its empty state until after we have finished extracting configuration information from chip SRAM. In the common case of using 16 bytes of chip SRAM to do untagged target lookups, we were trashing the last 8 targets configuration data. (actually only target 8 because of the bug in the BUILD_TCL macro). Cram the "bus reset delivered" message back under bootverbose. Fix the cleanup of the SCB busy target table when aborting commands. If the lun is wildcarded, we must loop through all possible luns. aic7xxx.h: Only bother supporting 64 luns right now. It doesn't seem like either this driver or any peripherals will be doing information unit transfers (where the lun number is a 32 bit integer) any time soon. aic7xxx.seq: Fix support for the aic7895. We must flush the data FIFO if performing a manual transfer that is not a multiple of 8 bytes. We were doing this quite regularly for embedded cdbs. Manaually flush the fifo on earlier adapters when dealing with embedded cdbs too. We were stuffing the FIFO with 16 bytes instead, but triggering the flush is more efficient and allows us to remove two instructions from the "copy_to_fifo" routine.
2000-07-24 22:27:40 +00:00
static void restart_sequencer(struct ahc_softc *ahc);
static u_int ahc_index_busy_tcl(struct ahc_softc *ahc,
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
u_int tcl, int unbusy);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static __inline void ahc_freeze_ccb(union ccb* ccb);
static __inline cam_status ahc_ccb_status(union ccb* ccb);
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
static __inline void ahcsetccbstatus(union ccb* ccb,
cam_status status);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
static void ahc_run_untagged_queues(struct ahc_softc *);
static void ahc_run_untagged_queue(struct ahc_softc *,
struct scb_tailq *);
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
static void ahc_run_tqinfifo(struct ahc_softc *ahc, int paused);
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
static void ahc_run_qoutfifo(struct ahc_softc *ahc);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
static __inline struct ahc_initiator_tinfo *
ahc_fetch_transinfo(struct ahc_softc *ahc,
char channel,
u_int our_id, u_int target,
struct tmode_tstate **tstate);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
static __inline struct ahc_dma_seg *
ahc_sg_bus_to_virt(struct scb *scb,
uint32_t sg_busaddr);
static __inline uint32_t
ahc_sg_virt_to_bus(struct scb *scb,
struct ahc_dma_seg *sg);
static __inline void ahc_queue_scb(struct ahc_softc *ahc,
struct scb *scb);
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
static void ahcfreescb(struct ahc_softc *ahc, struct scb *scb);
static __inline struct scb *ahcgetscb(struct ahc_softc *ahc);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
static __inline void ahc_freeze_untagged_queues(struct ahc_softc *ahc);
static __inline void ahc_release_untagged_queues(struct ahc_softc *ahc);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
static __inline uint32_t
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_hscb_busaddr(struct ahc_softc *ahc, u_int index)
{
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
return (ahc->scb_data->hscb_busaddr
+ (sizeof(struct hardware_scb) * index));
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
#define AHC_BUSRESET_DELAY 250 /* Reset delay in us */
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Restart the sequencer program from address zero
*/
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
static void
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
restart_sequencer(struct ahc_softc *ahc)
{
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
u_int i;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
pause_sequencer(ahc);
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
/*
* Everytime we restart the sequencer, there
* is the possiblitity that we have restarted
* within a three instruction window where an
* SCB has been marked free but has not made it
* onto the free list. Since SCSI events(bus reset,
* unexpected bus free) will always freeze the
* sequencer, we cannot close this window. To
* avoid losing an SCB, we reconsitute the free
* list every time we restart the sequencer.
*/
ahc_outb(ahc, FREE_SCBH, SCB_LIST_NULL);
for (i = 0; i < ahc->scb_data->maxhscbs; i++) {
ahc_outb(ahc, SCBPTR, i);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (ahc_inb(ahc, SCB_TAG) == SCB_LIST_NULL) {
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
ahc_add_curscb_to_free_list(ahc);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
}
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_outb(ahc, SEQCTL, FASTMODE|SEQRESET);
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
unpause_sequencer(ahc);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
aic7xxx.c: Correct the BUILD_TCL macro. It was placing the target id in the wrong bits. This was only an issue for adapters that do not perform SCB paging (aha-3940AUW for instance). Don't bother inlining ahc_index_busy_tcl. It is never used in a performance critical path and is a bit chunky. Correct ahc_index_busy_tcl to deal with "busy target tables" embedded in the latter half of 64byte SCBs. Don't initialize the busy target table to its empty state until after we have finished extracting configuration information from chip SRAM. In the common case of using 16 bytes of chip SRAM to do untagged target lookups, we were trashing the last 8 targets configuration data. (actually only target 8 because of the bug in the BUILD_TCL macro). Cram the "bus reset delivered" message back under bootverbose. Fix the cleanup of the SCB busy target table when aborting commands. If the lun is wildcarded, we must loop through all possible luns. aic7xxx.h: Only bother supporting 64 luns right now. It doesn't seem like either this driver or any peripherals will be doing information unit transfers (where the lun number is a 32 bit integer) any time soon. aic7xxx.seq: Fix support for the aic7895. We must flush the data FIFO if performing a manual transfer that is not a multiple of 8 bytes. We were doing this quite regularly for embedded cdbs. Manaually flush the fifo on earlier adapters when dealing with embedded cdbs too. We were stuffing the FIFO with 16 bytes instead, but triggering the flush is more efficient and allows us to remove two instructions from the "copy_to_fifo" routine.
2000-07-24 22:27:40 +00:00
static u_int
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
ahc_index_busy_tcl(struct ahc_softc *ahc, u_int tcl, int unbusy)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
u_int scbid;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int target_offset;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
aic7xxx.c: Correct the BUILD_TCL macro. It was placing the target id in the wrong bits. This was only an issue for adapters that do not perform SCB paging (aha-3940AUW for instance). Don't bother inlining ahc_index_busy_tcl. It is never used in a performance critical path and is a bit chunky. Correct ahc_index_busy_tcl to deal with "busy target tables" embedded in the latter half of 64byte SCBs. Don't initialize the busy target table to its empty state until after we have finished extracting configuration information from chip SRAM. In the common case of using 16 bytes of chip SRAM to do untagged target lookups, we were trashing the last 8 targets configuration data. (actually only target 8 because of the bug in the BUILD_TCL macro). Cram the "bus reset delivered" message back under bootverbose. Fix the cleanup of the SCB busy target table when aborting commands. If the lun is wildcarded, we must loop through all possible luns. aic7xxx.h: Only bother supporting 64 luns right now. It doesn't seem like either this driver or any peripherals will be doing information unit transfers (where the lun number is a 32 bit integer) any time soon. aic7xxx.seq: Fix support for the aic7895. We must flush the data FIFO if performing a manual transfer that is not a multiple of 8 bytes. We were doing this quite regularly for embedded cdbs. Manaually flush the fifo on earlier adapters when dealing with embedded cdbs too. We were stuffing the FIFO with 16 bytes instead, but triggering the flush is more efficient and allows us to remove two instructions from the "copy_to_fifo" routine.
2000-07-24 22:27:40 +00:00
if ((ahc->features & AHC_SCB_BTT) != 0) {
u_int saved_scbptr;
saved_scbptr = ahc_inb(ahc, SCBPTR);
ahc_outb(ahc, SCBPTR, TCL_LUN(tcl));
scbid = ahc_inb(ahc, SCB_64_BTT + TCL_TARGET_OFFSET(tcl));
if (unbusy)
ahc_outb(ahc, SCB_64_BTT + TCL_TARGET_OFFSET(tcl),
SCB_LIST_NULL);
ahc_outb(ahc, SCBPTR, saved_scbptr);
} else {
target_offset = TCL_TARGET_OFFSET(tcl);
scbid = ahc_inb(ahc, BUSY_TARGETS + target_offset);
if (unbusy)
ahc_outb(ahc, BUSY_TARGETS + target_offset,
SCB_LIST_NULL);
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
return (scbid);
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
static __inline int
ahc_check_residual(struct scb *scb)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
struct status_pkt *sp;
sp = &scb->hscb->shared_data.status;
if ((scb->hscb->sgptr & SG_RESID_VALID) != 0)
return (1);
return (0);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
static __inline void
ahc_freeze_ccb(union ccb* ccb)
{
if ((ccb->ccb_h.status & CAM_DEV_QFRZN) == 0) {
ccb->ccb_h.status |= CAM_DEV_QFRZN;
xpt_freeze_devq(ccb->ccb_h.path, /*count*/1);
}
}
static __inline cam_status
ahc_ccb_status(union ccb* ccb)
{
return (ccb->ccb_h.status & CAM_STATUS_MASK);
}
static __inline void
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
ahcsetccbstatus(union ccb* ccb, cam_status status)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
ccb->ccb_h.status &= ~CAM_STATUS_MASK;
ccb->ccb_h.status |= status;
}
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
static __inline struct ahc_initiator_tinfo *
ahc_fetch_transinfo(struct ahc_softc *ahc, char channel, u_int our_id,
u_int remote_id, struct tmode_tstate **tstate)
{
/*
* Transfer data structures are stored from the perspective
* of the target role. Since the parameters for a connection
* in the initiator role to a given target are the same as
* when the roles are reversed, we pretend we are the target.
*/
if (channel == 'B')
our_id += 8;
*tstate = ahc->enabled_targets[our_id];
return (&(*tstate)->transinfo[remote_id]);
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
static __inline struct ahc_dma_seg *
ahc_sg_bus_to_virt(struct scb *scb, uint32_t sg_busaddr)
{
int sg_index;
sg_index = (sg_busaddr - scb->sg_list_phys)/sizeof(struct ahc_dma_seg);
/* sg_list_phys points to entry 1, not 0 */
sg_index++;
return (&scb->sg_list[sg_index]);
}
static __inline uint32_t
ahc_sg_virt_to_bus(struct scb *scb, struct ahc_dma_seg *sg)
{
int sg_index;
/* sg_list_phys points to entry 1, not 0 */
sg_index = sg - &scb->sg_list[1];
return (scb->sg_list_phys + (sg_index * sizeof(*scb->sg_list)));
}
static __inline void
ahc_queue_scb(struct ahc_softc *ahc, struct scb *scb)
{
ahc->qinfifo[ahc->qinfifonext++] = scb->hscb->tag;
if ((ahc->features & AHC_QUEUE_REGS) != 0) {
ahc_outb(ahc, HNSCB_QOFF, ahc->qinfifonext);
} else {
pause_sequencer(ahc);
ahc_outb(ahc, KERNEL_QINPOS, ahc->qinfifonext);
unpause_sequencer(ahc);
}
}
static __inline void
ahc_freeze_untagged_queues(struct ahc_softc *ahc)
{
if ((ahc->features & AHC_SCB_BTT) == 0)
ahc->untagged_queue_lock++;
}
static __inline void
ahc_release_untagged_queues(struct ahc_softc *ahc)
{
if ((ahc->features & AHC_SCB_BTT) == 0) {
ahc->untagged_queue_lock--;
if (ahc->untagged_queue_lock == 0)
ahc_run_untagged_queues(ahc);
}
}
static void
ahc_run_untagged_queues(struct ahc_softc *ahc)
{
int i;
for (i = 0; i < 16; i++)
ahc_run_untagged_queue(ahc, &ahc->untagged_queues[i]);
}
static void
ahc_run_untagged_queue(struct ahc_softc *ahc, struct scb_tailq *queue)
{
struct scb *scb;
if (ahc->untagged_queue_lock != 0)
return;
if ((scb = TAILQ_FIRST(queue)) != NULL
&& (scb->flags & SCB_ACTIVE) == 0) {
scb->flags |= SCB_ACTIVE;
ahc_queue_scb(ahc, scb);
}
}
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
static void
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
ahc_run_tqinfifo(struct ahc_softc *ahc, int paused)
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
{
struct target_cmd *cmd;
while ((cmd = &ahc->targetcmds[ahc->tqinfifonext])->cmd_valid != 0) {
/*
* Only advance through the queue if we
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
* have the resources to process the command.
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
*/
if (ahc_handle_target_cmd(ahc, cmd) != 0)
break;
ahc->tqinfifonext++;
cmd->cmd_valid = 0;
/*
* Lazily update our position in the target mode incomming
* command queue as seen by the sequencer.
*/
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
if ((ahc->tqinfifonext & (HOST_TQINPOS - 1)) == 1) {
if ((ahc->features & AHC_HS_MAILBOX) != 0) {
u_int hs_mailbox;
hs_mailbox = ahc_inb(ahc, HS_MAILBOX);
hs_mailbox &= ~HOST_TQINPOS;
hs_mailbox |= ahc->tqinfifonext & HOST_TQINPOS;
ahc_outb(ahc, HS_MAILBOX, hs_mailbox);
} else {
if (!paused)
pause_sequencer(ahc);
ahc_outb(ahc, KERNEL_TQINPOS,
ahc->tqinfifonext & HOST_TQINPOS);
if (!paused)
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
unpause_sequencer(ahc);
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
}
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
}
}
}
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
static void
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
ahc_run_qoutfifo(struct ahc_softc *ahc)
{
struct scb *scb;
u_int scb_index;
while (ahc->qoutfifo[ahc->qoutfifonext] != SCB_LIST_NULL) {
scb_index = ahc->qoutfifo[ahc->qoutfifonext];
ahc->qoutfifo[ahc->qoutfifonext++] = SCB_LIST_NULL;
scb = &ahc->scb_data->scbarray[scb_index];
if (scb_index >= ahc->scb_data->numscbs
|| (scb->flags & SCB_ACTIVE) == 0) {
printf("%s: WARNING no command for scb %d "
"(cmdcmplt)\nQOUTPOS = %d\n",
ahc_name(ahc), scb_index,
ahc->qoutfifonext - 1);
continue;
}
/*
* Save off the residual
* if there is one.
*/
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (ahc_check_residual(scb) != 0)
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
ahc_calc_residual(scb);
else
scb->ccb->csio.resid = 0;
ahc_done(ahc, scb);
}
}
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
/*
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
* Return an SCB resource to the free list.
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
*/
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
static void
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
ahcfreescb(struct ahc_softc *ahc, struct scb *scb)
{
struct hardware_scb *hscb;
int opri;
hscb = scb->hscb;
opri = splcam();
if ((ahc->flags & AHC_RESOURCE_SHORTAGE) != 0
&& (scb->ccb->ccb_h.status & CAM_RELEASE_SIMQ) == 0) {
scb->ccb->ccb_h.status |= CAM_RELEASE_SIMQ;
ahc->flags &= ~AHC_RESOURCE_SHORTAGE;
}
/* Clean up for the next user */
scb->flags = SCB_FREE;
hscb->control = 0;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
SLIST_INSERT_HEAD(&ahc->scb_data->free_scbs, scb, links.sle);
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
splx(opri);
}
/*
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
* Get a free scb. If there are none, see if we can allocate a new SCB.
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
*/
static __inline struct scb *
ahcgetscb(struct ahc_softc *ahc)
{
struct scb *scbp;
int opri;
opri = splcam();
if ((scbp = SLIST_FIRST(&ahc->scb_data->free_scbs))) {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
SLIST_REMOVE_HEAD(&ahc->scb_data->free_scbs, links.sle);
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
} else {
ahcallocscbs(ahc);
scbp = SLIST_FIRST(&ahc->scb_data->free_scbs);
if (scbp != NULL)
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
SLIST_REMOVE_HEAD(&ahc->scb_data->free_scbs, links.sle);
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
}
splx(opri);
return (scbp);
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
char *
ahc_name(struct ahc_softc *ahc)
{
static char name[10];
Examine all occurrences of sprintf(), strcat(), and str[n]cpy() for possible buffer overflow problems. Replaced most sprintf()'s with snprintf(); for others cases, added terminating NUL bytes where appropriate, replaced constants like "16" with sizeof(), etc. These changes include several bug fixes, but most changes are for maintainability's sake. Any instance where it wasn't "immediately obvious" that a buffer overflow could not occur was made safer. Reviewed by: Bruce Evans <bde@zeta.org.au> Reviewed by: Matthew Dillon <dillon@apollo.backplane.com> Reviewed by: Mike Spengler <mks@networkcs.com>
1998-12-04 22:54:57 +00:00
snprintf(name, sizeof(name), "ahc%d", ahc->unit);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
return (name);
}
#ifdef AHC_DEBUG
static void
ahc_print_scb(struct scb *scb)
{
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
int i;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
struct hardware_scb *hscb = scb->hscb;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
printf("scb:%p control:0x%x scsiid:0x%x lun:%d cdb_len:%d\n",
scb,
hscb->control,
hscb->scsiid,
hscb->lun,
hscb->cdb_len);
i = 0;
printf("Shared Data: %#02x %#02x %#02x %#02x\n",
hscb->shared_data.cdb[i++],
hscb->shared_data.cdb[i++],
hscb->shared_data.cdb[i++],
hscb->shared_data.cdb[i++]);
printf(" %#02x %#02x %#02x %#02x\n",
hscb->shared_data.cdb[i++],
hscb->shared_data.cdb[i++],
hscb->shared_data.cdb[i++],
hscb->shared_data.cdb[i++]);
printf(" %#02x %#02x %#02x %#02x\n",
hscb->shared_data.cdb[i++],
hscb->shared_data.cdb[i++],
hscb->shared_data.cdb[i++],
hscb->shared_data.cdb[i++]);
printf(" dataptr:%#x datacnt:%#x sgptr:%#x tag:%#x\n",
hscb->dataptr,
hscb->datacnt,
hscb->sgptr,
hscb->tag);
if (scb->sg_count > 0) {
for (i = 0; i < scb->sg_count; i++) {
printf("sg[%d] - Addr 0x%x : Length %d\n",
i,
scb->sg_list[i].addr,
scb->sg_list[i].len);
}
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
#endif
static struct {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
uint8_t errno;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
char *errmesg;
} hard_error[] = {
{ ILLHADDR, "Illegal Host Access" },
{ ILLSADDR, "Illegal Sequencer Address referrenced" },
{ ILLOPCODE, "Illegal Opcode in sequencer program" },
{ SQPARERR, "Sequencer Parity Error" },
{ DPARERR, "Data-path Parity Error" },
{ MPARERR, "Scratch or SCB Memory Parity Error" },
{ PCIERRSTAT, "PCI Error detected" },
{ CIOPARERR, "CIOBUS Parity Error" },
};
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
static const int num_errors = sizeof(hard_error)/sizeof(hard_error[0]);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
static struct {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
uint8_t phase;
uint8_t mesg_out; /* Message response to parity errors */
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
char *phasemsg;
} phase_table[] = {
{ P_DATAOUT, MSG_NOOP, "in Data-out phase" },
{ P_DATAIN, MSG_INITIATOR_DET_ERR, "in Data-in phase" },
{ P_COMMAND, MSG_NOOP, "in Command phase" },
{ P_MESGOUT, MSG_NOOP, "in Message-out phase" },
{ P_STATUS, MSG_INITIATOR_DET_ERR, "in Status phase" },
{ P_MESGIN, MSG_PARITY_ERROR, "in Message-in phase" },
{ P_BUSFREE, MSG_NOOP, "while idle" },
{ 0, MSG_NOOP, "in unknown phase" }
};
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
static const u_int num_phases =
(sizeof(phase_table)/sizeof(phase_table[0])) - 1;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Valid SCSIRATE values. (p. 3-17)
* Provides a mapping of tranfer periods in ns to the proper value to
* stick in the scsiscfr reg to use that transfer rate.
*/
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
#define AHC_SYNCRATE_DT 0
#define AHC_SYNCRATE_ULTRA2 1
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
#define AHC_SYNCRATE_ULTRA 3
#define AHC_SYNCRATE_FAST 6
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static struct ahc_syncrate ahc_syncrates[] = {
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
/* ultra2 fast/ultra period rate */
{ 0x42, 0x000, 9, "80.0" },
{ 0x03, 0x000, 10, "40.0" },
{ 0x04, 0x000, 11, "33.0" },
{ 0x05, 0x100, 12, "20.0" },
{ 0x06, 0x110, 15, "16.0" },
{ 0x07, 0x120, 18, "13.4" },
{ 0x08, 0x000, 25, "10.0" },
{ 0x19, 0x010, 31, "8.0" },
{ 0x1a, 0x020, 37, "6.67" },
{ 0x1b, 0x030, 43, "5.7" },
{ 0x1c, 0x040, 50, "5.0" },
{ 0x00, 0x050, 56, "4.4" },
{ 0x00, 0x060, 62, "4.0" },
{ 0x00, 0x070, 68, "3.6" },
{ 0x00, 0x000, 0, NULL }
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
};
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
void
ahc_init_probe_config(struct ahc_probe_config *probe_config)
{
probe_config->description = NULL;
probe_config->channel = 'A';
probe_config->channel_b = 'B';
probe_config->chip = AHC_NONE;
probe_config->features = AHC_FENONE;
probe_config->bugs = AHC_BUGNONE;
probe_config->flags = AHC_FNONE;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Allocate a controller structure for a new device and initialize it.
*/
struct ahc_softc *
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
ahc_alloc(device_t dev, struct resource *regs, int regs_type, int regs_id,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
bus_dma_tag_t parent_dmat, struct ahc_probe_config *config,
struct scb_data *scb_data)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
/*
* find unit and check we have that many defined
*/
struct ahc_softc *ahc;
size_t alloc_size;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
int i;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
* Allocate a storage area for us.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
if (scb_data == NULL)
/*
* We are not sharing SCB space with another controller
* so allocate our own SCB data space.
*/
alloc_size = sizeof(struct full_ahc_softc);
else
alloc_size = sizeof(struct ahc_softc);
ahc = malloc(alloc_size, M_DEVBUF, M_NOWAIT);
if (!ahc) {
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
device_printf(dev, "cannot malloc softc!\n");
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
return NULL;
}
bzero(ahc, alloc_size);
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
LIST_INIT(&ahc->pending_ccbs);
ahc->device = dev;
ahc->unit = device_get_unit(dev);
ahc->regs_res_type = regs_type;
ahc->regs_res_id = regs_id;
ahc->regs = regs;
ahc->tag = rman_get_bustag(regs);
ahc->bsh = rman_get_bushandle(regs);
ahc->parent_dmat = parent_dmat;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahc->chip = config->chip;
ahc->features = config->features;
ahc->bugs = config->bugs;
ahc->flags = config->flags;
ahc->channel = config->channel;
for (i = 0; i < 16; i++)
TAILQ_INIT(&ahc->untagged_queues[i]);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (scb_data == NULL) {
struct full_ahc_softc* full_softc = (struct full_ahc_softc*)ahc;
ahc->scb_data = &full_softc->scb_data_storage;
} else
ahc->scb_data = scb_data;
ahc->unpause = (ahc_inb(ahc, HCNTRL) & IRQMS) | INTEN;
Use TCL_CHANNEL/SCSI_ID/etc. macros in more locations. Don't mess with the IRQMS bit in the host control register unless we are an aic7770 chip. Use calling context to determine if the card is already paused when we update the target message request bit field in controller scratch ram. Looking at the paused bit in the HCNTRL register opened up a race condition. Insert delays in the target message request update routine as a temporary work around for what looks like a chip bug. I'm still investigating this one. Fix the Abort/Abort Tag/BDR handler to pull its message from the message buffer in our softc instead of attempting to get it from a register on the controller. The message is never recorded by the controller in the new message scheme. Don't rely on having an SCB when a BDR occurs. We can issue these during invalid reconnects to. Fix a few cases where we were restarting the sequencer but then still falling out of a switch statement to unpause the sequencer again. This could cause us to mess up sequencer state if it generated another pausing interrupt between the time of the restart and unpause. Kill the 'transceiver settle' loop during card initialization. I failed to realize that a controller that is not connected to any cables will never settle or enable the SCSI transceivers at all. The correct solution is to monitor the IOERR interrupt which indicates that the transceiver state has changed (UW<->LVD). Modify the aic7xxx assembler to properly echo input when stdin is not a tty.
1999-03-23 07:24:29 +00:00
/* The IRQMS bit is only valid on VL and EISA chips */
if ((ahc->chip & AHC_PCI) != 0)
ahc->unpause &= ~IRQMS;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc->pause = ahc->unpause | PAUSE;
return (ahc);
}
void
ahc_free(ahc)
struct ahc_softc *ahc;
{
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
ahcfiniscbdata(ahc);
switch (ahc->init_level) {
case 3:
bus_dmamap_unload(ahc->shared_data_dmat,
ahc->shared_data_dmamap);
case 2:
bus_dmamem_free(ahc->shared_data_dmat, ahc->qoutfifo,
ahc->shared_data_dmamap);
bus_dmamap_destroy(ahc->shared_data_dmat,
ahc->shared_data_dmamap);
case 1:
bus_dma_tag_destroy(ahc->buffer_dmat);
break;
}
if (ahc->regs != NULL)
bus_release_resource(ahc->device, ahc->regs_res_type,
ahc->regs_res_id, ahc->regs);
if (ahc->irq != NULL)
bus_release_resource(ahc->device, ahc->irq_res_type,
0, ahc->irq);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
free(ahc, M_DEVBUF);
return;
}
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
static int
ahcinitscbdata(struct ahc_softc *ahc)
{
struct scb_data *scb_data;
int i;
scb_data = ahc->scb_data;
SLIST_INIT(&scb_data->free_scbs);
SLIST_INIT(&scb_data->sg_maps);
/* Allocate SCB resources */
scb_data->scbarray =
(struct scb *)malloc(sizeof(struct scb) * AHC_SCB_MAX,
M_DEVBUF, M_NOWAIT);
if (scb_data->scbarray == NULL)
return (ENOMEM);
bzero(scb_data->scbarray, sizeof(struct scb) * AHC_SCB_MAX);
/* Determine the number of hardware SCBs and initialize them */
scb_data->maxhscbs = ahc_probe_scbs(ahc);
/* SCB 0 heads the free list */
ahc_outb(ahc, FREE_SCBH, 0);
for (i = 0; i < ahc->scb_data->maxhscbs; i++) {
ahc_outb(ahc, SCBPTR, i);
/* Clear the control byte. */
ahc_outb(ahc, SCB_CONTROL, 0);
/* Set the next pointer */
ahc_outb(ahc, SCB_NEXT, i+1);
/* Make the tag number invalid */
ahc_outb(ahc, SCB_TAG, SCB_LIST_NULL);
}
/* Make sure that the last SCB terminates the free list */
ahc_outb(ahc, SCBPTR, i-1);
ahc_outb(ahc, SCB_NEXT, SCB_LIST_NULL);
/* Ensure we clear the 0 SCB's control byte. */
ahc_outb(ahc, SCBPTR, 0);
ahc_outb(ahc, SCB_CONTROL, 0);
scb_data->maxhscbs = i;
if (ahc->scb_data->maxhscbs == 0)
panic("%s: No SCB space found", ahc_name(ahc));
/*
* Create our DMA tags. These tags define the kinds of device
Fix typo (accessable --> accessible). PR: 18588 Submitted by: Anatoly Vorobey <mellon@pobox.com> Reviewed by: asmodai
2000-06-14 17:53:40 +00:00
* accessible memory allocations and memory mappings we will
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
* need to perform during normal operation.
*
* Unless we need to further restrict the allocation, we rely
* on the restrictions of the parent dmat, hence the common
* use of MAXADDR and MAXSIZE.
*/
/* DMA tag for our hardware scb structures */
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
if (bus_dma_tag_create(ahc->parent_dmat, /*alignment*/1, /*boundary*/0,
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
/*lowaddr*/BUS_SPACE_MAXADDR,
/*highaddr*/BUS_SPACE_MAXADDR,
/*filter*/NULL, /*filterarg*/NULL,
AHC_SCB_MAX * sizeof(struct hardware_scb),
/*nsegments*/1,
/*maxsegsz*/BUS_SPACE_MAXSIZE_32BIT,
/*flags*/0, &scb_data->hscb_dmat) != 0) {
goto error_exit;
}
scb_data->init_level++;
/* Allocation for our ccbs */
if (bus_dmamem_alloc(scb_data->hscb_dmat, (void **)&scb_data->hscbs,
BUS_DMA_NOWAIT, &scb_data->hscb_dmamap) != 0) {
goto error_exit;
}
scb_data->init_level++;
/* And permanently map them */
bus_dmamap_load(scb_data->hscb_dmat, scb_data->hscb_dmamap,
scb_data->hscbs,
AHC_SCB_MAX * sizeof(struct hardware_scb),
ahcdmamapcb, &scb_data->hscb_busaddr, /*flags*/0);
scb_data->init_level++;
/* DMA tag for our sense buffers */
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
if (bus_dma_tag_create(ahc->parent_dmat, /*alignment*/1, /*boundary*/0,
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
/*lowaddr*/BUS_SPACE_MAXADDR,
/*highaddr*/BUS_SPACE_MAXADDR,
/*filter*/NULL, /*filterarg*/NULL,
AHC_SCB_MAX * sizeof(struct scsi_sense_data),
/*nsegments*/1,
/*maxsegsz*/BUS_SPACE_MAXSIZE_32BIT,
/*flags*/0, &scb_data->sense_dmat) != 0) {
goto error_exit;
}
scb_data->init_level++;
/* Allocate them */
if (bus_dmamem_alloc(scb_data->sense_dmat, (void **)&scb_data->sense,
BUS_DMA_NOWAIT, &scb_data->sense_dmamap) != 0) {
goto error_exit;
}
scb_data->init_level++;
/* And permanently map them */
bus_dmamap_load(scb_data->sense_dmat, scb_data->sense_dmamap,
scb_data->sense,
AHC_SCB_MAX * sizeof(struct scsi_sense_data),
ahcdmamapcb, &scb_data->sense_busaddr, /*flags*/0);
scb_data->init_level++;
/* DMA tag for our S/G structures. We allocate in page sized chunks */
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
if (bus_dma_tag_create(ahc->parent_dmat, /*alignment*/1, /*boundary*/0,
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
/*lowaddr*/BUS_SPACE_MAXADDR,
/*highaddr*/BUS_SPACE_MAXADDR,
/*filter*/NULL, /*filterarg*/NULL,
PAGE_SIZE, /*nsegments*/1,
/*maxsegsz*/BUS_SPACE_MAXSIZE_32BIT,
/*flags*/0, &scb_data->sg_dmat) != 0) {
goto error_exit;
}
scb_data->init_level++;
/* Perform initial CCB allocation */
bzero(scb_data->hscbs, AHC_SCB_MAX * sizeof(struct hardware_scb));
ahcallocscbs(ahc);
if (scb_data->numscbs == 0) {
printf("%s: ahc_init_scb_data - "
"Unable to allocate initial scbs\n",
ahc_name(ahc));
goto error_exit;
}
/*
* Note that we were successfull
*/
return 0;
error_exit:
return ENOMEM;
}
static void
ahcfiniscbdata(struct ahc_softc *ahc)
{
struct scb_data *scb_data;
scb_data = ahc->scb_data;
switch (scb_data->init_level) {
default:
case 7:
{
struct sg_map_node *sg_map;
while ((sg_map = SLIST_FIRST(&scb_data->sg_maps))!= NULL) {
SLIST_REMOVE_HEAD(&scb_data->sg_maps, links);
bus_dmamap_unload(scb_data->sg_dmat,
sg_map->sg_dmamap);
bus_dmamem_free(scb_data->sg_dmat, sg_map->sg_vaddr,
sg_map->sg_dmamap);
free(sg_map, M_DEVBUF);
}
bus_dma_tag_destroy(scb_data->sg_dmat);
}
case 6:
bus_dmamap_unload(scb_data->sense_dmat,
scb_data->sense_dmamap);
case 5:
bus_dmamem_free(scb_data->sense_dmat, scb_data->sense,
scb_data->sense_dmamap);
bus_dmamap_destroy(scb_data->sense_dmat,
scb_data->sense_dmamap);
case 4:
bus_dma_tag_destroy(scb_data->sense_dmat);
case 3:
bus_dmamap_unload(scb_data->hscb_dmat, scb_data->hscb_dmamap);
case 2:
bus_dmamem_free(scb_data->hscb_dmat, scb_data->hscbs,
scb_data->hscb_dmamap);
bus_dmamap_destroy(scb_data->hscb_dmat, scb_data->hscb_dmamap);
case 1:
bus_dma_tag_destroy(scb_data->hscb_dmat);
break;
}
if (scb_data->scbarray != NULL)
free(scb_data->scbarray, M_DEVBUF);
}
static void
ahcdmamapcb(void *arg, bus_dma_segment_t *segs, int nseg, int error)
{
bus_addr_t *baddr;
baddr = (bus_addr_t *)arg;
*baddr = segs->ds_addr;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
int
ahc_reset(struct ahc_softc *ahc)
{
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
u_int sblkctl;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int sxfrctl1;
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
int wait;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
#ifdef AHC_DUMP_SEQ
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
if (ahc->init_level == 0)
ahc_dumpseq(ahc);
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
#endif
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
/* Cache STPWEN. It is cleared by a chip reset */
pause_sequencer(ahc);
sxfrctl1 = ahc_inb(ahc, SXFRCTL1) & STPWEN;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_outb(ahc, HCNTRL, CHIPRST | ahc->pause);
/*
* Ensure that the reset has finished
*/
wait = 1000;
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
do {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
DELAY(1000);
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
} while (--wait && !(ahc_inb(ahc, HCNTRL) & CHIPRSTACK));
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (wait == 0) {
printf("%s: WARNING - Failed chip reset! "
"Trying to initialize anyway.\n", ahc_name(ahc));
}
ahc_outb(ahc, HCNTRL, ahc->pause);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
/*
* Reload sxfrctl1 with the cached value of STPWEN
* to minimize the amount of time our terminators
* are disabled. If a BIOS has initialized the chip,
* then sxfrctl1 will have the correct value. If
* not, STPWEN will be false (the value after a POST)
* and this action will be harmless.
*
aic7xxx.c: Correct the BUILD_TCL macro. It was placing the target id in the wrong bits. This was only an issue for adapters that do not perform SCB paging (aha-3940AUW for instance). Don't bother inlining ahc_index_busy_tcl. It is never used in a performance critical path and is a bit chunky. Correct ahc_index_busy_tcl to deal with "busy target tables" embedded in the latter half of 64byte SCBs. Don't initialize the busy target table to its empty state until after we have finished extracting configuration information from chip SRAM. In the common case of using 16 bytes of chip SRAM to do untagged target lookups, we were trashing the last 8 targets configuration data. (actually only target 8 because of the bug in the BUILD_TCL macro). Cram the "bus reset delivered" message back under bootverbose. Fix the cleanup of the SCB busy target table when aborting commands. If the lun is wildcarded, we must loop through all possible luns. aic7xxx.h: Only bother supporting 64 luns right now. It doesn't seem like either this driver or any peripherals will be doing information unit transfers (where the lun number is a 32 bit integer) any time soon. aic7xxx.seq: Fix support for the aic7895. We must flush the data FIFO if performing a manual transfer that is not a multiple of 8 bytes. We were doing this quite regularly for embedded cdbs. Manaually flush the fifo on earlier adapters when dealing with embedded cdbs too. We were stuffing the FIFO with 16 bytes instead, but triggering the flush is more efficient and allows us to remove two instructions from the "copy_to_fifo" routine.
2000-07-24 22:27:40 +00:00
* We must always initialize STPWEN to 1 before we
* restore the saved value. STPWEN is initialized
* to a tri-state condition which is only be cleared
* by turning it on.
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
*/
ahc_outb(ahc, SXFRCTL1, sxfrctl1|STPWEN);
ahc_outb(ahc, SXFRCTL1, sxfrctl1);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/* Determine channel configuration */
sblkctl = ahc_inb(ahc, SBLKCTL) & (SELBUSB|SELWIDE);
/* No Twin Channel PCI cards */
if ((ahc->chip & AHC_PCI) != 0)
sblkctl &= ~SELBUSB;
switch (sblkctl) {
case 0:
/* Single Narrow Channel */
break;
case 2:
/* Wide Channel */
ahc->features |= AHC_WIDE;
break;
case 8:
/* Twin Channel */
ahc->features |= AHC_TWIN;
break;
default:
printf(" Unsupported adapter type. Ignoring\n");
return(-1);
}
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
return (0);
}
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
/*
* Called when we have an active connection to a target on the bus,
* this function finds the nearest syncrate to the input period limited
* by the capabilities of the bus connectivity of the target.
*/
static struct ahc_syncrate *
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahc_devlimited_syncrate(struct ahc_softc *ahc, u_int *period,
u_int *ppr_options) {
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
u_int maxsync;
if ((ahc->features & AHC_ULTRA2) != 0) {
if ((ahc_inb(ahc, SBLKCTL) & ENAB40) != 0
&& (ahc_inb(ahc, SSTAT2) & EXP_ACTIVE) == 0) {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
maxsync = AHC_SYNCRATE_DT;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
} else {
maxsync = AHC_SYNCRATE_ULTRA;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
/* Can't do DT on an SE bus */
*ppr_options &= ~MSG_EXT_PPR_DT_REQ;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
}
} else if ((ahc->features & AHC_ULTRA) != 0) {
maxsync = AHC_SYNCRATE_ULTRA;
} else {
maxsync = AHC_SYNCRATE_FAST;
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
return (ahc_find_syncrate(ahc, period, ppr_options, maxsync));
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Look up the valid period to SCSIRATE conversion in our table.
* Return the period and offset that should be sent to the target
* if this was the beginning of an SDTR.
*/
static struct ahc_syncrate *
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahc_find_syncrate(struct ahc_softc *ahc, u_int *period,
u_int *ppr_options, u_int maxsync)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
struct ahc_syncrate *syncrate;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if ((ahc->features & AHC_DT) == 0)
*ppr_options &= ~MSG_EXT_PPR_DT_REQ;
for (syncrate = &ahc_syncrates[maxsync];
syncrate->rate != NULL;
syncrate++) {
/*
* The Ultra2 table doesn't go as low
* as for the Fast/Ultra cards.
*/
if ((ahc->features & AHC_ULTRA2) != 0
&& (syncrate->sxfr_u2 == 0))
break;
/* Skip any DT entries if DT is not available */
if ((*ppr_options & MSG_EXT_PPR_DT_REQ) == 0
&& (syncrate->sxfr_u2 & DT_SXFR) != 0)
continue;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (*period <= syncrate->period) {
/*
* When responding to a target that requests
* sync, the requested rate may fall between
* two rates that we can output, but still be
* a rate that we can receive. Because of this,
* we want to respond to the target with
* the same rate that it sent to us even
* if the period we use to send data to it
* is lower. Only lower the response period
* if we must.
*/
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
if (syncrate == &ahc_syncrates[maxsync])
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*period = syncrate->period;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
/*
* At some speeds, we only support
* ST transfers.
*/
if ((syncrate->sxfr_u2 & ST_SXFR) != 0)
*ppr_options &= ~MSG_EXT_PPR_DT_REQ;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
break;
}
}
if ((*period == 0)
|| (syncrate->rate == NULL)
|| ((ahc->features & AHC_ULTRA2) != 0
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
&& (syncrate->sxfr_u2 == 0))) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/* Use asynchronous transfers. */
*period = 0;
syncrate = NULL;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
*ppr_options &= ~MSG_EXT_PPR_DT_REQ;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
return (syncrate);
}
static u_int
ahc_find_period(struct ahc_softc *ahc, u_int scsirate, u_int maxsync)
{
struct ahc_syncrate *syncrate;
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
if ((ahc->features & AHC_ULTRA2) != 0)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
scsirate &= SXFR_ULTRA2;
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
else
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
scsirate &= SXFR;
syncrate = &ahc_syncrates[maxsync];
while (syncrate->rate != NULL) {
if ((ahc->features & AHC_ULTRA2) != 0) {
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
if (syncrate->sxfr_u2 == 0)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
break;
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
else if (scsirate == (syncrate->sxfr_u2 & SXFR_ULTRA2))
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
return (syncrate->period);
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
} else if (scsirate == (syncrate->sxfr & SXFR)) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
return (syncrate->period);
}
syncrate++;
}
return (0); /* async */
}
static void
ahc_validate_offset(struct ahc_softc *ahc, struct ahc_syncrate *syncrate,
u_int *offset, int wide)
{
u_int maxoffset;
/* Limit offset to what we can do */
if (syncrate == NULL) {
maxoffset = 0;
} else if ((ahc->features & AHC_ULTRA2) != 0) {
maxoffset = MAX_OFFSET_ULTRA2;
} else {
if (wide)
maxoffset = MAX_OFFSET_16BIT;
else
maxoffset = MAX_OFFSET_8BIT;
}
*offset = MIN(*offset, maxoffset);
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
static void
ahc_validate_width(struct ahc_softc *ahc, u_int *bus_width)
{
switch (*bus_width) {
default:
if (ahc->features & AHC_WIDE) {
/* Respond Wide */
*bus_width = MSG_EXT_WDTR_BUS_16_BIT;
break;
}
/* FALLTHROUGH */
case MSG_EXT_WDTR_BUS_8_BIT:
bus_width = MSG_EXT_WDTR_BUS_8_BIT;
break;
}
}
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
static void
ahc_update_target_msg_request(struct ahc_softc *ahc,
struct ahc_devinfo *devinfo,
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
struct ahc_initiator_tinfo *tinfo,
Use TCL_CHANNEL/SCSI_ID/etc. macros in more locations. Don't mess with the IRQMS bit in the host control register unless we are an aic7770 chip. Use calling context to determine if the card is already paused when we update the target message request bit field in controller scratch ram. Looking at the paused bit in the HCNTRL register opened up a race condition. Insert delays in the target message request update routine as a temporary work around for what looks like a chip bug. I'm still investigating this one. Fix the Abort/Abort Tag/BDR handler to pull its message from the message buffer in our softc instead of attempting to get it from a register on the controller. The message is never recorded by the controller in the new message scheme. Don't rely on having an SCB when a BDR occurs. We can issue these during invalid reconnects to. Fix a few cases where we were restarting the sequencer but then still falling out of a switch statement to unpause the sequencer again. This could cause us to mess up sequencer state if it generated another pausing interrupt between the time of the restart and unpause. Kill the 'transceiver settle' loop during card initialization. I failed to realize that a controller that is not connected to any cables will never settle or enable the SCSI transceivers at all. The correct solution is to monitor the IOERR interrupt which indicates that the transceiver state has changed (UW<->LVD). Modify the aic7xxx assembler to properly echo input when stdin is not a tty.
1999-03-23 07:24:29 +00:00
int force, int paused)
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
{
u_int targ_msg_req_orig;
targ_msg_req_orig = ahc->targ_msg_req;
if (tinfo->current.period != tinfo->goal.period
|| tinfo->current.width != tinfo->goal.width
Add a number of interrelated CAM feature enhancements and bug fixes. NOTE: These changes will require recompilation of any userland applications, like cdrecord, xmcd, etc., that use the CAM passthrough interface. A make world is recommended. camcontrol.[c8]: - We now support two new commands, "tags" and "negotiate". - The tags commands allows users to view the number of tagged openings for a device as well as a number of other related parameters, and it allows users to set tagged openings for a device. - The negotiate command allows users to enable and disable disconnection and tagged queueing, set sync rates, offsets and bus width. Note that not all of those features are available for all controllers. Only the adv, ahc, and ncr drivers fully support all of the features at this point. Some cards do not allow the setting of sync rates, offsets and the like, and some of the drivers don't have any facilities to do so. Some drivers, like the adw driver, only support enabling or disabling sync negotiation, but do not support setting sync rates. - new description in the camcontrol man page of how to format a disk - cleanup of the camcontrol inquiry command - add support in the 'devlist' command for skipping unconfigured devices if -v was not specified on the command line. - make use of the new base_transfer_speed in the path inquiry CCB. - fix CCB bzero cases cam_xpt.c, cam_sim.[ch], cam_ccb.h: - new flags on many CCB function codes to designate whether they're non-immediate, use a user-supplied CCB, and can only be passed from userland programs via the xpt device. Use these flags in the transport layer and pass driver to categorize CCBs. - new flag in the transport layer device matching code for device nodes that indicates whether a device is unconfigured - bump the CAM version from 0x10 to 0x11 - Change the CAM ioctls to use the version as their group code, so we can force users to recompile code even when the CCB size doesn't change. - add + fill in a new value in the path inquiry CCB, base_transfer_speed. Remove a corresponding field from the cam_sim structure, and add code to every SIM to set this field to the proper value. - Fix the set transfer settings code in the transport layer. scsi_cd.c: - make some variables volatile instead of just casting them in various places - fix a race condition in the changer code - attach unless we get a "logical unit not supported" error. This should fix all of the cases where people have devices that return weird errors when they don't have media in the drive. scsi_da.c: - attach unless we get a "logical unit not supported" error scsi_pass.c: - for immediate CCBs, just malloc a CCB to send the user request in. This gets rid of the 'held' count problem in camcontrol tags. scsi_pass.h: - change the CAM ioctls to use the CAM version as their group code. adv driver: - Allow changing the sync rate and offset separately. adw driver - Allow changing the sync rate and offset separately. aha driver: - Don't return CAM_REQ_CMP for SET_TRAN_SETTINGS CCBs. ahc driver: - Allow setting offset and sync rate separately bt driver: - Don't return CAM_REQ_CMP for SET_TRAN_SETTINGS CCBs. NCR driver: - Fix the ultra/ultra 2 negotiation bug - allow setting both the sync rate and offset separately Other HBA drivers: - Put code in to set the base_transfer_speed field for XPT_GET_TRAN_SETTINGS CCBs. Reviewed by: gibbs, mjacob (isp), imp (aha)
1999-05-06 20:16:39 +00:00
|| tinfo->current.offset != tinfo->goal.offset
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
|| (force
&& (tinfo->goal.period != 0
|| tinfo->goal.width != MSG_EXT_WDTR_BUS_8_BIT)))
ahc->targ_msg_req |= devinfo->target_mask;
else
ahc->targ_msg_req &= ~devinfo->target_mask;
if (ahc->targ_msg_req != targ_msg_req_orig) {
/* Update the message request bit for this target */
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (!paused)
pause_sequencer(ahc);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahc_outb(ahc, TARGET_MSG_REQUEST,
ahc->targ_msg_req & 0xFF);
ahc_outb(ahc, TARGET_MSG_REQUEST + 1,
(ahc->targ_msg_req >> 8) & 0xFF);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (!paused)
unpause_sequencer(ahc);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
}
}
static int
ahc_create_path(struct ahc_softc *ahc, struct ahc_devinfo *devinfo,
struct cam_path **path)
{
path_id_t path_id;
if (devinfo->channel == 'B')
path_id = cam_sim_path(ahc->sim_b);
else
path_id = cam_sim_path(ahc->sim);
return (xpt_create_path(path, /*periph*/NULL,
path_id, devinfo->target,
devinfo->lun));
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static void
ahc_set_syncrate(struct ahc_softc *ahc, struct ahc_devinfo *devinfo,
struct cam_path *path, struct ahc_syncrate *syncrate,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int period, u_int offset, u_int ppr_options,
u_int type, int paused)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
Use TCL_CHANNEL/SCSI_ID/etc. macros in more locations. Don't mess with the IRQMS bit in the host control register unless we are an aic7770 chip. Use calling context to determine if the card is already paused when we update the target message request bit field in controller scratch ram. Looking at the paused bit in the HCNTRL register opened up a race condition. Insert delays in the target message request update routine as a temporary work around for what looks like a chip bug. I'm still investigating this one. Fix the Abort/Abort Tag/BDR handler to pull its message from the message buffer in our softc instead of attempting to get it from a register on the controller. The message is never recorded by the controller in the new message scheme. Don't rely on having an SCB when a BDR occurs. We can issue these during invalid reconnects to. Fix a few cases where we were restarting the sequencer but then still falling out of a switch statement to unpause the sequencer again. This could cause us to mess up sequencer state if it generated another pausing interrupt between the time of the restart and unpause. Kill the 'transceiver settle' loop during card initialization. I failed to realize that a controller that is not connected to any cables will never settle or enable the SCSI transceivers at all. The correct solution is to monitor the IOERR interrupt which indicates that the transceiver state has changed (UW<->LVD). Modify the aic7xxx assembler to properly echo input when stdin is not a tty.
1999-03-23 07:24:29 +00:00
struct ahc_initiator_tinfo *tinfo;
struct tmode_tstate *tstate;
u_int old_period;
u_int old_offset;
int active = (type & AHC_TRANS_ACTIVE) == AHC_TRANS_ACTIVE;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (syncrate == NULL) {
period = 0;
offset = 0;
}
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
tinfo = ahc_fetch_transinfo(ahc, devinfo->channel, devinfo->our_scsiid,
devinfo->target, &tstate);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
old_period = tinfo->current.period;
old_offset = tinfo->current.offset;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if ((type & AHC_TRANS_CUR) != 0
&& (old_period != period || old_offset != offset)) {
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
struct cam_path *path2;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
u_int scsirate;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
scsirate = tinfo->scsirate;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if ((ahc->features & AHC_ULTRA2) != 0) {
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
scsirate &= ~(SXFR_ULTRA2|SINGLE_EDGE|ENABLE_CRC);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (syncrate != NULL) {
scsirate |= syncrate->sxfr_u2;
if ((ppr_options & MSG_EXT_PPR_DT_REQ) != 0)
scsirate |= ENABLE_CRC;
else
scsirate |= SINGLE_EDGE;
}
Use TCL_CHANNEL/SCSI_ID/etc. macros in more locations. Don't mess with the IRQMS bit in the host control register unless we are an aic7770 chip. Use calling context to determine if the card is already paused when we update the target message request bit field in controller scratch ram. Looking at the paused bit in the HCNTRL register opened up a race condition. Insert delays in the target message request update routine as a temporary work around for what looks like a chip bug. I'm still investigating this one. Fix the Abort/Abort Tag/BDR handler to pull its message from the message buffer in our softc instead of attempting to get it from a register on the controller. The message is never recorded by the controller in the new message scheme. Don't rely on having an SCB when a BDR occurs. We can issue these during invalid reconnects to. Fix a few cases where we were restarting the sequencer but then still falling out of a switch statement to unpause the sequencer again. This could cause us to mess up sequencer state if it generated another pausing interrupt between the time of the restart and unpause. Kill the 'transceiver settle' loop during card initialization. I failed to realize that a controller that is not connected to any cables will never settle or enable the SCSI transceivers at all. The correct solution is to monitor the IOERR interrupt which indicates that the transceiver state has changed (UW<->LVD). Modify the aic7xxx assembler to properly echo input when stdin is not a tty.
1999-03-23 07:24:29 +00:00
if (active)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_outb(ahc, SCSIOFFSET, offset);
} else {
scsirate &= ~(SXFR|SOFS);
/*
* Ensure Ultra mode is set properly for
* this target.
*/
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
tstate->ultraenb &= ~devinfo->target_mask;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (syncrate != NULL) {
if (syncrate->sxfr & ULTRA_SXFR) {
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
tstate->ultraenb |=
devinfo->target_mask;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
scsirate |= syncrate->sxfr & SXFR;
scsirate |= offset & SOFS;
}
Use TCL_CHANNEL/SCSI_ID/etc. macros in more locations. Don't mess with the IRQMS bit in the host control register unless we are an aic7770 chip. Use calling context to determine if the card is already paused when we update the target message request bit field in controller scratch ram. Looking at the paused bit in the HCNTRL register opened up a race condition. Insert delays in the target message request update routine as a temporary work around for what looks like a chip bug. I'm still investigating this one. Fix the Abort/Abort Tag/BDR handler to pull its message from the message buffer in our softc instead of attempting to get it from a register on the controller. The message is never recorded by the controller in the new message scheme. Don't rely on having an SCB when a BDR occurs. We can issue these during invalid reconnects to. Fix a few cases where we were restarting the sequencer but then still falling out of a switch statement to unpause the sequencer again. This could cause us to mess up sequencer state if it generated another pausing interrupt between the time of the restart and unpause. Kill the 'transceiver settle' loop during card initialization. I failed to realize that a controller that is not connected to any cables will never settle or enable the SCSI transceivers at all. The correct solution is to monitor the IOERR interrupt which indicates that the transceiver state has changed (UW<->LVD). Modify the aic7xxx assembler to properly echo input when stdin is not a tty.
1999-03-23 07:24:29 +00:00
if (active) {
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
u_int sxfrctl0;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
sxfrctl0 = ahc_inb(ahc, SXFRCTL0);
sxfrctl0 &= ~FAST20;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
if (tstate->ultraenb & devinfo->target_mask)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
sxfrctl0 |= FAST20;
ahc_outb(ahc, SXFRCTL0, sxfrctl0);
}
}
Use TCL_CHANNEL/SCSI_ID/etc. macros in more locations. Don't mess with the IRQMS bit in the host control register unless we are an aic7770 chip. Use calling context to determine if the card is already paused when we update the target message request bit field in controller scratch ram. Looking at the paused bit in the HCNTRL register opened up a race condition. Insert delays in the target message request update routine as a temporary work around for what looks like a chip bug. I'm still investigating this one. Fix the Abort/Abort Tag/BDR handler to pull its message from the message buffer in our softc instead of attempting to get it from a register on the controller. The message is never recorded by the controller in the new message scheme. Don't rely on having an SCB when a BDR occurs. We can issue these during invalid reconnects to. Fix a few cases where we were restarting the sequencer but then still falling out of a switch statement to unpause the sequencer again. This could cause us to mess up sequencer state if it generated another pausing interrupt between the time of the restart and unpause. Kill the 'transceiver settle' loop during card initialization. I failed to realize that a controller that is not connected to any cables will never settle or enable the SCSI transceivers at all. The correct solution is to monitor the IOERR interrupt which indicates that the transceiver state has changed (UW<->LVD). Modify the aic7xxx assembler to properly echo input when stdin is not a tty.
1999-03-23 07:24:29 +00:00
if (active)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_outb(ahc, SCSIRATE, scsirate);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
tinfo->scsirate = scsirate;
tinfo->current.period = period;
tinfo->current.offset = offset;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
tinfo->current.ppr_options = ppr_options;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/* Update the syncrates in any pending scbs */
ahc_update_pending_syncrates(ahc);
/*
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
* If possible, tell the SCSI layer about the
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
* new transfer parameters.
*/
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
/* If possible, update the XPT's notion of our transfer rate */
path2 = NULL;
if (path == NULL) {
int error;
error = ahc_create_path(ahc, devinfo, &path2);
if (error == CAM_REQ_CMP)
path = path2;
else
path2 = NULL;
}
if (path != NULL) {
struct ccb_trans_settings neg;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
neg.flags = CCB_TRANS_CURRENT_SETTINGS;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
neg.sync_period = period;
neg.sync_offset = offset;
neg.valid = CCB_TRANS_SYNC_RATE_VALID
| CCB_TRANS_SYNC_OFFSET_VALID;
xpt_setup_ccb(&neg.ccb_h, path, /*priority*/1);
xpt_async(AC_TRANSFER_NEG, path, &neg);
}
if (path2 != NULL)
xpt_free_path(path2);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (bootverbose) {
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if (offset != 0) {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
printf("%s: target %d synchronous at %sMHz%s, "
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
"offset = 0x%x\n", ahc_name(ahc),
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
devinfo->target, syncrate->rate,
(ppr_options & MSG_EXT_PPR_DT_REQ)
? " DT" : "", offset);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
} else {
printf("%s: target %d using "
"asynchronous transfers\n",
ahc_name(ahc), devinfo->target);
}
}
}
if ((type & AHC_TRANS_GOAL) != 0) {
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
tinfo->goal.period = period;
tinfo->goal.offset = offset;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
tinfo->goal.ppr_options = ppr_options;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
if ((type & AHC_TRANS_USER) != 0) {
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
tinfo->user.period = period;
tinfo->user.offset = offset;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
tinfo->user.ppr_options = ppr_options;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
Use TCL_CHANNEL/SCSI_ID/etc. macros in more locations. Don't mess with the IRQMS bit in the host control register unless we are an aic7770 chip. Use calling context to determine if the card is already paused when we update the target message request bit field in controller scratch ram. Looking at the paused bit in the HCNTRL register opened up a race condition. Insert delays in the target message request update routine as a temporary work around for what looks like a chip bug. I'm still investigating this one. Fix the Abort/Abort Tag/BDR handler to pull its message from the message buffer in our softc instead of attempting to get it from a register on the controller. The message is never recorded by the controller in the new message scheme. Don't rely on having an SCB when a BDR occurs. We can issue these during invalid reconnects to. Fix a few cases where we were restarting the sequencer but then still falling out of a switch statement to unpause the sequencer again. This could cause us to mess up sequencer state if it generated another pausing interrupt between the time of the restart and unpause. Kill the 'transceiver settle' loop during card initialization. I failed to realize that a controller that is not connected to any cables will never settle or enable the SCSI transceivers at all. The correct solution is to monitor the IOERR interrupt which indicates that the transceiver state has changed (UW<->LVD). Modify the aic7xxx assembler to properly echo input when stdin is not a tty.
1999-03-23 07:24:29 +00:00
ahc_update_target_msg_request(ahc, devinfo, tinfo,
/*force*/FALSE,
Correct a bug where the chip could be unpaused in the middle of a bus or device reset error recovery operation.
1999-05-18 03:58:49 +00:00
paused);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
static void
ahc_set_width(struct ahc_softc *ahc, struct ahc_devinfo *devinfo,
Correct a bug where the chip could be unpaused in the middle of a bus or device reset error recovery operation.
1999-05-18 03:58:49 +00:00
struct cam_path *path, u_int width, u_int type, int paused)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
struct ahc_initiator_tinfo *tinfo;
struct tmode_tstate *tstate;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
u_int oldwidth;
Use TCL_CHANNEL/SCSI_ID/etc. macros in more locations. Don't mess with the IRQMS bit in the host control register unless we are an aic7770 chip. Use calling context to determine if the card is already paused when we update the target message request bit field in controller scratch ram. Looking at the paused bit in the HCNTRL register opened up a race condition. Insert delays in the target message request update routine as a temporary work around for what looks like a chip bug. I'm still investigating this one. Fix the Abort/Abort Tag/BDR handler to pull its message from the message buffer in our softc instead of attempting to get it from a register on the controller. The message is never recorded by the controller in the new message scheme. Don't rely on having an SCB when a BDR occurs. We can issue these during invalid reconnects to. Fix a few cases where we were restarting the sequencer but then still falling out of a switch statement to unpause the sequencer again. This could cause us to mess up sequencer state if it generated another pausing interrupt between the time of the restart and unpause. Kill the 'transceiver settle' loop during card initialization. I failed to realize that a controller that is not connected to any cables will never settle or enable the SCSI transceivers at all. The correct solution is to monitor the IOERR interrupt which indicates that the transceiver state has changed (UW<->LVD). Modify the aic7xxx assembler to properly echo input when stdin is not a tty.
1999-03-23 07:24:29 +00:00
int active = (type & AHC_TRANS_ACTIVE) == AHC_TRANS_ACTIVE;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
tinfo = ahc_fetch_transinfo(ahc, devinfo->channel, devinfo->our_scsiid,
devinfo->target, &tstate);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
oldwidth = tinfo->current.width;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if ((type & AHC_TRANS_CUR) != 0 && oldwidth != width) {
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
struct cam_path *path2;
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
u_int scsirate;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
scsirate = tinfo->scsirate;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
scsirate &= ~WIDEXFER;
if (width == MSG_EXT_WDTR_BUS_16_BIT)
scsirate |= WIDEXFER;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
tinfo->scsirate = scsirate;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Use TCL_CHANNEL/SCSI_ID/etc. macros in more locations. Don't mess with the IRQMS bit in the host control register unless we are an aic7770 chip. Use calling context to determine if the card is already paused when we update the target message request bit field in controller scratch ram. Looking at the paused bit in the HCNTRL register opened up a race condition. Insert delays in the target message request update routine as a temporary work around for what looks like a chip bug. I'm still investigating this one. Fix the Abort/Abort Tag/BDR handler to pull its message from the message buffer in our softc instead of attempting to get it from a register on the controller. The message is never recorded by the controller in the new message scheme. Don't rely on having an SCB when a BDR occurs. We can issue these during invalid reconnects to. Fix a few cases where we were restarting the sequencer but then still falling out of a switch statement to unpause the sequencer again. This could cause us to mess up sequencer state if it generated another pausing interrupt between the time of the restart and unpause. Kill the 'transceiver settle' loop during card initialization. I failed to realize that a controller that is not connected to any cables will never settle or enable the SCSI transceivers at all. The correct solution is to monitor the IOERR interrupt which indicates that the transceiver state has changed (UW<->LVD). Modify the aic7xxx assembler to properly echo input when stdin is not a tty.
1999-03-23 07:24:29 +00:00
if (active)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_outb(ahc, SCSIRATE, scsirate);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
tinfo->current.width = width;
/* If possible, update the XPT's notion of our transfer rate */
path2 = NULL;
if (path == NULL) {
int error;
error = ahc_create_path(ahc, devinfo, &path2);
if (error == CAM_REQ_CMP)
path = path2;
else
path2 = NULL;
}
if (path != NULL) {
struct ccb_trans_settings neg;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
neg.flags = CCB_TRANS_CURRENT_SETTINGS;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
neg.bus_width = width;
neg.valid = CCB_TRANS_BUS_WIDTH_VALID;
xpt_setup_ccb(&neg.ccb_h, path, /*priority*/1);
xpt_async(AC_TRANSFER_NEG, path, &neg);
}
if (path2 != NULL)
xpt_free_path(path2);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (bootverbose) {
printf("%s: target %d using %dbit transfers\n",
ahc_name(ahc), devinfo->target,
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
8 * (0x01 << width));
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
}
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if ((type & AHC_TRANS_GOAL) != 0)
tinfo->goal.width = width;
if ((type & AHC_TRANS_USER) != 0)
tinfo->user.width = width;
Use TCL_CHANNEL/SCSI_ID/etc. macros in more locations. Don't mess with the IRQMS bit in the host control register unless we are an aic7770 chip. Use calling context to determine if the card is already paused when we update the target message request bit field in controller scratch ram. Looking at the paused bit in the HCNTRL register opened up a race condition. Insert delays in the target message request update routine as a temporary work around for what looks like a chip bug. I'm still investigating this one. Fix the Abort/Abort Tag/BDR handler to pull its message from the message buffer in our softc instead of attempting to get it from a register on the controller. The message is never recorded by the controller in the new message scheme. Don't rely on having an SCB when a BDR occurs. We can issue these during invalid reconnects to. Fix a few cases where we were restarting the sequencer but then still falling out of a switch statement to unpause the sequencer again. This could cause us to mess up sequencer state if it generated another pausing interrupt between the time of the restart and unpause. Kill the 'transceiver settle' loop during card initialization. I failed to realize that a controller that is not connected to any cables will never settle or enable the SCSI transceivers at all. The correct solution is to monitor the IOERR interrupt which indicates that the transceiver state has changed (UW<->LVD). Modify the aic7xxx assembler to properly echo input when stdin is not a tty.
1999-03-23 07:24:29 +00:00
ahc_update_target_msg_request(ahc, devinfo, tinfo,
Correct a bug where the chip could be unpaused in the middle of a bus or device reset error recovery operation.
1999-05-18 03:58:49 +00:00
/*force*/FALSE, paused);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
static void
ahc_set_tags(struct ahc_softc *ahc, struct ahc_devinfo *devinfo, int enable)
{
struct ahc_initiator_tinfo *tinfo;
struct tmode_tstate *tstate;
tinfo = ahc_fetch_transinfo(ahc, devinfo->channel, devinfo->our_scsiid,
devinfo->target, &tstate);
if (enable)
tstate->tagenable |= devinfo->target_mask;
else
tstate->tagenable &= ~devinfo->target_mask;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Attach all the sub-devices we can find
*/
int
ahc_attach(struct ahc_softc *ahc)
{
struct ccb_setasync csa;
struct cam_devq *devq;
int bus_id;
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
int bus_id2;
struct cam_sim *sim;
struct cam_sim *sim2;
struct cam_path *path;
struct cam_path *path2;
int count;
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
int s;
int error;
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
count = 0;
sim = NULL;
sim2 = NULL;
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
s = splcam();
/* Hook up our interrupt handler */
Move the declaration of the interrupt type from the driver structure to the BUS_SETUP_INTR call.
1999-05-08 21:59:43 +00:00
if ((error = bus_setup_intr(ahc->device, ahc->irq, INTR_TYPE_CAM,
ahc_intr, ahc, &ahc->ih)) != 0) {
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
device_printf(ahc->device, "bus_setup_intr() failed: %d\n",
error);
goto fail;
}
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
/*
* Attach secondary channel first if the user has
* declared it the primary channel.
*/
if ((ahc->flags & AHC_CHANNEL_B_PRIMARY) != 0) {
bus_id = 1;
bus_id2 = 0;
} else {
bus_id = 0;
bus_id2 = 1;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
* Create the device queue for our SIM(s).
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
devq = cam_simq_alloc(AHC_SCB_MAX);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (devq == NULL)
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
goto fail;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
* Construct our first channel SIM entry
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
sim = cam_sim_alloc(ahc_action, ahc_poll, "ahc", ahc, ahc->unit,
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
1, AHC_SCB_MAX, devq);
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
if (sim == NULL) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
cam_simq_free(devq);
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
goto fail;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
if (xpt_bus_register(sim, bus_id) != CAM_SUCCESS) {
cam_sim_free(sim, /*free_devq*/TRUE);
sim = NULL;
goto fail;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
if (xpt_create_path(&path, /*periph*/NULL,
cam_sim_path(sim), CAM_TARGET_WILDCARD,
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
CAM_LUN_WILDCARD) != CAM_REQ_CMP) {
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
xpt_bus_deregister(cam_sim_path(sim));
cam_sim_free(sim, /*free_devq*/TRUE);
sim = NULL;
goto fail;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
xpt_setup_ccb(&csa.ccb_h, path, /*priority*/5);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
csa.ccb_h.func_code = XPT_SASYNC_CB;
csa.event_enable = AC_LOST_DEVICE;
csa.callback = ahc_async;
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
csa.callback_arg = sim;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
xpt_action((union ccb *)&csa);
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
count++;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (ahc->features & AHC_TWIN) {
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
sim2 = cam_sim_alloc(ahc_action, ahc_poll, "ahc",
ahc, ahc->unit, 1,
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
AHC_SCB_MAX, devq);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
if (sim2 == NULL) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
printf("ahc_attach: Unable to attach second "
"bus due to resource shortage");
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
goto fail;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
if (xpt_bus_register(sim2, bus_id2) != CAM_SUCCESS) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
printf("ahc_attach: Unable to attach second "
"bus due to resource shortage");
/*
* We do not want to destroy the device queue
* because the first bus is using it.
*/
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
cam_sim_free(sim2, /*free_devq*/FALSE);
goto fail;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
if (xpt_create_path(&path2, /*periph*/NULL,
cam_sim_path(sim2),
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
CAM_TARGET_WILDCARD,
CAM_LUN_WILDCARD) != CAM_REQ_CMP) {
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
xpt_bus_deregister(cam_sim_path(sim2));
cam_sim_free(sim2, /*free_devq*/FALSE);
sim2 = NULL;
goto fail;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
xpt_setup_ccb(&csa.ccb_h, path2, /*priority*/5);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
csa.ccb_h.func_code = XPT_SASYNC_CB;
csa.event_enable = AC_LOST_DEVICE;
csa.callback = ahc_async;
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
csa.callback_arg = sim2;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
xpt_action((union ccb *)&csa);
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
count++;
}
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
fail:
if ((ahc->flags & AHC_CHANNEL_B_PRIMARY) != 0) {
ahc->sim_b = sim;
ahc->path_b = path;
ahc->sim = sim2;
ahc->path = path2;
} else {
ahc->sim = sim;
ahc->path = path;
ahc->sim_b = sim2;
ahc->path_b = path2;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
splx(s);
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
return (count);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
#if UNUSED
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
static void
ahc_scb_devinfo(struct ahc_softc *ahc, struct ahc_devinfo *devinfo,
struct scb *scb)
{
role_t role;
int our_id;
if (scb->ccb->ccb_h.func_code == XPT_CONT_TARGET_IO) {
our_id = scb->ccb->ccb_h.target_id;
role = ROLE_TARGET;
} else {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
our_id = SCB_GET_CHANNEL(scb) == 'B' ? ahc->our_id_b : ahc->our_id;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
role = ROLE_INITIATOR;
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahc_compile_devinfo(devinfo, our_id, SCB_GET_TARGET(ahc, scb),
SCB_GET_LUN(scb), SCB_GET_CHANNEL(scb), role);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
}
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
#endif
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static void
ahc_fetch_devinfo(struct ahc_softc *ahc, struct ahc_devinfo *devinfo)
{
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int saved_scsiid;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
role_t role;
int our_id;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if (ahc_inb(ahc, SSTAT0) & TARGET)
role = ROLE_TARGET;
else
role = ROLE_INITIATOR;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
if (role == ROLE_TARGET
&& (ahc->features & AHC_MULTI_TID) != 0
&& (ahc_inb(ahc, SEQ_FLAGS) & CMDPHASE_PENDING) != 0) {
/* We were selected, so pull our id from TARGIDIN */
our_id = ahc_inb(ahc, TARGIDIN) & OID;
} else if ((ahc->features & AHC_ULTRA2) != 0)
our_id = ahc_inb(ahc, SCSIID_ULTRA2) & OID;
else
our_id = ahc_inb(ahc, SCSIID) & OID;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
saved_scsiid = ahc_inb(ahc, SAVED_SCSIID);
ahc_compile_devinfo(devinfo,
our_id,
SCSIID_TARGET(ahc, saved_scsiid),
ahc_inb(ahc, SAVED_LUN),
SCSIID_CHANNEL(ahc, saved_scsiid),
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
role);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
static void
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
ahc_compile_devinfo(struct ahc_devinfo *devinfo, u_int our_id, u_int target,
u_int lun, char channel, role_t role)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
devinfo->our_scsiid = our_id;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
devinfo->target = target;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
devinfo->lun = lun;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
devinfo->target_offset = target;
devinfo->channel = channel;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
devinfo->role = role;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (channel == 'B')
devinfo->target_offset += 8;
devinfo->target_mask = (0x01 << devinfo->target_offset);
}
/*
* Catch an interrupt from the adapter
*/
void
ahc_intr(void *arg)
{
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
struct ahc_softc *ahc;
u_int intstat;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc = (struct ahc_softc *)arg;
intstat = ahc_inb(ahc, INTSTAT);
/*
* Any interrupts to process?
*/
#if NPCI > 0
if ((intstat & INT_PEND) == 0) {
if ((ahc->chip & AHC_PCI) != 0
&& (ahc->unsolicited_ints > 500)) {
if ((ahc_inb(ahc, ERROR) & PCIERRSTAT) != 0)
ahc_pci_intr(ahc);
ahc->unsolicited_ints = 0;
} else {
ahc->unsolicited_ints++;
}
return;
} else {
ahc->unsolicited_ints = 0;
}
#else
if ((intstat & INT_PEND) == 0)
return;
#endif
if (intstat & CMDCMPLT) {
ahc_outb(ahc, CLRINT, CLRCMDINT);
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
ahc_run_qoutfifo(ahc);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if ((ahc->flags & AHC_TARGETMODE) != 0)
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
ahc_run_tqinfifo(ahc, /*paused*/FALSE);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
if (intstat & BRKADRINT) {
/*
* We upset the sequencer :-(
* Lookup the error message
*/
int i, error, num_errors;
error = ahc_inb(ahc, ERROR);
num_errors = sizeof(hard_error)/sizeof(hard_error[0]);
for (i = 0; error != 1 && i < num_errors; i++)
error >>= 1;
panic("%s: brkadrint, %s at seqaddr = 0x%x\n",
ahc_name(ahc), hard_error[i].errmesg,
ahc_inb(ahc, SEQADDR0) |
(ahc_inb(ahc, SEQADDR1) << 8));
/* Tell everyone that this HBA is no longer availible */
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc_abort_scbs(ahc, CAM_TARGET_WILDCARD, ALL_CHANNELS,
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
CAM_LUN_WILDCARD, SCB_LIST_NULL, ROLE_UNKNOWN,
CAM_NO_HBA);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if ((intstat & (SEQINT|SCSIINT)) != 0)
ahc_pause_bug_fix(ahc);
if ((intstat & SEQINT) != 0)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_handle_seqint(ahc, intstat);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if ((intstat & SCSIINT) != 0)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_handle_scsiint(ahc, intstat);
}
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
static struct tmode_tstate *
ahc_alloc_tstate(struct ahc_softc *ahc, u_int scsi_id, char channel)
{
struct tmode_tstate *master_tstate;
struct tmode_tstate *tstate;
Add necessary splcam() protection around target mode setup and teardown routines as well as card access that occurs for the AC_LOST_DEVICE async event.
1999-03-08 22:42:52 +00:00
int i, s;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
master_tstate = ahc->enabled_targets[ahc->our_id];
if (channel == 'B') {
scsi_id += 8;
master_tstate = ahc->enabled_targets[ahc->our_id_b + 8];
}
if (ahc->enabled_targets[scsi_id] != NULL
&& ahc->enabled_targets[scsi_id] != master_tstate)
panic("%s: ahc_alloc_tstate - Target already allocated",
ahc_name(ahc));
tstate = malloc(sizeof(*tstate), M_DEVBUF, M_NOWAIT);
if (tstate == NULL)
return (NULL);
/*
* If we have allocated a master tstate, copy user settings from
* the master tstate (taken from SRAM or the EEPROM) for this
* channel, but reset our current and goal settings to async/narrow
* until an initiator talks to us.
*/
if (master_tstate != NULL) {
bcopy(master_tstate, tstate, sizeof(*tstate));
bzero(tstate->enabled_luns, sizeof(tstate->enabled_luns));
tstate->ultraenb = 0;
for (i = 0; i < 16; i++) {
bzero(&tstate->transinfo[i].current,
sizeof(tstate->transinfo[i].current));
bzero(&tstate->transinfo[i].goal,
sizeof(tstate->transinfo[i].goal));
}
} else
bzero(tstate, sizeof(*tstate));
Add necessary splcam() protection around target mode setup and teardown routines as well as card access that occurs for the AC_LOST_DEVICE async event.
1999-03-08 22:42:52 +00:00
s = splcam();
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
ahc->enabled_targets[scsi_id] = tstate;
Add necessary splcam() protection around target mode setup and teardown routines as well as card access that occurs for the AC_LOST_DEVICE async event.
1999-03-08 22:42:52 +00:00
splx(s);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
return (tstate);
}
static void
ahc_free_tstate(struct ahc_softc *ahc, u_int scsi_id, char channel, int force)
{
struct tmode_tstate *tstate;
/* Don't clean up the entry for our initiator role */
if ((ahc->flags & AHC_INITIATORMODE) != 0
&& ((channel == 'B' && scsi_id == ahc->our_id_b)
|| (channel == 'A' && scsi_id == ahc->our_id))
&& force == FALSE)
return;
if (channel == 'B')
scsi_id += 8;
tstate = ahc->enabled_targets[scsi_id];
if (tstate != NULL)
free(tstate, M_DEVBUF);
ahc->enabled_targets[scsi_id] = NULL;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static void
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
ahc_handle_en_lun(struct ahc_softc *ahc, struct cam_sim *sim, union ccb *ccb)
{
struct tmode_tstate *tstate;
struct tmode_lstate *lstate;
struct ccb_en_lun *cel;
cam_status status;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int target;
u_int lun;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
u_int target_mask;
char channel;
Add necessary splcam() protection around target mode setup and teardown routines as well as card access that occurs for the AC_LOST_DEVICE async event.
1999-03-08 22:42:52 +00:00
int s;
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
status = ahc_find_tmode_devs(ahc, sim, ccb, &tstate, &lstate,
/* notfound_failure*/FALSE);
if (status != CAM_REQ_CMP) {
ccb->ccb_h.status = status;
return;
}
cel = &ccb->cel;
target = ccb->ccb_h.target_id;
lun = ccb->ccb_h.target_lun;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
channel = SIM_CHANNEL(ahc, sim);
target_mask = 0x01 << target;
if (channel == 'B')
target_mask <<= 8;
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
if (cel->enable != 0) {
u_int scsiseq;
/* Are we already enabled?? */
if (lstate != NULL) {
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
xpt_print_path(ccb->ccb_h.path);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
printf("Lun already enabled\n");
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
ccb->ccb_h.status = CAM_LUN_ALRDY_ENA;
return;
}
if (cel->grp6_len != 0
|| cel->grp7_len != 0) {
/*
* Don't (yet?) support vendor
* specific commands.
*/
ccb->ccb_h.status = CAM_REQ_INVALID;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
printf("Non-zero Group Codes\n");
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
return;
}
/*
* Seems to be okay.
* Setup our data structures.
*/
if (target != CAM_TARGET_WILDCARD && tstate == NULL) {
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
tstate = ahc_alloc_tstate(ahc, target, channel);
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
if (tstate == NULL) {
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
xpt_print_path(ccb->ccb_h.path);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
printf("Couldn't allocate tstate\n");
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
ccb->ccb_h.status = CAM_RESRC_UNAVAIL;
return;
}
}
lstate = malloc(sizeof(*lstate), M_DEVBUF, M_NOWAIT);
if (lstate == NULL) {
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
xpt_print_path(ccb->ccb_h.path);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
printf("Couldn't allocate lstate\n");
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
ccb->ccb_h.status = CAM_RESRC_UNAVAIL;
return;
}
bzero(lstate, sizeof(*lstate));
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
status = xpt_create_path(&lstate->path, /*periph*/NULL,
xpt_path_path_id(ccb->ccb_h.path),
xpt_path_target_id(ccb->ccb_h.path),
xpt_path_lun_id(ccb->ccb_h.path));
if (status != CAM_REQ_CMP) {
free(lstate, M_DEVBUF);
xpt_print_path(ccb->ccb_h.path);
printf("Couldn't allocate path\n");
ccb->ccb_h.status = CAM_RESRC_UNAVAIL;
return;
}
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
SLIST_INIT(&lstate->accept_tios);
SLIST_INIT(&lstate->immed_notifies);
Add necessary splcam() protection around target mode setup and teardown routines as well as card access that occurs for the AC_LOST_DEVICE async event.
1999-03-08 22:42:52 +00:00
s = splcam();
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
pause_sequencer(ahc);
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
if (target != CAM_TARGET_WILDCARD) {
tstate->enabled_luns[lun] = lstate;
ahc->enabled_luns++;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
if ((ahc->features & AHC_MULTI_TID) != 0) {
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
u_int targid_mask;
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
targid_mask = ahc_inb(ahc, TARGID)
| (ahc_inb(ahc, TARGID + 1) << 8);
targid_mask |= target_mask;
ahc_outb(ahc, TARGID, targid_mask);
ahc_outb(ahc, TARGID+1, (targid_mask >> 8));
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
ahc_update_scsiid(ahc, targid_mask);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
} else {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int our_id;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
char channel;
channel = SIM_CHANNEL(ahc, sim);
our_id = SIM_SCSI_ID(ahc, sim);
/*
* This can only happen if selections
* are not enabled
*/
if (target != our_id) {
u_int sblkctl;
char cur_channel;
int swap;
sblkctl = ahc_inb(ahc, SBLKCTL);
cur_channel = (sblkctl & SELBUSB)
? 'B' : 'A';
if ((ahc->features & AHC_TWIN) == 0)
cur_channel = 'A';
swap = cur_channel != channel;
if (channel == 'A')
ahc->our_id = target;
else
ahc->our_id_b = target;
if (swap)
ahc_outb(ahc, SBLKCTL,
sblkctl ^ SELBUSB);
ahc_outb(ahc, SCSIID, target);
if (swap)
ahc_outb(ahc, SBLKCTL, sblkctl);
}
}
} else
ahc->black_hole = lstate;
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
/* Allow select-in operations */
if (ahc->black_hole != NULL && ahc->enabled_luns > 0) {
scsiseq = ahc_inb(ahc, SCSISEQ_TEMPLATE);
scsiseq |= ENSELI;
ahc_outb(ahc, SCSISEQ_TEMPLATE, scsiseq);
scsiseq = ahc_inb(ahc, SCSISEQ);
scsiseq |= ENSELI;
ahc_outb(ahc, SCSISEQ, scsiseq);
}
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
unpause_sequencer(ahc);
Add necessary splcam() protection around target mode setup and teardown routines as well as card access that occurs for the AC_LOST_DEVICE async event.
1999-03-08 22:42:52 +00:00
splx(s);
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
ccb->ccb_h.status = CAM_REQ_CMP;
xpt_print_path(ccb->ccb_h.path);
printf("Lun now enabled for target mode\n");
} else {
struct ccb_hdr *elm;
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
int i, empty;
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
if (lstate == NULL) {
ccb->ccb_h.status = CAM_LUN_INVALID;
return;
}
Add necessary splcam() protection around target mode setup and teardown routines as well as card access that occurs for the AC_LOST_DEVICE async event.
1999-03-08 22:42:52 +00:00
s = splcam();
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
ccb->ccb_h.status = CAM_REQ_CMP;
LIST_FOREACH(elm, &ahc->pending_ccbs, sim_links.le) {
if (elm->func_code == XPT_CONT_TARGET_IO
&& !xpt_path_comp(elm->path, ccb->ccb_h.path)){
printf("CTIO pending\n");
ccb->ccb_h.status = CAM_REQ_INVALID;
Add necessary splcam() protection around target mode setup and teardown routines as well as card access that occurs for the AC_LOST_DEVICE async event.
1999-03-08 22:42:52 +00:00
splx(s);
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
return;
}
}
if (SLIST_FIRST(&lstate->accept_tios) != NULL) {
printf("ATIOs pending\n");
ccb->ccb_h.status = CAM_REQ_INVALID;
}
if (SLIST_FIRST(&lstate->immed_notifies) != NULL) {
printf("INOTs pending\n");
ccb->ccb_h.status = CAM_REQ_INVALID;
}
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
if (ccb->ccb_h.status != CAM_REQ_CMP) {
splx(s);
return;
}
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
xpt_print_path(ccb->ccb_h.path);
printf("Target mode disabled\n");
xpt_free_path(lstate->path);
free(lstate, M_DEVBUF);
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
pause_sequencer(ahc);
/* Can we clean up the target too? */
if (target != CAM_TARGET_WILDCARD) {
tstate->enabled_luns[lun] = NULL;
ahc->enabled_luns--;
for (empty = 1, i = 0; i < 8; i++)
if (tstate->enabled_luns[i] != NULL) {
empty = 0;
break;
}
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
if (empty) {
ahc_free_tstate(ahc, target, channel,
/*force*/FALSE);
if (ahc->features & AHC_MULTI_TID) {
u_int targid_mask;
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
targid_mask = ahc_inb(ahc, TARGID)
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
| (ahc_inb(ahc, TARGID + 1)
<< 8);
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
targid_mask &= ~target_mask;
ahc_outb(ahc, TARGID, targid_mask);
ahc_outb(ahc, TARGID+1,
(targid_mask >> 8));
ahc_update_scsiid(ahc, targid_mask);
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
}
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
}
} else {
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
ahc->black_hole = NULL;
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
/*
* We can't allow selections without
* our black hole device.
*/
empty = TRUE;
Add necessary splcam() protection around target mode setup and teardown routines as well as card access that occurs for the AC_LOST_DEVICE async event.
1999-03-08 22:42:52 +00:00
}
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
if (ahc->enabled_luns == 0) {
/* Disallow select-in */
u_int scsiseq;
scsiseq = ahc_inb(ahc, SCSISEQ_TEMPLATE);
scsiseq &= ~ENSELI;
ahc_outb(ahc, SCSISEQ_TEMPLATE, scsiseq);
scsiseq = ahc_inb(ahc, SCSISEQ);
scsiseq &= ~ENSELI;
ahc_outb(ahc, SCSISEQ, scsiseq);
}
unpause_sequencer(ahc);
Add necessary splcam() protection around target mode setup and teardown routines as well as card access that occurs for the AC_LOST_DEVICE async event.
1999-03-08 22:42:52 +00:00
splx(s);
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
}
}
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
static void
ahc_update_scsiid(struct ahc_softc *ahc, u_int targid_mask)
{
u_int scsiid_mask;
u_int scsiid;
if ((ahc->features & AHC_MULTI_TID) == 0)
panic("ahc_update_scsiid called on non-multitid unit\n");
/*
* Since we will rely on the the TARGID mask
* for selection enables, ensure that OID
* in SCSIID is not set to some other ID
* that we don't want to allow selections on.
*/
if ((ahc->features & AHC_ULTRA2) != 0)
scsiid = ahc_inb(ahc, SCSIID_ULTRA2);
else
scsiid = ahc_inb(ahc, SCSIID);
scsiid_mask = 0x1 << (scsiid & OID);
if ((targid_mask & scsiid_mask) == 0) {
u_int our_id;
/* ffs counts from 1 */
our_id = ffs(targid_mask);
if (our_id == 0)
our_id = ahc->our_id;
else
our_id--;
scsiid &= TID;
scsiid |= our_id;
}
if ((ahc->features & AHC_ULTRA2) != 0)
ahc_outb(ahc, SCSIID_ULTRA2, scsiid);
else
ahc_outb(ahc, SCSIID, scsiid);
}
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
static int
Change the delivery mechanism for incoming target commands. We now use a 256 entry ring buffer of descriptersfor this purpose. This allows the use of a simple 8bit counter in the sequencer code for tracking start location. Entries in the ring buffer now contain a "cmd_valid" byte at their tail. As an entry is serviced, this byte is cleared by the kernel and set by the sequencer during its dma of a new entry. Since this byte is the last portion of the command touched during a dma, the kernel can use this byte to ensure the command it processes is completely valid. The new command format requires a fixed sized DMA from the controller to deliver which allowed for additional simplification of the sequencer code. The hack that required 1 SCB slot to be stolen for incoming command delivery notification is also gone.
1998-11-23 01:33:47 +00:00
ahc_handle_target_cmd(struct ahc_softc *ahc, struct target_cmd *cmd)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
struct tmode_tstate *tstate;
struct tmode_lstate *lstate;
struct ccb_accept_tio *atio;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
uint8_t *byte;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
int initiator;
int target;
int lun;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
initiator = SCSIID_TARGET(ahc, cmd->scsiid);
target = SCSIID_OUR_ID(cmd->scsiid);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
lun = (cmd->identify & MSG_IDENTIFY_LUNMASK);
byte = cmd->bytes;
tstate = ahc->enabled_targets[target];
lstate = NULL;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (tstate != NULL)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
lstate = tstate->enabled_luns[lun];
/*
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
* Commands for disabled luns go to the black hole driver.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
if (lstate == NULL)
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
lstate = ahc->black_hole;
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
atio = (struct ccb_accept_tio*)SLIST_FIRST(&lstate->accept_tios);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (atio == NULL) {
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
ahc->flags |= AHC_TQINFIFO_BLOCKED;
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
/*
* Wait for more ATIOs from the peripheral driver for this lun.
*/
return (1);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
} else
ahc->flags &= ~AHC_TQINFIFO_BLOCKED;
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
#if 0
printf("Incoming command from %d for %d:%d%s\n",
initiator, target, lun,
lstate == ahc->black_hole ? "(Black Holed)" : "");
#endif
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
SLIST_REMOVE_HEAD(&lstate->accept_tios, sim_links.sle);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
if (lstate == ahc->black_hole) {
/* Fill in the wildcards */
atio->ccb_h.target_id = target;
atio->ccb_h.target_lun = lun;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Package it up and send it off to
* whomever has this lun enabled.
*/
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
atio->sense_len = 0;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
atio->init_id = initiator;
if (byte[0] != 0xFF) {
/* Tag was included */
atio->tag_action = *byte++;
atio->tag_id = *byte++;
atio->ccb_h.flags = CAM_TAG_ACTION_VALID;
} else {
atio->ccb_h.flags = 0;
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
byte++;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/* Okay. Now determine the cdb size based on the command code */
switch (*byte >> CMD_GROUP_CODE_SHIFT) {
case 0:
atio->cdb_len = 6;
break;
case 1:
case 2:
atio->cdb_len = 10;
break;
case 4:
atio->cdb_len = 16;
break;
case 5:
atio->cdb_len = 12;
break;
case 3:
default:
/* Only copy the opcode. */
atio->cdb_len = 1;
printf("Reserved or VU command code type encountered\n");
break;
}
bcopy(byte, atio->cdb_io.cdb_bytes, atio->cdb_len);
atio->ccb_h.status |= CAM_CDB_RECVD;
if ((cmd->identify & MSG_IDENTIFY_DISCFLAG) == 0) {
/*
* We weren't allowed to disconnect.
* We're hanging on the bus until a
* continue target I/O comes in response
* to this accept tio.
*/
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
#if 0
printf("Received Immediate Command %d:%d:%d - %p\n",
initiator, target, lun, ahc->pending_device);
#endif
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc->pending_device = lstate;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahc_freeze_ccb((union ccb *)atio);
atio->ccb_h.flags |= CAM_DIS_DISCONNECT;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
xpt_done((union ccb*)atio);
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
return (0);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
static void
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
ahc_handle_seqint(struct ahc_softc *ahc, u_int intstat)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
struct scb *scb;
struct ahc_devinfo devinfo;
ahc_fetch_devinfo(ahc, &devinfo);
/*
* Clear the upper byte that holds SEQINT status
* codes and clear the SEQINT bit. We will unpause
* the sequencer, if appropriate, after servicing
* the request.
*/
ahc_outb(ahc, CLRINT, CLRSEQINT);
switch (intstat & SEQINT_MASK) {
case BAD_STATUS:
{
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
u_int scb_index;
struct hardware_scb *hscb;
struct ccb_scsiio *csio;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* The sequencer will notify us when a command
* has an error that would be of interest to
* the kernel. This allows us to leave the sequencer
* running in the common case of command completes
* without error. The sequencer will already have
* dma'd the SCB back up to us, so we can reference
* the in kernel copy directly.
*/
scb_index = ahc_inb(ahc, SCB_TAG);
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
scb = &ahc->scb_data->scbarray[scb_index];
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Set the default return value to 0 (don't
* send sense). The sense code will change
* this if needed.
*/
ahc_outb(ahc, RETURN_1, 0);
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
if (!(scb_index < ahc->scb_data->numscbs
&& (scb->flags & SCB_ACTIVE) != 0)) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
printf("%s:%c:%d: ahc_intr - referenced scb "
"not valid during seqint 0x%x scb(%d)\n",
ahc_name(ahc), devinfo.channel,
devinfo.target, intstat, scb_index);
goto unpause;
}
Don't reference our SCB until we have validated that the firmware has returned an SCB that is in range.
1999-05-23 18:55:58 +00:00
hscb = scb->hscb;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/* Don't want to clobber the original sense code */
if ((scb->flags & SCB_SENSE) != 0) {
/*
* Clear the SCB_SENSE Flag and have
* the sequencer do a normal command
* complete.
*/
scb->flags &= ~SCB_SENSE;
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
ahcsetccbstatus(scb->ccb, CAM_AUTOSENSE_FAIL);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
break;
}
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
ahcsetccbstatus(scb->ccb, CAM_SCSI_STATUS_ERROR);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
/* Freeze the queue until the client sees the error. */
Correct an invalid (and unused) #define in 93cx6.c Freeze the transaction queue for any type of scsi status other than "okay".
1999-04-07 23:02:45 +00:00
ahc_freeze_devq(ahc, scb->ccb->ccb_h.path);
ahc_freeze_ccb(scb->ccb);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
csio = &scb->ccb->csio;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
csio->scsi_status = hscb->shared_data.status.scsi_status;
switch (csio->scsi_status) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
case SCSI_STATUS_OK:
printf("%s: Interrupted for staus of 0???\n",
ahc_name(ahc));
break;
case SCSI_STATUS_CMD_TERMINATED:
case SCSI_STATUS_CHECK_COND:
#ifdef AHC_DEBUG
if (ahc_debug & AHC_SHOWSENSE) {
xpt_print_path(csio->ccb_h.path);
printf("SCB %d: requests Check Status\n",
scb->hscb->tag);
}
#endif
if ((csio->ccb_h.flags & CAM_DIS_AUTOSENSE) == 0) {
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
struct ahc_dma_seg *sg;
struct scsi_sense *sc;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
struct ahc_initiator_tinfo *targ_info;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
struct tmode_tstate *tstate;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
struct ahc_transinfo *tinfo;
targ_info =
ahc_fetch_transinfo(ahc,
devinfo.channel,
devinfo.our_scsiid,
devinfo.target,
&tstate);
tinfo = &targ_info->current;
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
sg = scb->sg_list;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
sc = (struct scsi_sense *)
(&hscb->shared_data.cdb);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Save off the residual if there is one.
*/
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (ahc_check_residual(scb))
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_calc_residual(scb);
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
else
scb->ccb->csio.resid = 0;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
#ifdef AHC_DEBUG
if (ahc_debug & AHC_SHOWSENSE) {
xpt_print_path(csio->ccb_h.path);
printf("Sending Sense\n");
}
#endif
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
sg->addr = ahc->scb_data->sense_busaddr
+ (hscb->tag*sizeof(struct scsi_sense_data));
sg->len = MIN(sizeof(struct scsi_sense_data),
csio->sense_len);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
sg->len |= AHC_DMA_LAST_SEG;
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
sc->opcode = REQUEST_SENSE;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
sc->byte2 = 0;
if (tinfo->protocol_version <= SCSI_REV_2
&& SCB_GET_LUN(scb) < 8)
sc->byte2 = SCB_GET_LUN(scb) << 5;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
sc->unused[0] = 0;
sc->unused[1] = 0;
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
sc->length = sg->len;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
sc->control = 0;
/*
* Would be nice to preserve DISCENB here,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
* but due to the way we manage busy targets,
* we can't.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
hscb->control = 0;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* This request sense could be because the
* the device lost power or in some other
* way has lost our transfer negotiations.
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
* Renegotiate if appropriate. Unit attention
* errors will be reported before any data
* phases occur.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
if (scb->ccb->csio.resid
== scb->ccb->csio.dxfer_len) {
ahc_update_target_msg_request(ahc,
&devinfo,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
targ_info,
Use TCL_CHANNEL/SCSI_ID/etc. macros in more locations. Don't mess with the IRQMS bit in the host control register unless we are an aic7770 chip. Use calling context to determine if the card is already paused when we update the target message request bit field in controller scratch ram. Looking at the paused bit in the HCNTRL register opened up a race condition. Insert delays in the target message request update routine as a temporary work around for what looks like a chip bug. I'm still investigating this one. Fix the Abort/Abort Tag/BDR handler to pull its message from the message buffer in our softc instead of attempting to get it from a register on the controller. The message is never recorded by the controller in the new message scheme. Don't rely on having an SCB when a BDR occurs. We can issue these during invalid reconnects to. Fix a few cases where we were restarting the sequencer but then still falling out of a switch statement to unpause the sequencer again. This could cause us to mess up sequencer state if it generated another pausing interrupt between the time of the restart and unpause. Kill the 'transceiver settle' loop during card initialization. I failed to realize that a controller that is not connected to any cables will never settle or enable the SCSI transceivers at all. The correct solution is to monitor the IOERR interrupt which indicates that the transceiver state has changed (UW<->LVD). Modify the aic7xxx assembler to properly echo input when stdin is not a tty.
1999-03-23 07:24:29 +00:00
/*force*/TRUE,
/*paused*/TRUE);
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
hscb->cdb_len = sizeof(*sc);
hscb->dataptr = sg->addr;
hscb->datacnt = sg->len;
hscb->sgptr = scb->sg_list_phys | SG_FULL_RESID;
scb->sg_count = 1;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
scb->flags |= SCB_SENSE;
ahc_outb(ahc, RETURN_1, SEND_SENSE);
/*
* Ensure we have enough time to actually
* retrieve the sense.
*/
untimeout(ahc_timeout, (caddr_t)scb,
scb->ccb->ccb_h.timeout_ch);
scb->ccb->ccb_h.timeout_ch =
timeout(ahc_timeout, (caddr_t)scb, 5 * hz);
}
break;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
default:
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
break;
}
break;
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
case NO_MATCH:
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
{
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
/* Ensure we don't leave the selection hardware on */
ahc_outb(ahc, SCSISEQ,
ahc_inb(ahc, SCSISEQ) & (ENSELI|ENRSELI|ENAUTOATNP));
printf("%s:%c:%d: no active SCB for reconnecting "
"target - issuing BUS DEVICE RESET\n",
ahc_name(ahc), devinfo.channel, devinfo.target);
printf("SAVED_SCSIID == 0x%x, SAVED_LUN == 0x%x, "
"ARG_1 == 0x%x ARG_2 = 0x%x, SEQ_FLAGS == 0x%x\n",
ahc_inb(ahc, SAVED_SCSIID), ahc_inb(ahc, SAVED_LUN),
ahc_inb(ahc, ARG_1), ahc_inb(ahc, ARG_2),
ahc_inb(ahc, SEQ_FLAGS));
printf("SCB_SCSIID == 0x%x, SCB_LUN == 0x%x, "
"SCB_TAG == 0x%x\n",
ahc_inb(ahc, SCB_SCSIID), ahc_inb(ahc, SCB_LUN),
ahc_inb(ahc, SCB_TAG));
ahc->msgout_buf[0] = MSG_BUS_DEV_RESET;
ahc->msgout_len = 1;
ahc->msgout_index = 0;
ahc->msg_type = MSG_TYPE_INITIATOR_MSGOUT;
ahc_outb(ahc, MSG_OUT, HOST_MSG);
ahc_outb(ahc, SCSISIGO, ahc_inb(ahc, LASTPHASE) | ATNO);
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
break;
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
case SEND_REJECT:
{
u_int rejbyte = ahc_inb(ahc, ACCUM);
printf("%s:%c:%d: Warning - unknown message received from "
"target (0x%x). Rejecting\n",
ahc_name(ahc), devinfo.channel, devinfo.target, rejbyte);
break;
}
case NO_IDENT:
{
/*
* The reconnecting target either did not send an identify
* message, or did, but we didn't find an SCB to match and
* before it could respond to our ATN/abort, it hit a dataphase.
* The only safe thing to do is to blow it away with a bus
* reset.
*/
int found;
printf("%s:%c:%d: Target did not send an IDENTIFY message. "
"LASTPHASE = 0x%x, SAVED_SCSIID == 0x%x\n",
ahc_name(ahc), devinfo.channel, devinfo.target,
ahc_inb(ahc, LASTPHASE), ahc_inb(ahc, SAVED_SCSIID));
found = ahc_reset_channel(ahc, devinfo.channel,
/*initiate reset*/TRUE);
printf("%s: Issued Channel %c Bus Reset. "
"%d SCBs aborted\n", ahc_name(ahc), devinfo.channel,
found);
return;
}
case IGN_WIDE_RES:
ahc_handle_ign_wide_residue(ahc, &devinfo);
break;
case BAD_PHASE:
{
u_int lastphase;
lastphase = ahc_inb(ahc, LASTPHASE);
if (lastphase == P_BUSFREE) {
printf("%s:%c:%d: Missed busfree. Curphase = 0x%x\n",
ahc_name(ahc), devinfo.channel, devinfo.target,
ahc_inb(ahc, SCSISIGI));
restart_sequencer(ahc);
return;
} else {
printf("%s:%c:%d: unknown scsi bus phase %x. "
"Attempting to continue\n",
ahc_name(ahc), devinfo.channel, devinfo.target,
ahc_inb(ahc, SCSISIGI));
}
break;
}
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
case HOST_MSG_LOOP:
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
/*
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
* The sequencer has encountered a message phase
* that requires host assistance for completion.
* While handling the message phase(s), we will be
* notified by the sequencer after each byte is
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
* transfered so we can track bus phase changes.
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
*
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
* If this is the first time we've seen a HOST_MSG_LOOP
* interrupt, initialize the state of the host message
* loop.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if (ahc->msg_type == MSG_TYPE_NONE) {
u_int bus_phase;
bus_phase = ahc_inb(ahc, SCSISIGI) & PHASE_MASK;
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
if (bus_phase != P_MESGIN
&& bus_phase != P_MESGOUT) {
printf("ahc_intr: HOST_MSG_LOOP bad "
"phase 0x%x\n",
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
bus_phase);
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
/*
* Probably transitioned to bus free before
* we got here. Just punt the message.
*/
ahc_clear_intstat(ahc);
restart_sequencer(ahc);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
return;
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
}
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if (devinfo.role == ROLE_INITIATOR) {
struct scb *scb;
u_int scb_index;
scb_index = ahc_inb(ahc, SCB_TAG);
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
scb = &ahc->scb_data->scbarray[scb_index];
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if (bus_phase == P_MESGOUT)
ahc_setup_initiator_msgout(ahc,
&devinfo,
scb);
else {
ahc->msg_type =
MSG_TYPE_INITIATOR_MSGIN;
ahc->msgin_index = 0;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
} else {
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if (bus_phase == P_MESGOUT) {
ahc->msg_type =
MSG_TYPE_TARGET_MSGOUT;
ahc->msgin_index = 0;
} else
/* XXX Ever executed??? */
ahc_setup_target_msgin(ahc, &devinfo);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
}
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
/* Pass a NULL path so that handlers generate their own */
ahc_handle_message_phase(ahc, /*path*/NULL);
break;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
case PERR_DETECTED:
{
/*
* If we've cleared the parity error interrupt
* but the sequencer still believes that SCSIPERR
* is true, it must be that the parity error is
* for the currently presented byte on the bus,
* and we are not in a phase (data-in) where we will
* eventually ack this byte. Ack the byte and
* throw it away in the hope that the target will
* take us to message out to deliver the appropriate
* error message.
*/
if ((intstat & SCSIINT) == 0
&& (ahc_inb(ahc, SSTAT1) & SCSIPERR) != 0) {
u_int curphase;
/*
* The hardware will only let you ack bytes
* if the expected phase in SCSISIGO matches
* the current phase. Make sure this is
* currently the case.
*/
curphase = ahc_inb(ahc, SCSISIGI) & PHASE_MASK;
ahc_outb(ahc, LASTPHASE, curphase);
ahc_outb(ahc, SCSISIGO, curphase);
ahc_inb(ahc, SCSIDATL);
}
break;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
case DATA_OVERRUN:
{
/*
* When the sequencer detects an overrun, it
* places the controller in "BITBUCKET" mode
* and allows the target to complete its transfer.
* Unfortunately, none of the counters get updated
* when the controller is in this mode, so we have
* no way of knowing how large the overrun was.
*/
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
u_int scbindex = ahc_inb(ahc, SCB_TAG);
u_int lastphase = ahc_inb(ahc, LASTPHASE);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int i;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
scb = &ahc->scb_data->scbarray[scbindex];
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
for (i = 0; i < num_phases; i++) {
if (lastphase == phase_table[i].phase)
break;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
xpt_print_path(scb->ccb->ccb_h.path);
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
printf("data overrun detected %s."
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
" Tag == 0x%x.\n",
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
phase_table[i].phasemsg,
scb->hscb->tag);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
xpt_print_path(scb->ccb->ccb_h.path);
printf("%s seen Data Phase. Length = %d. NumSGs = %d.\n",
ahc_inb(ahc, SEQ_FLAGS) & DPHASE ? "Have" : "Haven't",
scb->ccb->csio.dxfer_len, scb->sg_count);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
if (scb->sg_count > 0) {
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
for (i = 0; i < scb->sg_count; i++) {
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
printf("sg[%d] - Addr 0x%x : Length %d\n",
i,
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
scb->sg_list[i].addr,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
scb->sg_list[i].len & AHC_SG_LEN_MASK);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
/*
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
* Set this and it will take effect when the
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
* target does a command complete.
*/
ahc_freeze_devq(ahc, scb->ccb->ccb_h.path);
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
ahcsetccbstatus(scb->ccb, CAM_DATA_RUN_ERR);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_freeze_ccb(scb->ccb);
break;
}
case TRACEPOINT:
{
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
printf("SAVED_SCSIID %x, SAVED_LUN %x, SCBPTR %x\n",
ahc_inb(ahc, SAVED_SCSIID), ahc_inb(ahc, SAVED_LUN),
ahc_inb(ahc, SCBPTR));
#if 0
printf("%s: SCB_DATAPTR = %x, SCB_DATACNT = %x\n",
ahc_name(ahc),
ahc_inb(ahc, SCB_DATAPTR)
| (ahc_inb(ahc, SCB_DATAPTR + 1) << 8)
| (ahc_inb(ahc, SCB_DATAPTR + 2) << 16)
| (ahc_inb(ahc, SCB_DATAPTR + 3) << 24),
ahc_inb(ahc, SCB_DATACNT)
| (ahc_inb(ahc, SCB_DATACNT + 1) << 8)
| (ahc_inb(ahc, SCB_DATACNT + 2) << 16)
| (ahc_inb(ahc, SCB_DATACNT + 3) << 24));
printf("SCSIRATE = %x\n", ahc_inb(ahc, SCSIRATE));
printf("SG_CACHEPTR = %x\n", ahc_inb(ahc, SINDEX));
printf("DFCNTRL = %x, DFSTATUS = %x\n",
ahc_inb(ahc, DFCNTRL),
ahc_inb(ahc, DFSTATUS));
if ((ahc->features & AHC_CMD_CHAN) != 0) {
printf("CCHADDR = 0x%x\n",
ahc_inb(ahc, CCHADDR)
| (ahc_inb(ahc, CCHADDR + 1) << 8)
| (ahc_inb(ahc, CCHADDR + 2) << 16)
| (ahc_inb(ahc, CCHADDR + 3) << 24));
} else {
printf("HADDR = 0x%x\n",
ahc_inb(ahc, HADDR)
| (ahc_inb(ahc, HADDR + 1) << 8)
| (ahc_inb(ahc, HADDR + 2) << 16)
| (ahc_inb(ahc, HADDR + 3) << 24));
}
#endif
break;
}
case TRACEPOINT2:
{
printf("SINDEX = %x\n", ahc_inb(ahc, SINDEX));
printf("SCSIRATE = %x\n", ahc_inb(ahc, SCSIRATE));
#if 0
printf("SCB_RESIDUAL_SGPTR = %x, SCB_RESIDUAL_DATACNT = %x\n",
ahc_inb(ahc, SCB_RESIDUAL_SGPTR)
| (ahc_inb(ahc, SCB_RESIDUAL_SGPTR + 1) << 8)
| (ahc_inb(ahc, SCB_RESIDUAL_SGPTR + 2) << 16)
| (ahc_inb(ahc, SCB_RESIDUAL_SGPTR + 3) << 24),
ahc_inb(ahc, SCB_RESIDUAL_DATACNT)
| (ahc_inb(ahc, SCB_RESIDUAL_DATACNT + 1) << 8)
| (ahc_inb(ahc, SCB_RESIDUAL_DATACNT + 2) << 16)
| (ahc_inb(ahc, SCB_RESIDUAL_DATACNT + 3) << 24));
printf("DATA_COUNT_ODD = %x\n", ahc_inb(ahc, DATA_COUNT_ODD));
printf("SINDEX = %x\n", ahc_inb(ahc, SINDEX));
printf("SCB_SGPTR %x, SCB_RESIDUAL_SGPTR %x\n",
ahc_inb(ahc, SCB_SGPTR),
ahc_inb(ahc, SCB_RESIDUAL_SGPTR));
printf("SAVED_SCSIID %x, SAVED_LUN %d, "
"DISCONNECTED_SCBH %d\n",
ahc_inb(ahc, SAVED_SCSIID),
ahc_inb(ahc, SAVED_LUN),
ahc_inb(ahc, DISCONNECTED_SCBH));
int i;
if (ahc->unit != 1)
break;
for (i = 0; i < 32;) {
printf("0x%x 0x%x 0x%x 0x%x\n",
ahc_inb(ahc, SCB_CONTROL + i),
ahc_inb(ahc, SCB_CONTROL + i + 1),
ahc_inb(ahc, SCB_CONTROL + i + 2),
ahc_inb(ahc, SCB_CONTROL + i + 3));
i += 4;
}
#endif
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
#if 0
printf("SSTAT1 == 0x%x\n", ahc_inb(ahc, SSTAT1));
printf("SSTAT0 == 0x%x\n", ahc_inb(ahc, SSTAT0));
printf(", SCSISIGI == 0x%x\n", ahc_inb(ahc, SCSISIGI));
printf("TRACEPOINT: CCHCNT = %d, SG_COUNT = %d\n",
ahc_inb(ahc, CCHCNT), ahc_inb(ahc, SG_COUNT));
printf("TRACEPOINT: SCB_TAG = %d\n", ahc_inb(ahc, SCB_TAG));
printf("TRACEPOINT1: CCHADDR = %d, CCHCNT = %d, SCBPTR = %d\n",
ahc_inb(ahc, CCHADDR)
| (ahc_inb(ahc, CCHADDR+1) << 8)
| (ahc_inb(ahc, CCHADDR+2) << 16)
| (ahc_inb(ahc, CCHADDR+3) << 24),
ahc_inb(ahc, CCHCNT)
| (ahc_inb(ahc, CCHCNT+1) << 8)
| (ahc_inb(ahc, CCHCNT+2) << 16),
ahc_inb(ahc, SCBPTR));
printf("TRACEPOINT: WAITING_SCBH = %d\n", ahc_inb(ahc, WAITING_SCBH));
printf("TRACEPOINT: SCB_TAG = %d\n", ahc_inb(ahc, SCB_TAG));
#endif
break;
}
default:
printf("ahc_intr: seqint, "
"intstat == 0x%x, scsisigi = 0x%x\n",
intstat, ahc_inb(ahc, SCSISIGI));
break;
}
unpause:
/*
* The sequencer is paused immediately on
* a SEQINT, so we should restart it when
* we're done.
*/
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
unpause_sequencer(ahc);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
static void
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
ahc_handle_scsiint(struct ahc_softc *ahc, u_int intstat)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
u_int scb_index;
u_int status;
struct scb *scb;
-Wunused cleanup. Correct a problem where an external bus reset on the 'background' channel of a Twin Channel EISA controller could put the driver into an infinite loop. Noticed by: Twin Channel bug, Joerg Wunsch <joerg@FreeBSD.org> Submitted by: -Wunused, Poul-Henning Kamp <phk@freebsd.org>
1998-10-15 23:49:27 +00:00
char cur_channel;
char intr_channel;
if ((ahc->features & AHC_TWIN) != 0
&& ((ahc_inb(ahc, SBLKCTL) & SELBUSB) != 0))
cur_channel = 'B';
else
cur_channel = 'A';
intr_channel = cur_channel;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
status = ahc_inb(ahc, SSTAT1);
-Wunused cleanup. Correct a problem where an external bus reset on the 'background' channel of a Twin Channel EISA controller could put the driver into an infinite loop. Noticed by: Twin Channel bug, Joerg Wunsch <joerg@FreeBSD.org> Submitted by: -Wunused, Poul-Henning Kamp <phk@freebsd.org>
1998-10-15 23:49:27 +00:00
if (status == 0) {
if ((ahc->features & AHC_TWIN) != 0) {
/* Try the other channel */
ahc_outb(ahc, SBLKCTL, ahc_inb(ahc, SBLKCTL) ^ SELBUSB);
status = ahc_inb(ahc, SSTAT1);
ahc_outb(ahc, SBLKCTL, ahc_inb(ahc, SBLKCTL) ^ SELBUSB);
intr_channel = (cur_channel == 'A') ? 'B' : 'A';
}
if (status == 0) {
printf("%s: Spurious SCSI interrupt\n", ahc_name(ahc));
return;
}
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
-Wunused cleanup. Correct a problem where an external bus reset on the 'background' channel of a Twin Channel EISA controller could put the driver into an infinite loop. Noticed by: Twin Channel bug, Joerg Wunsch <joerg@FreeBSD.org> Submitted by: -Wunused, Poul-Henning Kamp <phk@freebsd.org>
1998-10-15 23:49:27 +00:00
scb_index = ahc_inb(ahc, SCB_TAG);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (scb_index < ahc->scb_data->numscbs) {
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
scb = &ahc->scb_data->scbarray[scb_index];
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
if ((scb->flags & SCB_ACTIVE) == 0
|| (ahc_inb(ahc, SEQ_FLAGS) & IDENTIFY_SEEN) == 0)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
scb = NULL;
} else
scb = NULL;
if ((status & SCSIRSTI) != 0) {
printf("%s: Someone reset channel %c\n",
-Wunused cleanup. Correct a problem where an external bus reset on the 'background' channel of a Twin Channel EISA controller could put the driver into an infinite loop. Noticed by: Twin Channel bug, Joerg Wunsch <joerg@FreeBSD.org> Submitted by: -Wunused, Poul-Henning Kamp <phk@freebsd.org>
1998-10-15 23:49:27 +00:00
ahc_name(ahc), intr_channel);
ahc_reset_channel(ahc, intr_channel, /* Initiate Reset */FALSE);
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
} else if ((status & SCSIPERR) != 0) {
/*
* Determine the bus phase and queue an appropriate message.
* SCSIPERR is latched true as soon as a parity error
* occurs. If the sequencer acked the transfer that
* caused the parity error and the currently presented
* transfer on the bus has correct parity, SCSIPERR will
* be cleared by CLRSCSIPERR. Use this to determine if
* we should look at the last phase the sequencer recorded,
* or the current phase presented on the bus.
*/
u_int mesg_out;
u_int curphase;
u_int errorphase;
u_int lastphase;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int i;
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
lastphase = ahc_inb(ahc, LASTPHASE);
curphase = ahc_inb(ahc, SCSISIGI) & PHASE_MASK;
ahc_outb(ahc, CLRSINT1, CLRSCSIPERR);
/*
* For all phases save DATA, the sequencer won't
* automatically ack a byte that has a parity error
* in it. So the only way that the current phase
* could be 'data-in' is if the parity error is for
* an already acked byte in the data phase. During
* synchronous data-in transfers, we may actually
* ack bytes before latching the current phase in
* LASTPHASE, leading to the discrepancy between
* curphase and lastphase.
*/
if ((ahc_inb(ahc, SSTAT1) & SCSIPERR) != 0
|| curphase == P_DATAIN)
errorphase = curphase;
else
errorphase = lastphase;
for (i = 0; i < num_phases; i++) {
if (errorphase == phase_table[i].phase)
break;
}
mesg_out = phase_table[i].mesg_out;
if (scb != NULL)
xpt_print_path(scb->ccb->ccb_h.path);
else
printf("%s:%c:%d: ", ahc_name(ahc),
intr_channel,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
SCSIID_TARGET(ahc, ahc_inb(ahc, SAVED_SCSIID)));
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
printf("parity error detected %s. "
"SEQADDR(0x%x) SCSIRATE(0x%x)\n",
phase_table[i].phasemsg,
ahc_inb(ahc, SEQADDR0) | (ahc_inb(ahc, SEQADDR1) << 8),
ahc_inb(ahc, SCSIRATE));
/*
* We've set the hardware to assert ATN if we
* get a parity error on "in" phases, so all we
* need to do is stuff the message buffer with
* the appropriate message. "In" phases have set
* mesg_out to something other than MSG_NOP.
*/
if (mesg_out != MSG_NOOP) {
if (ahc->msg_type != MSG_TYPE_NONE)
ahc->send_msg_perror = TRUE;
else
ahc_outb(ahc, MSG_OUT, mesg_out);
}
ahc_outb(ahc, CLRINT, CLRSCSIINT);
unpause_sequencer(ahc);
} else if ((status & BUSFREE) != 0
&& (ahc_inb(ahc, SIMODE1) & ENBUSFREE) != 0) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* First look at what phase we were last in.
* If its message out, chances are pretty good
* that the busfree was in response to one of
* our abort requests.
*/
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
u_int lastphase = ahc_inb(ahc, LASTPHASE);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int saved_scsiid = ahc_inb(ahc, SAVED_SCSIID);
u_int saved_lun = ahc_inb(ahc, SAVED_LUN);
u_int target = SCSIID_TARGET(ahc, saved_scsiid);
u_int initiator_role_id = SCSIID_OUR_ID(saved_scsiid);
char channel = SCSIID_CHANNEL(ahc, saved_scsiid);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
int printerror = 1;
ahc_outb(ahc, SCSISEQ,
ahc_inb(ahc, SCSISEQ) & (ENSELI|ENRSELI|ENAUTOATNP));
if (lastphase == P_MESGOUT) {
u_int message;
u_int tag;
Use TCL_CHANNEL/SCSI_ID/etc. macros in more locations. Don't mess with the IRQMS bit in the host control register unless we are an aic7770 chip. Use calling context to determine if the card is already paused when we update the target message request bit field in controller scratch ram. Looking at the paused bit in the HCNTRL register opened up a race condition. Insert delays in the target message request update routine as a temporary work around for what looks like a chip bug. I'm still investigating this one. Fix the Abort/Abort Tag/BDR handler to pull its message from the message buffer in our softc instead of attempting to get it from a register on the controller. The message is never recorded by the controller in the new message scheme. Don't rely on having an SCB when a BDR occurs. We can issue these during invalid reconnects to. Fix a few cases where we were restarting the sequencer but then still falling out of a switch statement to unpause the sequencer again. This could cause us to mess up sequencer state if it generated another pausing interrupt between the time of the restart and unpause. Kill the 'transceiver settle' loop during card initialization. I failed to realize that a controller that is not connected to any cables will never settle or enable the SCSI transceivers at all. The correct solution is to monitor the IOERR interrupt which indicates that the transceiver state has changed (UW<->LVD). Modify the aic7xxx assembler to properly echo input when stdin is not a tty.
1999-03-23 07:24:29 +00:00
message = ahc->msgout_buf[ahc->msgout_index - 1];
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
tag = SCB_LIST_NULL;
switch (message) {
case MSG_ABORT_TAG:
tag = scb->hscb->tag;
/* FALLTRHOUGH */
case MSG_ABORT:
xpt_print_path(scb->ccb->ccb_h.path);
printf("SCB %d - Abort %s Completed.\n",
Fix compile warnings.
1998-09-16 03:26:10 +00:00
scb->hscb->tag, tag == SCB_LIST_NULL ?
"" : "Tag");
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
ahc_abort_scbs(ahc, target, channel,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
saved_lun, tag,
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
ROLE_INITIATOR,
CAM_REQ_ABORTED);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
printerror = 0;
break;
case MSG_BUS_DEV_RESET:
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
{
struct ahc_devinfo devinfo;
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
/*
* Don't mark the user's request for this BDR
* as completing with CAM_BDR_SENT. CAM3
* specifies CAM_REQ_CMP.
*/
if (scb != NULL
&& scb->ccb->ccb_h.func_code == XPT_RESET_DEV
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
&& ahc_match_scb(ahc, scb, target, channel,
saved_lun,
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
SCB_LIST_NULL,
ROLE_INITIATOR)) {
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
ahcsetccbstatus(scb->ccb, CAM_REQ_CMP);
}
Use TCL_CHANNEL/SCSI_ID/etc. macros in more locations. Don't mess with the IRQMS bit in the host control register unless we are an aic7770 chip. Use calling context to determine if the card is already paused when we update the target message request bit field in controller scratch ram. Looking at the paused bit in the HCNTRL register opened up a race condition. Insert delays in the target message request update routine as a temporary work around for what looks like a chip bug. I'm still investigating this one. Fix the Abort/Abort Tag/BDR handler to pull its message from the message buffer in our softc instead of attempting to get it from a register on the controller. The message is never recorded by the controller in the new message scheme. Don't rely on having an SCB when a BDR occurs. We can issue these during invalid reconnects to. Fix a few cases where we were restarting the sequencer but then still falling out of a switch statement to unpause the sequencer again. This could cause us to mess up sequencer state if it generated another pausing interrupt between the time of the restart and unpause. Kill the 'transceiver settle' loop during card initialization. I failed to realize that a controller that is not connected to any cables will never settle or enable the SCSI transceivers at all. The correct solution is to monitor the IOERR interrupt which indicates that the transceiver state has changed (UW<->LVD). Modify the aic7xxx assembler to properly echo input when stdin is not a tty.
1999-03-23 07:24:29 +00:00
ahc_compile_devinfo(&devinfo,
initiator_role_id,
target,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
saved_lun,
Use TCL_CHANNEL/SCSI_ID/etc. macros in more locations. Don't mess with the IRQMS bit in the host control register unless we are an aic7770 chip. Use calling context to determine if the card is already paused when we update the target message request bit field in controller scratch ram. Looking at the paused bit in the HCNTRL register opened up a race condition. Insert delays in the target message request update routine as a temporary work around for what looks like a chip bug. I'm still investigating this one. Fix the Abort/Abort Tag/BDR handler to pull its message from the message buffer in our softc instead of attempting to get it from a register on the controller. The message is never recorded by the controller in the new message scheme. Don't rely on having an SCB when a BDR occurs. We can issue these during invalid reconnects to. Fix a few cases where we were restarting the sequencer but then still falling out of a switch statement to unpause the sequencer again. This could cause us to mess up sequencer state if it generated another pausing interrupt between the time of the restart and unpause. Kill the 'transceiver settle' loop during card initialization. I failed to realize that a controller that is not connected to any cables will never settle or enable the SCSI transceivers at all. The correct solution is to monitor the IOERR interrupt which indicates that the transceiver state has changed (UW<->LVD). Modify the aic7xxx assembler to properly echo input when stdin is not a tty.
1999-03-23 07:24:29 +00:00
channel,
ROLE_INITIATOR);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
ahc_handle_devreset(ahc, &devinfo,
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
CAM_BDR_SENT, AC_SENT_BDR,
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
"Bus Device Reset",
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
/*verbose_level*/0);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
printerror = 0;
break;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
default:
break;
}
}
if (printerror != 0) {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int i;
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (scb != NULL) {
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
u_int tag;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if ((scb->hscb->control & TAG_ENB) != 0)
tag = scb->hscb->tag;
else
tag = SCB_LIST_NULL;
ahc_abort_scbs(ahc, target, channel,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
SCB_GET_LUN(scb), tag,
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
ROLE_INITIATOR,
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
CAM_UNEXP_BUSFREE);
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
xpt_print_path(scb->ccb->ccb_h.path);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
} else {
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
/*
* We had not fully identified this connection,
* so we cannot abort anything.
*/
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
printf("%s: ", ahc_name(ahc));
}
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
for (i = 0; i < num_phases; i++) {
if (lastphase == phase_table[i].phase)
break;
}
printf("Unexpected busfree %s\n"
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
"SEQADDR == 0x%x\n",
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
phase_table[i].phasemsg, ahc_inb(ahc, SEQADDR0)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
| (ahc_inb(ahc, SEQADDR1) << 8));
}
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc_clear_msg_state(ahc);
ahc_outb(ahc, SIMODE1, ahc_inb(ahc, SIMODE1) & ~ENBUSFREE);
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
ahc_outb(ahc, CLRSINT1, CLRBUSFREE|CLRSCSIPERR);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_outb(ahc, CLRINT, CLRSCSIINT);
restart_sequencer(ahc);
} else if ((status & SELTO) != 0) {
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
u_int scbptr;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
scbptr = ahc_inb(ahc, WAITING_SCBH);
ahc_outb(ahc, SCBPTR, scbptr);
scb_index = ahc_inb(ahc, SCB_TAG);
if (scb_index < ahc->scb_data->numscbs) {
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
scb = &ahc->scb_data->scbarray[scb_index];
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if ((scb->flags & SCB_ACTIVE) == 0)
scb = NULL;
} else
scb = NULL;
if (scb == NULL) {
printf("%s: ahc_intr - referenced scb not "
"valid during SELTO scb(%d, %d)\n",
ahc_name(ahc), scbptr, scb_index);
} else {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahcsetccbstatus(scb->ccb, CAM_SEL_TIMEOUT);
ahc_freeze_devq(ahc, scb->ccb->ccb_h.path);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
/* Stop the selection */
ahc_outb(ahc, SCSISEQ, 0);
Clear the SELINGO bit after a selection timeout occurs. SELINGO is usually cleared by a successful selection, but there is no guarantee that a future successful selection will ever occur (e.g. empty bus). The driver never looks at SELINGO, but the busy LED does, so this change has the cosmetic effect of fixing the rare instance where the busy LED was left on, confusing the user.
1999-09-20 19:04:22 +00:00
/* No more pending messages */
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc_clear_msg_state(ahc);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Clear the SELINGO bit after a selection timeout occurs. SELINGO is usually cleared by a successful selection, but there is no guarantee that a future successful selection will ever occur (e.g. empty bus). The driver never looks at SELINGO, but the busy LED does, so this change has the cosmetic effect of fixing the rare instance where the busy LED was left on, confusing the user.
1999-09-20 19:04:22 +00:00
/*
* Although the driver does not care about the
* 'Selection in Progress' status bit, the busy
* LED does. SELINGO is only cleared by a sucessful
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
* selection, so we must manually clear it to insure
Clear the SELINGO bit after a selection timeout occurs. SELINGO is usually cleared by a successful selection, but there is no guarantee that a future successful selection will ever occur (e.g. empty bus). The driver never looks at SELINGO, but the busy LED does, so this change has the cosmetic effect of fixing the rare instance where the busy LED was left on, confusing the user.
1999-09-20 19:04:22 +00:00
* the LED turns off just incase no future successful
* selections occur (e.g. no devices on the bus).
*/
ahc_outb(ahc, CLRSINT0, CLRSELINGO);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Clear the SELINGO bit after a selection timeout occurs. SELINGO is usually cleared by a successful selection, but there is no guarantee that a future successful selection will ever occur (e.g. empty bus). The driver never looks at SELINGO, but the busy LED does, so this change has the cosmetic effect of fixing the rare instance where the busy LED was left on, confusing the user.
1999-09-20 19:04:22 +00:00
/* Clear interrupt state */
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
ahc_outb(ahc, CLRSINT1, CLRSELTIMEO|CLRBUSFREE|CLRSCSIPERR);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_outb(ahc, CLRINT, CLRSCSIINT);
restart_sequencer(ahc);
} else {
xpt_print_path(scb->ccb->ccb_h.path);
printf("Unknown SCSIINT. Status = 0x%x\n", status);
ahc_outb(ahc, CLRSINT1, status);
ahc_outb(ahc, CLRINT, CLRSCSIINT);
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
unpause_sequencer(ahc);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
}
static void
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc_build_transfer_msg(struct ahc_softc *ahc, struct ahc_devinfo *devinfo)
{
/*
* We need to initiate transfer negotiations.
* If our current and goal settings are identical,
* we want to renegotiate due to a check condition.
*/
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
struct ahc_initiator_tinfo *tinfo;
struct tmode_tstate *tstate;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
struct ahc_syncrate *rate;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
int dowide;
int dosync;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
int doppr;
int use_ppr;
u_int period;
u_int ppr_options;
u_int offset;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
tinfo = ahc_fetch_transinfo(ahc, devinfo->channel, devinfo->our_scsiid,
devinfo->target, &tstate);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
dowide = tinfo->current.width != tinfo->goal.width;
dosync = tinfo->current.period != tinfo->goal.period;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
doppr = tinfo->current.ppr_options != tinfo->goal.ppr_options;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (!dowide && !dosync && !doppr) {
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
dowide = tinfo->goal.width != MSG_EXT_WDTR_BUS_8_BIT;
dosync = tinfo->goal.period != 0;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
doppr = tinfo->goal.ppr_options != 0;
}
if (!dowide && !dosync && !doppr) {
panic("ahc_intr: AWAITING_MSG for negotiation, "
"but no negotiation needed\n");
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
use_ppr = (tinfo->current.transport_version >= 3) || doppr;
if (use_ppr) {
ahc_construct_ppr(ahc, tinfo->goal.period, tinfo->goal.offset,
tinfo->goal.width, tinfo->goal.ppr_options);
} else if (dowide) {
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc_construct_wdtr(ahc, tinfo->goal.width);
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
} else if (dosync) {
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
period = tinfo->goal.period;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ppr_options = 0;
rate = ahc_devlimited_syncrate(ahc, &period, &ppr_options);
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
offset = tinfo->goal.offset;
ahc_validate_offset(ahc, rate, &offset,
tinfo->current.width);
ahc_construct_sdtr(ahc, period, offset);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
}
}
static void
ahc_setup_initiator_msgout(struct ahc_softc *ahc, struct ahc_devinfo *devinfo,
struct scb *scb)
{
/*
* To facilitate adding multiple messages together,
* each routine should increment the index and len
* variables instead of setting them explicitly.
*/
ahc->msgout_index = 0;
ahc->msgout_len = 0;
if ((scb->flags & SCB_DEVICE_RESET) == 0
&& ahc_inb(ahc, MSG_OUT) == MSG_IDENTIFYFLAG) {
u_int identify_msg;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
identify_msg = MSG_IDENTIFYFLAG | SCB_GET_LUN(scb);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if ((scb->hscb->control & DISCENB) != 0)
identify_msg |= MSG_IDENTIFY_DISCFLAG;
ahc->msgout_buf[ahc->msgout_index++] = identify_msg;
ahc->msgout_len++;
if ((scb->hscb->control & TAG_ENB) != 0) {
ahc->msgout_buf[ahc->msgout_index++] =
scb->ccb->csio.tag_action;
ahc->msgout_buf[ahc->msgout_index++] = scb->hscb->tag;
ahc->msgout_len += 2;
}
}
if (scb->flags & SCB_DEVICE_RESET) {
ahc->msgout_buf[ahc->msgout_index++] = MSG_BUS_DEV_RESET;
ahc->msgout_len++;
xpt_print_path(scb->ccb->ccb_h.path);
printf("Bus Device Reset Message Sent\n");
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
} else if ((scb->flags & SCB_ABORT) != 0) {
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if ((scb->hscb->control & TAG_ENB) != 0)
ahc->msgout_buf[ahc->msgout_index++] = MSG_ABORT_TAG;
else
ahc->msgout_buf[ahc->msgout_index++] = MSG_ABORT;
ahc->msgout_len++;
xpt_print_path(scb->ccb->ccb_h.path);
printf("Abort Message Sent\n");
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
} else if ((ahc->targ_msg_req & devinfo->target_mask) != 0
|| (scb->flags & SCB_NEGOTIATE) != 0) {
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc_build_transfer_msg(ahc, devinfo);
} else {
printf("ahc_intr: AWAITING_MSG for an SCB that "
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
"does not have a waiting message\n");
printf("SCSIID = %x, target_mask = %x\n", scb->hscb->scsiid,
devinfo->target_mask);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
panic("SCB = %d, SCB Control = %x, MSG_OUT = %x "
"SCB flags = %x", scb->hscb->tag, scb->hscb->control,
ahc_inb(ahc, MSG_OUT), scb->flags);
}
/*
* Clear the MK_MESSAGE flag from the SCB so we aren't
* asked to send this message again.
*/
ahc_outb(ahc, SCB_CONTROL, ahc_inb(ahc, SCB_CONTROL) & ~MK_MESSAGE);
ahc->msgout_index = 0;
ahc->msg_type = MSG_TYPE_INITIATOR_MSGOUT;
}
static void
ahc_setup_target_msgin(struct ahc_softc *ahc, struct ahc_devinfo *devinfo)
{
/*
* To facilitate adding multiple messages together,
* each routine should increment the index and len
* variables instead of setting them explicitly.
*/
ahc->msgout_index = 0;
ahc->msgout_len = 0;
if ((ahc->targ_msg_req & devinfo->target_mask) != 0)
ahc_build_transfer_msg(ahc, devinfo);
else
panic("ahc_intr: AWAITING target message with no message");
ahc->msgout_index = 0;
ahc->msg_type = MSG_TYPE_TARGET_MSGIN;
}
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
static int
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc_handle_msg_reject(struct ahc_softc *ahc, struct ahc_devinfo *devinfo)
{
/*
* What we care about here is if we had an
* outstanding SDTR or WDTR message for this
* target. If we did, this is a signal that
* the target is refusing negotiation.
*/
struct scb *scb;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
struct ahc_initiator_tinfo *tinfo;
struct tmode_tstate *tstate;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
u_int scb_index;
u_int last_msg;
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
int response = 0;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
scb_index = ahc_inb(ahc, SCB_TAG);
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
scb = &ahc->scb_data->scbarray[scb_index];
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
tinfo = ahc_fetch_transinfo(ahc, devinfo->channel,
devinfo->our_scsiid,
devinfo->target, &tstate);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
/* Might be necessary */
last_msg = ahc_inb(ahc, LAST_MSG);
if (ahc_sent_msg(ahc, MSG_EXT_WDTR, /*full*/FALSE)) {
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
/* note 8bit xfers */
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
printf("%s:%c:%d: refuses WIDE negotiation. Using "
"8bit transfers\n", ahc_name(ahc),
devinfo->channel, devinfo->target);
ahc_set_width(ahc, devinfo, scb->ccb->ccb_h.path,
MSG_EXT_WDTR_BUS_8_BIT,
Correct a bug where the chip could be unpaused in the middle of a bus or device reset error recovery operation.
1999-05-18 03:58:49 +00:00
AHC_TRANS_ACTIVE|AHC_TRANS_GOAL,
/*paused*/TRUE);
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
/*
* No need to clear the sync rate. If the target
* did not accept the command, our syncrate is
* unaffected. If the target started the negotiation,
* but rejected our response, we already cleared the
* sync rate before sending our WDTR.
*/
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if (tinfo->goal.period) {
u_int period;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int ppr_options;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
/* Start the sync negotiation */
period = tinfo->goal.period;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ppr_options = 0;
ahc_devlimited_syncrate(ahc, &period, &ppr_options);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc->msgout_index = 0;
ahc->msgout_len = 0;
ahc_construct_sdtr(ahc, period, tinfo->goal.offset);
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
ahc->msgout_index = 0;
response = 1;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
}
} else if (ahc_sent_msg(ahc, MSG_EXT_SDTR, /*full*/FALSE)) {
/* note asynch xfers and clear flag */
ahc_set_syncrate(ahc, devinfo, scb->ccb->ccb_h.path,
/*syncrate*/NULL, /*period*/0,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
/*offset*/0, /*ppr_options*/0,
Correct a bug where the chip could be unpaused in the middle of a bus or device reset error recovery operation.
1999-05-18 03:58:49 +00:00
AHC_TRANS_ACTIVE|AHC_TRANS_GOAL,
/*paused*/TRUE);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
printf("%s:%c:%d: refuses synchronous negotiation. "
"Using asynchronous transfers\n",
ahc_name(ahc),
devinfo->channel, devinfo->target);
} else if ((scb->hscb->control & MSG_SIMPLE_Q_TAG) != 0) {
struct ccb_trans_settings neg;
printf("%s:%c:%d: refuses tagged commands. Performing "
"non-tagged I/O\n", ahc_name(ahc),
devinfo->channel, devinfo->target);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
ahc_set_tags(ahc, devinfo, FALSE);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
neg.flags = CCB_TRANS_CURRENT_SETTINGS;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
neg.valid = CCB_TRANS_TQ_VALID;
xpt_setup_ccb(&neg.ccb_h, scb->ccb->ccb_h.path, /*priority*/1);
xpt_async(AC_TRANSFER_NEG, scb->ccb->ccb_h.path, &neg);
/*
* Resend the identify for this CCB as the target
* may believe that the selection is invalid otherwise.
*/
ahc_outb(ahc, SCB_CONTROL, ahc_inb(ahc, SCB_CONTROL)
& ~MSG_SIMPLE_Q_TAG);
scb->hscb->control &= ~MSG_SIMPLE_Q_TAG;
scb->ccb->ccb_h.flags &= ~CAM_TAG_ACTION_VALID;
ahc_outb(ahc, MSG_OUT, MSG_IDENTIFYFLAG);
ahc_outb(ahc, SCSISIGO, ahc_inb(ahc, SCSISIGO) | ATNO);
/*
* Requeue all tagged commands for this target
* currently in our posession so they can be
* converted to untagged commands.
*/
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahc_search_qinfifo(ahc, SCB_GET_TARGET(ahc, scb),
SCB_GET_CHANNEL(ahc, scb),
SCB_GET_LUN(scb), /*tag*/SCB_LIST_NULL,
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
ROLE_INITIATOR, CAM_REQUEUE_REQ,
SEARCH_COMPLETE);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
} else {
/*
* Otherwise, we ignore it.
*/
printf("%s:%c:%d: Message reject for %x -- ignored\n",
ahc_name(ahc), devinfo->channel, devinfo->target,
last_msg);
}
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
return (response);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
}
static void
ahc_clear_msg_state(struct ahc_softc *ahc)
{
ahc->msgout_len = 0;
ahc->msgin_index = 0;
ahc->msg_type = MSG_TYPE_NONE;
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
ahc_outb(ahc, MSG_OUT, MSG_NOOP);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
}
static void
ahc_handle_message_phase(struct ahc_softc *ahc, struct cam_path *path)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
struct ahc_devinfo devinfo;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
u_int bus_phase;
int end_session;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_fetch_devinfo(ahc, &devinfo);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
end_session = FALSE;
bus_phase = ahc_inb(ahc, SCSISIGI) & PHASE_MASK;
reswitch:
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
switch (ahc->msg_type) {
case MSG_TYPE_INITIATOR_MSGOUT:
{
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
int lastbyte;
int phasemis;
int msgdone;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if (ahc->msgout_len == 0)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
panic("REQINIT interrupt with no active message");
phasemis = bus_phase != P_MESGOUT;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if (phasemis) {
if (bus_phase == P_MESGIN) {
/*
* Change gears and see if
* this messages is of interest to
* us or should be passed back to
* the sequencer.
*/
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
ahc_outb(ahc, CLRSINT1, CLRATNO);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc->send_msg_perror = FALSE;
ahc->msg_type = MSG_TYPE_INITIATOR_MSGIN;
ahc->msgin_index = 0;
goto reswitch;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
end_session = TRUE;
break;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if (ahc->send_msg_perror) {
ahc_outb(ahc, CLRSINT1, CLRATNO);
ahc_outb(ahc, CLRSINT1, CLRREQINIT);
ahc_outb(ahc, SCSIDATL, MSG_PARITY_ERROR);
break;
}
msgdone = ahc->msgout_index == ahc->msgout_len;
if (msgdone) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
* The target has requested a retry.
* Re-assert ATN, reset our message index to
* 0, and try again.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc->msgout_index = 0;
ahc_outb(ahc, SCSISIGO, ahc_inb(ahc, SCSISIGO) | ATNO);
}
lastbyte = ahc->msgout_index == (ahc->msgout_len - 1);
if (lastbyte) {
/* Last byte is signified by dropping ATN */
ahc_outb(ahc, CLRSINT1, CLRATNO);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
/*
* Clear our interrupt status and present
* the next byte on the bus.
*/
ahc_outb(ahc, CLRSINT1, CLRREQINIT);
ahc_outb(ahc, SCSIDATL, ahc->msgout_buf[ahc->msgout_index++]);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
break;
}
case MSG_TYPE_INITIATOR_MSGIN:
{
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
int phasemis;
int message_done;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
phasemis = bus_phase != P_MESGIN;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if (phasemis) {
ahc->msgin_index = 0;
if (bus_phase == P_MESGOUT
&& (ahc->send_msg_perror == TRUE
|| (ahc->msgout_len != 0
&& ahc->msgout_index == 0))) {
ahc->msg_type = MSG_TYPE_INITIATOR_MSGOUT;
goto reswitch;
}
end_session = TRUE;
break;
}
/* Pull the byte in without acking it */
ahc->msgin_buf[ahc->msgin_index] = ahc_inb(ahc, SCSIBUSL);
message_done = ahc_parse_msg(ahc, path, &devinfo);
if (message_done) {
/*
* Clear our incoming message buffer in case there
* is another message following this one.
*/
ahc->msgin_index = 0;
/*
* If this message illicited a response,
* assert ATN so the target takes us to the
* message out phase.
*/
if (ahc->msgout_len != 0)
ahc_outb(ahc, SCSISIGO,
ahc_inb(ahc, SCSISIGO) | ATNO);
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
} else
ahc->msgin_index++;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
/* Ack the byte */
ahc_outb(ahc, CLRSINT1, CLRREQINIT);
ahc_inb(ahc, SCSIDATL);
break;
}
case MSG_TYPE_TARGET_MSGIN:
{
int msgdone;
int msgout_request;
if (ahc->msgout_len == 0)
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
panic("Target MSGIN with no active message");
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
/*
* If we interrupted a mesgout session, the initiator
* will not know this until our first REQ. So, we
* only honor mesgout requests after we've sent our
* first byte.
*/
if ((ahc_inb(ahc, SCSISIGI) & ATNI) != 0
&& ahc->msgout_index > 0)
msgout_request = TRUE;
else
msgout_request = FALSE;
if (msgout_request) {
/*
* Change gears and see if
* this messages is of interest to
* us or should be passed back to
* the sequencer.
*/
ahc->msg_type = MSG_TYPE_TARGET_MSGOUT;
ahc_outb(ahc, SCSISIGO, P_MESGOUT | BSYO);
ahc->msgin_index = 0;
/* Dummy read to REQ for first byte */
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_inb(ahc, SCSIDATL);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc_outb(ahc, SXFRCTL0,
ahc_inb(ahc, SXFRCTL0) | SPIOEN);
break;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
msgdone = ahc->msgout_index == ahc->msgout_len;
if (msgdone) {
ahc_outb(ahc, SXFRCTL0,
ahc_inb(ahc, SXFRCTL0) & ~SPIOEN);
end_session = TRUE;
break;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
/*
* Present the next byte on the bus.
*/
ahc_outb(ahc, SXFRCTL0, ahc_inb(ahc, SXFRCTL0) | SPIOEN);
ahc_outb(ahc, SCSIDATL, ahc->msgout_buf[ahc->msgout_index++]);
break;
}
case MSG_TYPE_TARGET_MSGOUT:
{
int lastbyte;
int msgdone;
/*
* The initiator signals that this is
* the last byte by dropping ATN.
*/
lastbyte = (ahc_inb(ahc, SCSISIGI) & ATNI) == 0;
/*
* Read the latched byte, but turn off SPIOEN first
* so that we don't inadvertantly cause a REQ for the
* next byte.
*/
ahc_outb(ahc, SXFRCTL0, ahc_inb(ahc, SXFRCTL0) & ~SPIOEN);
ahc->msgin_buf[ahc->msgin_index] = ahc_inb(ahc, SCSIDATL);
msgdone = ahc_parse_msg(ahc, path, &devinfo);
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
if (msgdone == MSGLOOP_TERMINATED) {
/*
* The message is *really* done in that it caused
* us to go to bus free. The sequencer has already
* been reset at this point, so pull the ejection
* handle.
*/
return;
}
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc->msgin_index++;
/*
* XXX Read spec about initiator dropping ATN too soon
* and use msgdone to detect it.
*/
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
if (msgdone == MSGLOOP_MSGCOMPLETE) {
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc->msgin_index = 0;
/*
* If this message illicited a response, transition
* to the Message in phase and send it.
*/
if (ahc->msgout_len != 0) {
ahc_outb(ahc, SCSISIGO, P_MESGIN | BSYO);
ahc_outb(ahc, SXFRCTL0,
ahc_inb(ahc, SXFRCTL0) | SPIOEN);
ahc->msg_type = MSG_TYPE_TARGET_MSGIN;
ahc->msgin_index = 0;
break;
}
}
if (lastbyte)
end_session = TRUE;
else {
/* Ask for the next byte. */
ahc_outb(ahc, SXFRCTL0,
ahc_inb(ahc, SXFRCTL0) | SPIOEN);
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
break;
}
default:
panic("Unknown REQINIT message type");
}
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if (end_session) {
ahc_clear_msg_state(ahc);
ahc_outb(ahc, RETURN_1, EXIT_MSG_LOOP);
} else
ahc_outb(ahc, RETURN_1, CONT_MSG_LOOP);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
/*
* See if we sent a particular extended message to the target.
* If "full" is true, the target saw the full message.
* If "full" is false, the target saw at least the first
* byte of the message.
*/
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static int
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc_sent_msg(struct ahc_softc *ahc, u_int msgtype, int full)
{
int found;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int index;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
found = FALSE;
index = 0;
while (index < ahc->msgout_len) {
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
if (ahc->msgout_buf[index] == MSG_EXTENDED) {
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
/* Found a candidate */
if (ahc->msgout_buf[index+2] == msgtype) {
u_int end_index;
end_index = index + 1
+ ahc->msgout_buf[index + 1];
if (full) {
if (ahc->msgout_index > end_index)
found = TRUE;
} else if (ahc->msgout_index > index)
found = TRUE;
}
break;
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
} else if (ahc->msgout_buf[index] >= MSG_SIMPLE_Q_TAG
&& ahc->msgout_buf[index] <= MSG_IGN_WIDE_RESIDUE) {
/* Skip tag type and tag id or residue param*/
index += 2;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
} else {
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
/* Single byte message */
index++;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
}
}
return (found);
}
static int
ahc_parse_msg(struct ahc_softc *ahc, struct cam_path *path,
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
struct ahc_devinfo *devinfo)
{
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
struct ahc_initiator_tinfo *tinfo;
struct tmode_tstate *tstate;
int reject;
int done;
int response;
u_int targ_scsirate;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
done = MSGLOOP_IN_PROG;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
response = FALSE;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
reject = FALSE;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
tinfo = ahc_fetch_transinfo(ahc, devinfo->channel, devinfo->our_scsiid,
devinfo->target, &tstate);
targ_scsirate = tinfo->scsirate;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Parse as much of the message as is availible,
* rejecting it if we don't support it. When
* the entire message is availible and has been
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
* handled, return MSGLOOP_MSGCOMPLETE, indicating
* that we have parsed an entire message.
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
*
* In the case of extended messages, we accept the length
* byte outright and perform more checking once we know the
* extended message type.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
switch (ahc->msgin_buf[0]) {
case MSG_MESSAGE_REJECT:
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
response = ahc_handle_msg_reject(ahc, devinfo);
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
/* FALLTHROUGH */
case MSG_NOOP:
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
done = MSGLOOP_MSGCOMPLETE;
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
break;
case MSG_EXTENDED:
{
/* Wait for enough of the message to begin validation */
if (ahc->msgin_index < 2)
break;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
switch (ahc->msgin_buf[2]) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
case MSG_EXT_SDTR:
{
struct ahc_syncrate *syncrate;
u_int period;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int ppr_options;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
u_int offset;
u_int saved_offset;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if (ahc->msgin_buf[1] != MSG_EXT_SDTR_LEN) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
reject = TRUE;
break;
}
/*
* Wait until we have both args before validating
* and acting on this message.
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
*
* Add one to MSG_EXT_SDTR_LEN to account for
* the extended message preamble.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if (ahc->msgin_index < (MSG_EXT_SDTR_LEN + 1))
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
break;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
period = ahc->msgin_buf[3];
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ppr_options = 0;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
saved_offset = offset = ahc->msgin_buf[4];
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
syncrate = ahc_devlimited_syncrate(ahc, &period,
&ppr_options);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_validate_offset(ahc, syncrate, &offset,
targ_scsirate & WIDEXFER);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc_set_syncrate(ahc, devinfo, path,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
syncrate, period,
offset, ppr_options,
Correct a bug where the chip could be unpaused in the middle of a bus or device reset error recovery operation.
1999-05-18 03:58:49 +00:00
AHC_TRANS_ACTIVE|AHC_TRANS_GOAL,
/*paused*/TRUE);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* See if we initiated Sync Negotiation
* and didn't have to fall down to async
* transfers.
*/
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if (ahc_sent_msg(ahc, MSG_EXT_SDTR, /*full*/TRUE)) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/* We started it */
if (saved_offset != offset) {
/* Went too low - force async */
reject = TRUE;
}
} else {
/*
* Send our own SDTR in reply
*/
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if (bootverbose)
printf("Sending SDTR!\n");
ahc->msgout_index = 0;
ahc->msgout_len = 0;
ahc_construct_sdtr(ahc, period, offset);
ahc->msgout_index = 0;
response = TRUE;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
done = MSGLOOP_MSGCOMPLETE;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
break;
}
case MSG_EXT_WDTR:
{
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int bus_width;
u_int saved_width;
u_int sending_reply;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
sending_reply = FALSE;
if (ahc->msgin_buf[1] != MSG_EXT_WDTR_LEN) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
reject = TRUE;
break;
}
/*
* Wait until we have our arg before validating
* and acting on this message.
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
*
* Add one to MSG_EXT_WDTR_LEN to account for
* the extended message preamble.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if (ahc->msgin_index < (MSG_EXT_WDTR_LEN + 1))
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
break;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
bus_width = ahc->msgin_buf[3];
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
saved_width = bus_width;
ahc_validate_width(ahc, &bus_width);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if (ahc_sent_msg(ahc, MSG_EXT_WDTR, /*full*/TRUE)) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Don't send a WDTR back to the
* target, since we asked first.
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
* If the width went higher than our
* request, reject it.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (saved_width > bus_width) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
reject = TRUE;
printf("%s: target %d requested %dBit "
"transfers. Rejecting...\n",
ahc_name(ahc), devinfo->target,
8 * (0x01 << bus_width));
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
bus_width = 0;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
} else {
/*
* Send our own WDTR in reply
*/
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if (bootverbose)
printf("Sending WDTR!\n");
ahc->msgout_index = 0;
ahc->msgout_len = 0;
ahc_construct_wdtr(ahc, bus_width);
ahc->msgout_index = 0;
response = TRUE;
sending_reply = TRUE;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc_set_width(ahc, devinfo, path, bus_width,
Correct a bug where the chip could be unpaused in the middle of a bus or device reset error recovery operation.
1999-05-18 03:58:49 +00:00
AHC_TRANS_ACTIVE|AHC_TRANS_GOAL,
/*paused*/TRUE);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/* After a wide message, we are async */
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc_set_syncrate(ahc, devinfo, path,
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*syncrate*/NULL, /*period*/0,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
/*offset*/0, /*ppr_options*/0,
AHC_TRANS_ACTIVE, /*paused*/TRUE);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if (sending_reply == FALSE && reject == FALSE) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
/* XXX functionalize */
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (tinfo->goal.period) {
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
struct ahc_syncrate *rate;
u_int period;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int ppr;
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
u_int offset;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/* Start the sync negotiation */
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
period = tinfo->goal.period;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ppr = 0;
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
rate = ahc_devlimited_syncrate(ahc,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
&period,
&ppr);
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
offset = tinfo->goal.offset;
ahc_validate_offset(ahc, rate, &offset,
tinfo->current.width);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc->msgout_index = 0;
ahc->msgout_len = 0;
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
ahc_construct_sdtr(ahc, period, offset);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc->msgout_index = 0;
response = TRUE;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
done = MSGLOOP_MSGCOMPLETE;
break;
}
case MSG_EXT_PPR:
{
struct ahc_syncrate *syncrate;
u_int period;
u_int offset;
u_int bus_width;
u_int ppr_options;
u_int saved_width;
u_int saved_offset;
u_int saved_ppr_options;
if (ahc->msgin_buf[1] != MSG_EXT_PPR_LEN) {
reject = TRUE;
break;
}
/*
* Wait until we have all args before validating
* and acting on this message.
*
* Add one to MSG_EXT_PPR_LEN to account for
* the extended message preamble.
*/
if (ahc->msgin_index < (MSG_EXT_PPR_LEN + 1))
break;
period = ahc->msgin_buf[3];
offset = ahc->msgin_buf[5];
bus_width = ahc->msgin_buf[6];
saved_width = bus_width;
ppr_options = ahc->msgin_buf[7];
/*
* According to the spec, a DT only
* period factor with no DT option
* set implies async.
*/
if ((ppr_options & MSG_EXT_PPR_DT_REQ) == 0
&& period == 9)
offset = 0;
saved_ppr_options = ppr_options;
saved_offset = offset;
/*
* Mask out any options we don't support
* on any controller. Transfer options are
* only available if we are negotiating wide.
*/
ppr_options &= MSG_EXT_PPR_DT_REQ;
if (bus_width == 0)
ppr_options = 0;
ahc_validate_width(ahc, &bus_width);
syncrate = ahc_devlimited_syncrate(ahc, &period,
&ppr_options);
ahc_validate_offset(ahc, syncrate, &offset, bus_width);
if (ahc_sent_msg(ahc, MSG_EXT_PPR, /*full*/TRUE)) {
/*
* If we are unable to do any of the
* requested options (we went too low),
* then we'll have to reject the message.
*/
if (saved_width > bus_width
|| saved_offset != offset
|| saved_ppr_options != ppr_options)
reject = TRUE;
} else {
printf("Target Initated PPR detected!\n");
response = TRUE;
}
ahc_set_syncrate(ahc, devinfo, path,
syncrate, period,
offset, ppr_options,
AHC_TRANS_ACTIVE|AHC_TRANS_GOAL,
/*paused*/TRUE);
ahc_set_width(ahc, devinfo, path, bus_width,
AHC_TRANS_ACTIVE|AHC_TRANS_GOAL,
/*paused*/TRUE);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
break;
}
default:
/* Unknown extended message. Reject it. */
reject = TRUE;
break;
}
Add two missing break statements.
1999-01-15 23:24:23 +00:00
break;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
case MSG_BUS_DEV_RESET:
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
ahc_handle_devreset(ahc, devinfo,
CAM_BDR_SENT, AC_SENT_BDR,
"Bus Device Reset Received",
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
/*verbose_level*/0);
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
restart_sequencer(ahc);
done = MSGLOOP_TERMINATED;
break;
case MSG_ABORT_TAG:
case MSG_ABORT:
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
case MSG_CLEAR_QUEUE:
/* Target mode messages */
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
if (devinfo->role != ROLE_TARGET) {
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
reject = TRUE;
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
break;
}
ahc_abort_scbs(ahc, devinfo->target, devinfo->channel,
devinfo->lun,
ahc->msgin_buf[0] == MSG_ABORT_TAG
? SCB_LIST_NULL
: ahc_inb(ahc, INITIATOR_TAG),
ROLE_TARGET, CAM_REQ_ABORTED);
tstate = ahc->enabled_targets[devinfo->our_scsiid];
if (tstate != NULL) {
struct tmode_lstate* lstate;
lstate = tstate->enabled_luns[devinfo->lun];
if (lstate != NULL) {
ahc_queue_lstate_event(ahc, lstate,
devinfo->our_scsiid,
ahc->msgin_buf[0],
/*arg*/0);
ahc_send_lstate_events(ahc, lstate);
}
}
done = MSGLOOP_MSGCOMPLETE;
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
break;
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
case MSG_TERM_IO_PROC:
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
default:
reject = TRUE;
break;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (reject) {
/*
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
* Setup to reject the message.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc->msgout_index = 0;
ahc->msgout_len = 1;
ahc->msgout_buf[0] = MSG_MESSAGE_REJECT;
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
done = MSGLOOP_MSGCOMPLETE;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
response = TRUE;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
if (done != MSGLOOP_IN_PROG && !response)
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
/* Clear the outgoing message buffer */
ahc->msgout_len = 0;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
return (done);
}
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
static void
ahc_handle_ign_wide_residue(struct ahc_softc *ahc, struct ahc_devinfo *devinfo)
{
u_int scb_index;
struct scb *scb;
scb_index = ahc_inb(ahc, SCB_TAG);
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
scb = &ahc->scb_data->scbarray[scb_index];
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
if ((ahc_inb(ahc, SEQ_FLAGS) & DPHASE) == 0
|| (scb->ccb->ccb_h.flags & CAM_DIR_MASK) != CAM_DIR_IN) {
/*
* Ignore the message if we haven't
* seen an appropriate data phase yet.
*/
} else {
/*
* If the residual occurred on the last
* transfer and the transfer request was
* expected to end on an odd count, do
* nothing. Otherwise, subtract a byte
* and update the residual count accordingly.
*/
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
uint32_t sgptr;
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
sgptr = ahc_inb(ahc, SCB_RESIDUAL_SGPTR);
if ((sgptr & SG_LIST_NULL) != 0
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
&& ahc_inb(ahc, DATA_COUNT_ODD) == 1) {
/*
* If the residual occurred on the last
* transfer and the transfer request was
* expected to end on an odd count, do
* nothing.
*/
} else {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
struct ahc_dma_seg *sg;
uint32_t data_cnt;
uint32_t data_addr;
/* Pull in the rest of the sgptr */
sgptr |= (ahc_inb(ahc, SCB_RESIDUAL_SGPTR + 3) << 24)
| (ahc_inb(ahc, SCB_RESIDUAL_SGPTR + 2) << 16)
| (ahc_inb(ahc, SCB_RESIDUAL_SGPTR + 1) << 8);
sgptr &= SG_PTR_MASK;
data_cnt = (ahc_inb(ahc, SCB_RESIDUAL_DATACNT+2) << 16)
| (ahc_inb(ahc, SCB_RESIDUAL_DATACNT+1) << 8)
| (ahc_inb(ahc, SCB_RESIDUAL_DATACNT));
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
data_addr = (ahc_inb(ahc, SHADDR + 3) << 24)
| (ahc_inb(ahc, SHADDR + 2) << 16)
| (ahc_inb(ahc, SHADDR + 1) << 8)
| (ahc_inb(ahc, SHADDR));
data_cnt += 1;
data_addr -= 1;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
sg = ahc_sg_bus_to_virt(scb, sgptr);
/*
* The residual sg ptr points to the next S/G
* to load so we must go back one.
*/
sg--;
if (sg != scb->sg_list
&& (sg->len & AHC_SG_LEN_MASK) < data_cnt) {
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
sg--;
data_cnt = 1 | (sg->len & AHC_DMA_LAST_SEG);
data_addr = sg->addr
+ (sg->len & AHC_SG_LEN_MASK) - 1;
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
/*
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
* Increment sg so it points to the
* "next" sg.
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
*/
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
sg++;
sgptr = ahc_sg_virt_to_bus(scb, sg);
ahc_outb(ahc, SCB_RESIDUAL_SGPTR + 3,
sgptr >> 24);
ahc_outb(ahc, SCB_RESIDUAL_SGPTR + 2,
sgptr >> 16);
ahc_outb(ahc, SCB_RESIDUAL_SGPTR + 1,
sgptr >> 8);
ahc_outb(ahc, SCB_RESIDUAL_SGPTR, sgptr);
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
/* XXX What about high address byte??? */
ahc_outb(ahc, SCB_RESIDUAL_DATACNT + 3, data_cnt >> 24);
ahc_outb(ahc, SCB_RESIDUAL_DATACNT + 2, data_cnt >> 16);
ahc_outb(ahc, SCB_RESIDUAL_DATACNT + 1, data_cnt >> 8);
ahc_outb(ahc, SCB_RESIDUAL_DATACNT, data_cnt);
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
/* XXX Perhaps better to just keep the saved address in sram */
if ((ahc->features & AHC_ULTRA2) != 0) {
ahc_outb(ahc, HADDR + 3, data_addr >> 24);
ahc_outb(ahc, HADDR + 2, data_addr >> 16);
ahc_outb(ahc, HADDR + 1, data_addr >> 8);
ahc_outb(ahc, HADDR, data_addr);
ahc_outb(ahc, DFCNTRL, PRELOADEN);
ahc_outb(ahc, SXFRCTL0,
ahc_inb(ahc, SXFRCTL0) | CLRCHN);
} else {
ahc_outb(ahc, SHADDR + 3, data_addr >> 24);
ahc_outb(ahc, SHADDR + 2, data_addr >> 16);
ahc_outb(ahc, SHADDR + 1, data_addr >> 8);
ahc_outb(ahc, SHADDR, data_addr);
}
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
}
}
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static void
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
ahc_handle_devreset(struct ahc_softc *ahc, struct ahc_devinfo *devinfo,
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
cam_status status, ac_code acode, char *message,
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
int verbose_level)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
struct cam_path *path;
int found;
int error;
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
struct tmode_tstate* tstate;
u_int lun;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
error = ahc_create_path(ahc, devinfo, &path);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
found = ahc_abort_scbs(ahc, devinfo->target, devinfo->channel,
CAM_LUN_WILDCARD, SCB_LIST_NULL, devinfo->role,
status);
aic7xxx.c: Clean out some #if 0'ed debugging cruft. aic7xxx.h: Definitions for the aic7855 and aic7859.
1999-08-30 16:12:39 +00:00
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
/*
* Send an immediate notify ccb to all target more peripheral
* drivers affected by this action.
*/
tstate = ahc->enabled_targets[devinfo->our_scsiid];
if (tstate != NULL) {
for (lun = 0; lun <= 7; lun++) {
struct tmode_lstate* lstate;
lstate = tstate->enabled_luns[lun];
if (lstate == NULL)
continue;
ahc_queue_lstate_event(ahc, lstate, devinfo->our_scsiid,
MSG_BUS_DEV_RESET, /*arg*/0);
ahc_send_lstate_events(ahc, lstate);
}
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Go back to async/narrow transfers and renegotiate.
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
* ahc_set_width and ahc_set_syncrate can cope with NULL
* paths.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
ahc_set_width(ahc, devinfo, path, MSG_EXT_WDTR_BUS_8_BIT,
Correct a bug where the chip could be unpaused in the middle of a bus or device reset error recovery operation.
1999-05-18 03:58:49 +00:00
AHC_TRANS_CUR, /*paused*/TRUE);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
ahc_set_syncrate(ahc, devinfo, path, /*syncrate*/NULL,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
/*period*/0, /*offset*/0, /*ppr_options*/0,
AHC_TRANS_CUR, /*paused*/TRUE);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (error == CAM_REQ_CMP && acode != 0)
xpt_async(AC_SENT_BDR, path, NULL);
if (error == CAM_REQ_CMP)
xpt_free_path(path);
if (message != NULL
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
&& (verbose_level <= bootverbose))
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
printf("%s: %s on %c:%d. %d SCBs aborted\n", ahc_name(ahc),
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
message, devinfo->channel, devinfo->target, found);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Add a missing newline.
1998-10-07 03:34:13 +00:00
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* We have an scb which has been processed by the
* adaptor, now we look to see how the operation
* went.
*/
static void
ahc_done(struct ahc_softc *ahc, struct scb *scb)
{
union ccb *ccb;
CAM_DEBUG(scb->ccb->ccb_h.path, CAM_DEBUG_TRACE,
("ahc_done - scb %d\n", scb->hscb->tag));
ccb = scb->ccb;
LIST_REMOVE(&ccb->ccb_h, sim_links.le);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (ccb->ccb_h.func_code == XPT_SCSI_IO
&& ((ccb->ccb_h.flags & CAM_TAG_ACTION_VALID) == 0
|| ccb->csio.tag_action == CAM_TAG_ACTION_NONE)
&& (ahc->features & AHC_SCB_BTT) == 0) {
struct scb_tailq *untagged_q;
untagged_q = &ahc->untagged_queues[ccb->ccb_h.target_id];
TAILQ_REMOVE(untagged_q, scb, links.tqe);
ahc_run_untagged_queue(ahc, untagged_q);
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
untimeout(ahc_timeout, (caddr_t)scb, ccb->ccb_h.timeout_ch);
if ((ccb->ccb_h.flags & CAM_DIR_MASK) != CAM_DIR_NONE) {
bus_dmasync_op_t op;
if ((ccb->ccb_h.flags & CAM_DIR_MASK) == CAM_DIR_IN)
op = BUS_DMASYNC_POSTREAD;
else
op = BUS_DMASYNC_POSTWRITE;
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
bus_dmamap_sync(ahc->buffer_dmat, scb->dmamap, op);
bus_dmamap_unload(ahc->buffer_dmat, scb->dmamap);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
if (ccb->ccb_h.func_code == XPT_CONT_TARGET_IO) {
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
if (ahc_ccb_status(ccb) == CAM_REQ_INPROG)
ccb->ccb_h.status |= CAM_REQ_CMP;
ccb->ccb_h.status &= ~CAM_SIM_QUEUED;
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
ahcfreescb(ahc, scb);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
xpt_done(ccb);
return;
}
/*
* If the recovery SCB completes, we have to be
* out of our timeout.
*/
if ((scb->flags & SCB_RECOVERY_SCB) != 0) {
struct ccb_hdr *ccbh;
/*
* We were able to complete the command successfully,
* so reinstate the timeouts for all other pending
* commands.
*/
ccbh = ahc->pending_ccbs.lh_first;
while (ccbh != NULL) {
struct scb *pending_scb;
pending_scb = (struct scb *)ccbh->ccb_scb_ptr;
ccbh->timeout_ch =
timeout(ahc_timeout, pending_scb,
(ccbh->timeout * hz)/1000);
ccbh = LIST_NEXT(ccbh, sim_links.le);
}
/*
* Ensure that we didn't put a second instance of this
* SCB into the QINFIFO.
*/
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahc_search_qinfifo(ahc, SCB_GET_TARGET(ahc, scb),
SCB_GET_CHANNEL(ahc, scb),
SCB_GET_LUN(scb), scb->hscb->tag,
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
ROLE_INITIATOR, /*status*/0,
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
SEARCH_REMOVE);
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
if (ahc_ccb_status(ccb) == CAM_BDR_SENT
|| ahc_ccb_status(ccb) == CAM_REQ_ABORTED)
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
ahcsetccbstatus(ccb, CAM_CMD_TIMEOUT);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
xpt_print_path(ccb->ccb_h.path);
printf("no longer in timeout, status = %x\n",
ccb->ccb_h.status);
}
/* Don't clobber any existing error state */
if (ahc_ccb_status(ccb) == CAM_REQ_INPROG) {
ccb->ccb_h.status |= CAM_REQ_CMP;
} else if ((scb->flags & SCB_SENSE) != 0) {
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
/*
* We performed autosense retrieval.
*
Update copyright. Correct some whitespace.
1999-05-17 21:53:51 +00:00
* bzero the sense data before having
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
* the drive fill it. The SCSI spec mandates
* that any untransfered data should be
* assumed to be zero. Complete the 'bounce'
* of sense information through buffers accessible
* via bus-space by copying it into the clients
* csio.
*/
bzero(&ccb->csio.sense_data, sizeof(ccb->csio.sense_data));
bcopy(&ahc->scb_data->sense[scb->hscb->tag],
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
&ccb->csio.sense_data,
scb->sg_list->len & AHC_SG_LEN_MASK);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
scb->ccb->ccb_h.status |= CAM_AUTOSNS_VALID;
}
ccb->ccb_h.status &= ~CAM_SIM_QUEUED;
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
ahcfreescb(ahc, scb);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
xpt_done(ccb);
}
/*
* Determine the number of SCBs available on the controller
*/
int
ahc_probe_scbs(struct ahc_softc *ahc) {
int i;
for (i = 0; i < AHC_SCB_MAX; i++) {
ahc_outb(ahc, SCBPTR, i);
ahc_outb(ahc, SCB_CONTROL, i);
if (ahc_inb(ahc, SCB_CONTROL) != i)
break;
ahc_outb(ahc, SCBPTR, 0);
if (ahc_inb(ahc, SCB_CONTROL) != 0)
break;
}
return (i);
}
/*
* Start the board, ready for normal operation
*/
int
ahc_init(struct ahc_softc *ahc)
{
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
int max_targ = 15;
int i;
int term;
u_int scsi_conf;
u_int scsiseq_template;
u_int ultraenb;
u_int discenable;
u_int tagenable;
size_t driver_data_size;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
uint32_t physaddr;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
#ifdef AHC_PRINT_SRAM
printf("Scratch Ram:");
for (i = 0x20; i < 0x5f; i++) {
if (((i % 8) == 0) && (i != 0)) {
printf ("\n ");
}
printf (" 0x%x", ahc_inb(ahc, i));
}
if ((ahc->features & AHC_MORE_SRAM) != 0) {
for (i = 0x70; i < 0x7f; i++) {
if (((i % 8) == 0) && (i != 0)) {
printf ("\n ");
}
printf (" 0x%x", ahc_inb(ahc, i));
}
}
printf ("\n");
#endif
/*
* Assume we have a board at this stage and it has been reset.
*/
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
if ((ahc->flags & AHC_USEDEFAULTS) != 0)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc->our_id = ahc->our_id_b = 7;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
/*
* Default to allowing initiator operations.
*/
ahc->flags |= AHC_INITIATORMODE;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* XXX Would be better to use a per device flag, but PCI and EISA
* devices don't have them yet.
*/
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if ((AHC_TMODE_ENABLE & (0x01 << ahc->unit)) != 0) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc->flags |= AHC_TARGETMODE;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
/*
* Although we have space for both the initiator and
* target roles on ULTRA2 chips, we currently disable
* the initiator role to allow multi-scsi-id target mode
* configurations. We can only respond on the same SCSI
* ID as our initiator role if we allow initiator operation.
* At some point, we should add a configuration knob to
* allow both roles to be loaded.
*/
ahc->flags &= ~AHC_INITIATORMODE;
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
}
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
/* DMA tag for mapping buffers into device visible space. */
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
if (bus_dma_tag_create(ahc->parent_dmat, /*alignment*/1, /*boundary*/0,
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
/*lowaddr*/BUS_SPACE_MAXADDR,
/*highaddr*/BUS_SPACE_MAXADDR,
/*filter*/NULL, /*filterarg*/NULL,
/*maxsize*/MAXBSIZE, /*nsegments*/AHC_NSEG,
/*maxsegsz*/AHC_MAXTRANSFER_SIZE,
/*flags*/BUS_DMA_ALLOCNOW,
&ahc->buffer_dmat) != 0) {
return (ENOMEM);
}
ahc->init_level++;
/*
* DMA tag for our command fifos and other data in system memory
* the card's sequencer must be able to access. For initiator
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
* roles, we need to allocate space for the the qinfifo and qoutfifo.
* The qinfifo and qoutfifo are composed of 256 1 byte elements.
* When providing for the target mode role, we additionally must
* provide space for the incoming target command fifo and an extra
* byte to deal with a dma bug in some chip versions.
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
*/
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
driver_data_size = 2 * 256 * sizeof(uint8_t);
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
if ((ahc->flags & AHC_TARGETMODE) != 0)
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
driver_data_size += AHC_TMODE_CMDS * sizeof(struct target_cmd)
+ /*DMA WideOdd Bug Buffer*/1;
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
if (bus_dma_tag_create(ahc->parent_dmat, /*alignment*/1, /*boundary*/0,
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
/*lowaddr*/BUS_SPACE_MAXADDR,
/*highaddr*/BUS_SPACE_MAXADDR,
/*filter*/NULL, /*filterarg*/NULL,
driver_data_size,
/*nsegments*/1,
/*maxsegsz*/BUS_SPACE_MAXSIZE_32BIT,
/*flags*/0, &ahc->shared_data_dmat) != 0) {
return (ENOMEM);
}
ahc->init_level++;
/* Allocation of driver data */
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (bus_dmamem_alloc(ahc->shared_data_dmat,
(void **)&ahc->qoutfifo,
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
BUS_DMA_NOWAIT, &ahc->shared_data_dmamap) != 0) {
return (ENOMEM);
}
ahc->init_level++;
/* And permanently map it in */
bus_dmamap_load(ahc->shared_data_dmat, ahc->shared_data_dmamap,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahc->qoutfifo, driver_data_size, ahcdmamapcb,
&ahc->shared_data_busaddr, /*flags*/0);
if ((ahc->flags & AHC_TARGETMODE) != 0) {
ahc->targetcmds = (struct target_cmd *)ahc->qoutfifo;
ahc->qoutfifo = (uint8_t *)&ahc->targetcmds[256];
ahc->dma_bug_buf = ahc->shared_data_busaddr
+ driver_data_size - 1;
/* All target command blocks start out invalid. */
for (i = 0; i < AHC_TMODE_CMDS; i++)
ahc->targetcmds[i].cmd_valid = 0;
ahc->tqinfifonext = 1;
ahc_outb(ahc, KERNEL_TQINPOS, ahc->tqinfifonext - 1);
ahc_outb(ahc, TQINPOS, ahc->tqinfifonext);
ahc->qoutfifo = (uint8_t *)&ahc->targetcmds[256];
}
ahc->qinfifo = &ahc->qoutfifo[256];
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
ahc->init_level++;
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
/* Allocate SCB data now that buffer_dmat is initialized */
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
if (ahc->scb_data->maxhscbs == 0)
if (ahcinitscbdata(ahc) != 0)
return (ENOMEM);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
/*
* Allocate a tstate to house information for our
* initiator presence on the bus as well as the user
* data for any target mode initiator.
*/
if (ahc_alloc_tstate(ahc, ahc->our_id, 'A') == NULL) {
printf("%s: unable to allocate tmode_tstate. "
"Failing attach\n", ahc_name(ahc));
return (-1);
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if ((ahc->features & AHC_TWIN) != 0) {
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
if (ahc_alloc_tstate(ahc, ahc->our_id_b, 'B') == NULL) {
printf("%s: unable to allocate tmode_tstate. "
"Failing attach\n", ahc_name(ahc));
return (-1);
}
Bring back support for honoring the primary channel setting for twin channel EISA adapters. This could have caused the system to find the wrong root disk. Reviewed by: J Wunsch <j@ida.interface-business.de>
1998-10-15 18:21:50 +00:00
printf("Twin Channel, A SCSI Id=%d, B SCSI Id=%d, primary %c, ",
ahc->our_id, ahc->our_id_b,
ahc->flags & AHC_CHANNEL_B_PRIMARY? 'B': 'A');
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
} else {
if ((ahc->features & AHC_WIDE) != 0) {
printf("Wide ");
} else {
printf("Single ");
}
printf("Channel %c, SCSI Id=%d, ", ahc->channel, ahc->our_id);
}
ahc_outb(ahc, SEQ_FLAGS, 0);
if (ahc->scb_data->maxhscbs < AHC_SCB_MAX) {
ahc->flags |= AHC_PAGESCBS;
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
printf("%d/%d SCBs\n", ahc->scb_data->maxhscbs, AHC_SCB_MAX);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
} else {
ahc->flags &= ~AHC_PAGESCBS;
printf("%d SCBs\n", ahc->scb_data->maxhscbs);
}
#ifdef AHC_DEBUG
if (ahc_debug & AHC_SHOWMISC) {
printf("%s: hardware scb %d bytes; kernel scb %d bytes; "
"ahc_dma %d bytes\n",
ahc_name(ahc),
sizeof(struct hardware_scb),
sizeof(struct scb),
sizeof(struct ahc_dma_seg));
}
#endif /* AHC_DEBUG */
/* Set the SCSI Id, SXFRCTL0, SXFRCTL1, and SIMODE1, for both channels*/
if (ahc->features & AHC_TWIN) {
/*
* The device is gated to channel B after a chip reset,
* so set those values first
*/
term = (ahc->flags & AHC_TERM_ENB_B) != 0 ? STPWEN : 0;
if ((ahc->features & AHC_ULTRA2) != 0)
ahc_outb(ahc, SCSIID_ULTRA2, ahc->our_id_b);
else
ahc_outb(ahc, SCSIID, ahc->our_id_b);
scsi_conf = ahc_inb(ahc, SCSICONF + 1);
ahc_outb(ahc, SXFRCTL1, (scsi_conf & (ENSPCHK|STIMESEL))
Remove unused variables and functions. In the rejected message handler, it was possible for us to confuse a rejected SDTR or WDTR for a rejected tag.
1998-09-20 23:30:14 +00:00
|term|ENSTIMER|ACTNEGEN);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_outb(ahc, SIMODE1, ENSELTIMO|ENSCSIRST|ENSCSIPERR);
ahc_outb(ahc, SXFRCTL0, DFON|SPIOEN);
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
if ((scsi_conf & RESET_SCSI) != 0
&& (ahc->flags & AHC_INITIATORMODE) != 0)
ahc->flags |= AHC_RESET_BUS_B;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/* Select Channel A */
ahc_outb(ahc, SBLKCTL, ahc_inb(ahc, SBLKCTL) & ~SELBUSB);
}
term = (ahc->flags & AHC_TERM_ENB_A) != 0 ? STPWEN : 0;
if ((ahc->features & AHC_ULTRA2) != 0)
ahc_outb(ahc, SCSIID_ULTRA2, ahc->our_id);
else
ahc_outb(ahc, SCSIID, ahc->our_id);
scsi_conf = ahc_inb(ahc, SCSICONF);
ahc_outb(ahc, SXFRCTL1, (scsi_conf & (ENSPCHK|STIMESEL))
|term
|ENSTIMER|ACTNEGEN);
ahc_outb(ahc, SIMODE1, ENSELTIMO|ENSCSIRST|ENSCSIPERR);
ahc_outb(ahc, SXFRCTL0, DFON|SPIOEN);
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
if ((scsi_conf & RESET_SCSI) != 0
&& (ahc->flags & AHC_INITIATORMODE) != 0)
ahc->flags |= AHC_RESET_BUS_A;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Look at the information that board initialization or
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
* the board bios has left us.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
ultraenb = 0;
tagenable = ALL_TARGETS_MASK;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/* Grab the disconnection disable table and invert it for our needs */
if (ahc->flags & AHC_USEDEFAULTS) {
printf("%s: Host Adapter Bios disabled. Using default SCSI "
"device parameters\n", ahc_name(ahc));
When 'using defaults' ensure that termination is enabled. Removed some unused code.
1998-10-09 17:41:39 +00:00
ahc->flags |= AHC_EXTENDED_TRANS_A|AHC_EXTENDED_TRANS_B|
AHC_TERM_ENB_A|AHC_TERM_ENB_B;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
discenable = ALL_TARGETS_MASK;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if ((ahc->features & AHC_ULTRA) != 0)
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
ultraenb = ALL_TARGETS_MASK;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
} else {
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
discenable = ~((ahc_inb(ahc, DISC_DSB + 1) << 8)
| ahc_inb(ahc, DISC_DSB));
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if ((ahc->features & (AHC_ULTRA|AHC_ULTRA2)) != 0)
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
ultraenb = (ahc_inb(ahc, ULTRA_ENB + 1) << 8)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
| ahc_inb(ahc, ULTRA_ENB);
}
if ((ahc->features & (AHC_WIDE|AHC_TWIN)) == 0)
max_targ = 7;
for (i = 0; i <= max_targ; i++) {
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
struct ahc_initiator_tinfo *tinfo;
struct tmode_tstate *tstate;
u_int our_id;
u_int target_id;
char channel;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
channel = 'A';
our_id = ahc->our_id;
target_id = i;
if (i > 7 && (ahc->features & AHC_TWIN) != 0) {
channel = 'B';
our_id = ahc->our_id_b;
target_id = i % 8;
}
tinfo = ahc_fetch_transinfo(ahc, channel, our_id,
target_id, &tstate);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/* Default to async narrow across the board */
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
bzero(tinfo, sizeof(*tinfo));
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (ahc->flags & AHC_USEDEFAULTS) {
if ((ahc->features & AHC_WIDE) != 0)
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
tinfo->user.width = MSG_EXT_WDTR_BUS_16_BIT;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* These will be truncated when we determine the
* connection type we have with the target.
*/
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
tinfo->user.period = ahc_syncrates->period;
tinfo->user.offset = ~0;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
} else {
u_int scsirate;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
uint16_t mask;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/* Take the settings leftover in scratch RAM. */
scsirate = ahc_inb(ahc, TARG_SCSIRATE + i);
mask = (0x01 << i);
if ((ahc->features & AHC_ULTRA2) != 0) {
u_int offset;
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
u_int maxsync;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if ((scsirate & SOFS) == 0x0F) {
/*
* Haven't negotiated yet,
* so the format is different.
*/
scsirate = (scsirate & SXFR) >> 4
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
| (ultraenb & mask)
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
? 0x08 : 0x0
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
| (scsirate & WIDEXFER);
offset = MAX_OFFSET_ULTRA2;
} else
offset = ahc_inb(ahc, TARG_OFFSET + i);
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
maxsync = AHC_SYNCRATE_ULTRA2;
if ((ahc->features & AHC_DT) != 0)
maxsync = AHC_SYNCRATE_DT;
aic7xxx.c: Honor the 'bus reset at startup' option now that the XPT properly handles transfer negotiation in this scenario. Honor the sync rate settings on Ultra2 controllers. We would always negotiate at the fastest speed. Oops. aic7xxx.h: Whitespace. aic7xxx.seq: Fix a minor nit that would cause the controller to miss the update of the negotiation required bitmask causing the negotiation to be delayed by a command.
1999-05-22 22:04:11 +00:00
tinfo->user.period =
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
ahc_find_period(ahc, scsirate, maxsync);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (offset == 0)
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
tinfo->user.period = 0;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
else
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
tinfo->user.offset = ~0;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if ((scsirate & SXFR_ULTRA2) <= 8/*10MHz*/
&& (ahc->features & AHC_DT) != 0)
tinfo->user.ppr_options =
MSG_EXT_PPR_DT_REQ;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
} else if ((scsirate & SOFS) != 0) {
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
tinfo->user.period =
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_find_period(ahc, scsirate,
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
(ultraenb & mask)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
? AHC_SYNCRATE_ULTRA
: AHC_SYNCRATE_FAST);
Fix from Justin for transfer negotiations for targets up to target ID 7.
1999-04-26 22:03:44 +00:00
if (tinfo->user.period != 0)
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
tinfo->user.offset = ~0;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
if ((scsirate & WIDEXFER) != 0
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
&& (ahc->features & AHC_WIDE) != 0)
tinfo->user.width = MSG_EXT_WDTR_BUS_16_BIT;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
tinfo->user.protocol_version = 4;
if ((ahc->features & AHC_DT) != 0)
tinfo->user.transport_version = 3;
else
tinfo->user.transport_version = 2;
tinfo->goal.protocol_version = 2;
tinfo->goal.transport_version = 2;
tinfo->current.protocol_version = 2;
tinfo->current.transport_version = 2;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
tstate->ultraenb = ultraenb;
tstate->discenable = discenable;
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
tstate->tagenable = 0; /* Wait until the XPT says its okay */
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
ahc->user_discenable = discenable;
ahc->user_tagenable = tagenable;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
aic7xxx.c: Correct the BUILD_TCL macro. It was placing the target id in the wrong bits. This was only an issue for adapters that do not perform SCB paging (aha-3940AUW for instance). Don't bother inlining ahc_index_busy_tcl. It is never used in a performance critical path and is a bit chunky. Correct ahc_index_busy_tcl to deal with "busy target tables" embedded in the latter half of 64byte SCBs. Don't initialize the busy target table to its empty state until after we have finished extracting configuration information from chip SRAM. In the common case of using 16 bytes of chip SRAM to do untagged target lookups, we were trashing the last 8 targets configuration data. (actually only target 8 because of the bug in the BUILD_TCL macro). Cram the "bus reset delivered" message back under bootverbose. Fix the cleanup of the SCB busy target table when aborting commands. If the lun is wildcarded, we must loop through all possible luns. aic7xxx.h: Only bother supporting 64 luns right now. It doesn't seem like either this driver or any peripherals will be doing information unit transfers (where the lun number is a 32 bit integer) any time soon. aic7xxx.seq: Fix support for the aic7895. We must flush the data FIFO if performing a manual transfer that is not a multiple of 8 bytes. We were doing this quite regularly for embedded cdbs. Manaually flush the fifo on earlier adapters when dealing with embedded cdbs too. We were stuffing the FIFO with 16 bytes instead, but triggering the flush is more efficient and allows us to remove two instructions from the "copy_to_fifo" routine.
2000-07-24 22:27:40 +00:00
/* There are no untagged SCBs active yet. */
for (i = 0; i < 16; i++) {
ahc_index_busy_tcl(ahc, BUILD_TCL(i << 4, 0), /*unbusy*/TRUE);
if ((ahc->features & AHC_SCB_BTT) != 0) {
int lun;
/*
* The SCB based BTT allows an entry per
* target and lun pair.
*/
for (lun = 1; lun < AHC_NUM_LUNS; lun++) {
ahc_index_busy_tcl(ahc,
BUILD_TCL(i << 4, lun),
/*unbusy*/TRUE);
}
}
}
/* All of our queues are empty */
for (i = 0; i < 256; i++)
ahc->qoutfifo[i] = SCB_LIST_NULL;
for (i = 0; i < 256; i++)
ahc->qinfifo[i] = SCB_LIST_NULL;
if ((ahc->features & AHC_MULTI_TID) != 0) {
ahc_outb(ahc, TARGID, 0);
ahc_outb(ahc, TARGID + 1, 0);
}
Fix from Justin for transfer negotiations for targets up to target ID 7.
1999-04-26 22:03:44 +00:00
/*
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
* Tell the sequencer where it can find our arrays in memory.
Fix from Justin for transfer negotiations for targets up to target ID 7.
1999-04-26 22:03:44 +00:00
*/
physaddr = ahc->scb_data->hscb_busaddr;
ahc_outb(ahc, HSCB_ADDR, physaddr & 0xFF);
ahc_outb(ahc, HSCB_ADDR + 1, (physaddr >> 8) & 0xFF);
ahc_outb(ahc, HSCB_ADDR + 2, (physaddr >> 16) & 0xFF);
ahc_outb(ahc, HSCB_ADDR + 3, (physaddr >> 24) & 0xFF);
physaddr = ahc->shared_data_busaddr;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahc_outb(ahc, SHARED_DATA_ADDR, physaddr & 0xFF);
ahc_outb(ahc, SHARED_DATA_ADDR + 1, (physaddr >> 8) & 0xFF);
ahc_outb(ahc, SHARED_DATA_ADDR + 2, (physaddr >> 16) & 0xFF);
ahc_outb(ahc, SHARED_DATA_ADDR + 3, (physaddr >> 24) & 0xFF);
Fix from Justin for transfer negotiations for targets up to target ID 7.
1999-04-26 22:03:44 +00:00
/*
* Initialize the group code to command length table.
* This overrides the values in TARG_SCSIRATE, so only
* setup the table after we have processed that information.
*/
ahc_outb(ahc, CMDSIZE_TABLE, 5);
ahc_outb(ahc, CMDSIZE_TABLE + 1, 9);
ahc_outb(ahc, CMDSIZE_TABLE + 2, 9);
ahc_outb(ahc, CMDSIZE_TABLE + 3, 0);
ahc_outb(ahc, CMDSIZE_TABLE + 4, 15);
ahc_outb(ahc, CMDSIZE_TABLE + 5, 11);
ahc_outb(ahc, CMDSIZE_TABLE + 6, 0);
ahc_outb(ahc, CMDSIZE_TABLE + 7, 0);
/* Tell the sequencer of our initial queue positions */
ahc_outb(ahc, KERNEL_QINPOS, 0);
ahc_outb(ahc, QINPOS, 0);
ahc_outb(ahc, QOUTPOS, 0);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
/* Don't have any special messages to send to targets */
ahc_outb(ahc, TARGET_MSG_REQUEST, 0);
ahc_outb(ahc, TARGET_MSG_REQUEST + 1, 0);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Use the built in queue management registers
* if they are available.
*/
if ((ahc->features & AHC_QUEUE_REGS) != 0) {
ahc_outb(ahc, QOFF_CTLSTA, SCB_QSIZE_256);
ahc_outb(ahc, SDSCB_QOFF, 0);
ahc_outb(ahc, SNSCB_QOFF, 0);
ahc_outb(ahc, HNSCB_QOFF, 0);
}
/* We don't have any waiting selections */
ahc_outb(ahc, WAITING_SCBH, SCB_LIST_NULL);
/* Our disconnection list is empty too */
ahc_outb(ahc, DISCONNECTED_SCBH, SCB_LIST_NULL);
/* Message out buffer starts empty */
ahc_outb(ahc, MSG_OUT, MSG_NOOP);
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
/*
* Setup the allowed SCSI Sequences based on operational mode.
* If we are a target, we'll enalbe select in operations once
* we've had a lun enabled.
*/
scsiseq_template = ENSELO|ENAUTOATNO|ENAUTOATNP;
if ((ahc->flags & AHC_INITIATORMODE) != 0)
scsiseq_template |= ENRSELI;
ahc_outb(ahc, SCSISEQ_TEMPLATE, scsiseq_template);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Load the Sequencer program and Enable the adapter
* in "fast" mode.
*/
if (bootverbose)
printf("%s: Downloading Sequencer Program...",
ahc_name(ahc));
ahc_loadseq(ahc);
/* We have to wait until after any system dumps... */
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
EVENTHANDLER_REGISTER(shutdown_final, ahc_shutdown,
ahc, SHUTDOWN_PRI_DEFAULT);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if ((ahc->features & AHC_ULTRA2) != 0) {
int wait;
/*
* Wait for up to 500ms for our transceivers
* to settle. If the adapter does not have
* a cable attached, the tranceivers may
* never settle, so don't complain if we
* fail here.
*/
pause_sequencer(ahc);
for (wait = 5000;
(ahc_inb(ahc, SBLKCTL) & (ENAB40|ENAB20)) == 0 && wait;
wait--)
DELAY(100);
unpause_sequencer(ahc);
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
return (0);
}
static cam_status
ahc_find_tmode_devs(struct ahc_softc *ahc, struct cam_sim *sim, union ccb *ccb,
struct tmode_tstate **tstate, struct tmode_lstate **lstate,
int notfound_failure)
{
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int our_id;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* If we are not configured for target mode, someone
* is really confused to be sending this to us.
*/
if ((ahc->flags & AHC_TARGETMODE) == 0)
return (CAM_REQ_INVALID);
/* Range check target and lun */
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
/*
* Handle the 'black hole' device that sucks up
* requests to unattached luns on enabled targets.
*/
if (ccb->ccb_h.target_id == CAM_TARGET_WILDCARD
&& ccb->ccb_h.target_lun == CAM_LUN_WILDCARD) {
*tstate = NULL;
*lstate = ahc->black_hole;
} else {
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
u_int max_id;
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
if (cam_sim_bus(sim) == 0)
our_id = ahc->our_id;
else
our_id = ahc->our_id_b;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
max_id = (ahc->features & AHC_WIDE) ? 15 : 7;
if (ccb->ccb_h.target_id > max_id)
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
return (CAM_TID_INVALID);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
if (ccb->ccb_h.target_lun > 7)
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
return (CAM_LUN_INVALID);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
if (ccb->ccb_h.target_id != our_id) {
if ((ahc->features & AHC_MULTI_TID) != 0) {
/*
* Only allow additional targets if
* the initiator role is disabled.
* The hardware cannot handle a re-select-in
* on the initiator id during a re-select-out
* on a different target id.
*/
if ((ahc->flags & AHC_INITIATORMODE) != 0)
return (CAM_TID_INVALID);
} else {
/*
* Only allow our target id to change
* if the initiator role is not configured
* and there are no enabled luns which
* are attached to the currently registered
* scsi id.
*/
if ((ahc->flags & AHC_INITIATORMODE) != 0
|| ahc->enabled_luns > 0)
return (CAM_TID_INVALID);
}
}
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
*tstate = ahc->enabled_targets[ccb->ccb_h.target_id];
*lstate = NULL;
if (*tstate != NULL)
*lstate =
(*tstate)->enabled_luns[ccb->ccb_h.target_lun];
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (notfound_failure != 0 && *lstate == NULL)
return (CAM_PATH_INVALID);
return (CAM_REQ_CMP);
}
static void
ahc_action(struct cam_sim *sim, union ccb *ccb)
{
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
struct ahc_softc *ahc;
struct tmode_lstate *lstate;
u_int target_id;
u_int our_id;
int s;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
CAM_DEBUG(ccb->ccb_h.path, CAM_DEBUG_TRACE, ("ahc_action\n"));
ahc = (struct ahc_softc *)cam_sim_softc(sim);
target_id = ccb->ccb_h.target_id;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
our_id = SIM_SCSI_ID(ahc, sim);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
switch (ccb->ccb_h.func_code) {
/* Common cases first */
case XPT_ACCEPT_TARGET_IO: /* Accept Host Target Mode CDB */
case XPT_CONT_TARGET_IO:/* Continue Host Target I/O Connection*/
{
struct tmode_tstate *tstate;
cam_status status;
status = ahc_find_tmode_devs(ahc, sim, ccb, &tstate,
&lstate, TRUE);
if (status != CAM_REQ_CMP) {
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
if (ccb->ccb_h.func_code == XPT_CONT_TARGET_IO) {
/* Response from the black hole device */
tstate = NULL;
lstate = ahc->black_hole;
} else {
ccb->ccb_h.status = status;
xpt_done(ccb);
break;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
if (ccb->ccb_h.func_code == XPT_ACCEPT_TARGET_IO) {
Add necessary splcam() protection around target mode setup and teardown routines as well as card access that occurs for the AC_LOST_DEVICE async event.
1999-03-08 22:42:52 +00:00
int s;
s = splcam();
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
SLIST_INSERT_HEAD(&lstate->accept_tios, &ccb->ccb_h,
sim_links.sle);
ccb->ccb_h.status = CAM_REQ_INPROG;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
if ((ahc->flags & AHC_TQINFIFO_BLOCKED) != 0)
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
ahc_run_tqinfifo(ahc, /*paused*/FALSE);
Add necessary splcam() protection around target mode setup and teardown routines as well as card access that occurs for the AC_LOST_DEVICE async event.
1999-03-08 22:42:52 +00:00
splx(s);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
break;
}
/*
* The target_id represents the target we attempt to
* select. In target mode, this is the initiator of
* the original command.
*/
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
our_id = target_id;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
target_id = ccb->csio.init_id;
/* FALLTHROUGH */
}
case XPT_SCSI_IO: /* Execute the requested I/O operation */
case XPT_RESET_DEV: /* Bus Device Reset the specified SCSI device */
{
struct scb *scb;
struct hardware_scb *hscb;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
struct ahc_initiator_tinfo *tinfo;
struct tmode_tstate *tstate;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
uint16_t mask;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* get an scb to use.
*/
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
if ((scb = ahcgetscb(ahc)) == NULL) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
int s;
s = splcam();
ahc->flags |= AHC_RESOURCE_SHORTAGE;
splx(s);
xpt_freeze_simq(ahc->sim, /*count*/1);
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
ahcsetccbstatus(ccb, CAM_REQUEUE_REQ);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
xpt_done(ccb);
return;
}
hscb = scb->hscb;
CAM_DEBUG(ccb->ccb_h.path, CAM_DEBUG_SUBTRACE,
("start scb(%p)\n", scb));
scb->ccb = ccb;
/*
* So we can find the SCB when an abort is requested
*/
ccb->ccb_h.ccb_scb_ptr = scb;
ccb->ccb_h.ccb_ahc_ptr = ahc;
/*
* Put all the arguments for the xfer in the scb
*/
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
hscb->control = 0;
hscb->scsiid = BUILD_SCSIID(ahc, sim, target_id, our_id);
hscb->lun = ccb->ccb_h.target_lun;
mask = SCB_GET_TARGET_MASK(ahc, scb);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
tinfo = ahc_fetch_transinfo(ahc, SIM_CHANNEL(ahc, sim), our_id,
target_id, &tstate);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
hscb->scsirate = tinfo->scsirate;
hscb->scsioffset = tinfo->current.offset;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
if ((tstate->ultraenb & mask) != 0)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
hscb->control |= ULTRAENB;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
if ((tstate->discenable & mask) != 0
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
&& (ccb->ccb_h.flags & CAM_DIS_DISCONNECT) == 0)
hscb->control |= DISCENB;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if ((ccb->ccb_h.flags & CAM_NEGOTIATE) != 0
&& (tinfo->current.width != 0 || tinfo->current.period != 0)) {
scb->flags |= SCB_NEGOTIATE;
hscb->control |= MK_MESSAGE;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (ccb->ccb_h.func_code == XPT_RESET_DEV) {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
hscb->cdb_len = 0;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
scb->flags |= SCB_DEVICE_RESET;
hscb->control |= MK_MESSAGE;
ahc_execute_scb(scb, NULL, 0, 0);
} else {
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
if (ccb->ccb_h.func_code == XPT_CONT_TARGET_IO) {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
struct target_data *tdata;
tdata = &hscb->shared_data.tdata;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (ahc->pending_device == lstate) {
scb->flags |= SCB_TARGET_IMMEDIATE;
ahc->pending_device = NULL;
}
hscb->control |= TARGET_SCB;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
tdata->target_phases = IDENTIFY_SEEN;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if ((ccb->ccb_h.flags & CAM_SEND_STATUS) != 0) {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
tdata->target_phases |= SPHASE_PENDING;
tdata->scsi_status =
ccb->csio.scsi_status;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
tdata->initiator_tag = ccb->csio.tag_id;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
if (ccb->ccb_h.flags & CAM_TAG_ACTION_VALID)
hscb->control |= ccb->csio.tag_action;
ahc_setup_data(ahc, &ccb->csio, scb);
}
break;
}
case XPT_NOTIFY_ACK:
case XPT_IMMED_NOTIFY:
{
struct tmode_tstate *tstate;
struct tmode_lstate *lstate;
cam_status status;
status = ahc_find_tmode_devs(ahc, sim, ccb, &tstate,
&lstate, TRUE);
if (status != CAM_REQ_CMP) {
ccb->ccb_h.status = status;
xpt_done(ccb);
break;
}
SLIST_INSERT_HEAD(&lstate->immed_notifies, &ccb->ccb_h,
sim_links.sle);
ccb->ccb_h.status = CAM_REQ_INPROG;
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
ahc_send_lstate_events(ahc, lstate);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
break;
}
case XPT_EN_LUN: /* Enable LUN as a target */
Add support for routing initiator transactions to disabled luns to the black hole device. The controller will now only accept selections if the black hole device is present and some other target/lun is enabled for target mode. Handle the IGNORE WIDE RESIDUE message. This support has not been tested. Checkpoint work on handling ABORT, BUS DEVICE RESET, TERMINATE I/O PROCESS, and CLEAR QUEUE messages as a target. Fix a few problems with tagged command handling in target mode. Wait until the sync offset counter falls to 0 before changing phase after a data-in transfer completes as the DMA logic seems to indicate transfer complete as soon as our last REQ is issued. Simplify some of the target mode message handling code in the sequencer.
1999-01-14 06:14:15 +00:00
ahc_handle_en_lun(ahc, sim, ccb);
xpt_done(ccb);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
break;
case XPT_ABORT: /* Abort the specified CCB */
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
{
ahc_abort_ccb(ahc, sim, ccb);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
break;
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
case XPT_SET_TRAN_SETTINGS:
{
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
struct ahc_devinfo devinfo;
struct ccb_trans_settings *cts;
struct ahc_initiator_tinfo *tinfo;
struct tmode_tstate *tstate;
uint16_t *discenable;
uint16_t *tagenable;
u_int update_type;
int s;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
cts = &ccb->cts;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
ahc_compile_devinfo(&devinfo, SIM_SCSI_ID(ahc, sim),
cts->ccb_h.target_id,
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
cts->ccb_h.target_lun,
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
SIM_CHANNEL(ahc, sim),
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ROLE_UNKNOWN);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
tinfo = ahc_fetch_transinfo(ahc, devinfo.channel,
devinfo.our_scsiid,
devinfo.target, &tstate);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
update_type = 0;
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
if ((cts->flags & CCB_TRANS_CURRENT_SETTINGS) != 0) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
update_type |= AHC_TRANS_GOAL;
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
discenable = &tstate->discenable;
tagenable = &tstate->tagenable;
} else if ((cts->flags & CCB_TRANS_USER_SETTINGS) != 0) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
update_type |= AHC_TRANS_USER;
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
discenable = &ahc->user_discenable;
tagenable = &ahc->user_tagenable;
} else {
ccb->ccb_h.status = CAM_REQ_INVALID;
xpt_done(ccb);
break;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
s = splcam();
if ((cts->valid & CCB_TRANS_DISC_VALID) != 0) {
if ((cts->flags & CCB_TRANS_DISC_ENB) != 0)
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
*discenable |= devinfo.target_mask;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
else
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
*discenable &= ~devinfo.target_mask;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
if ((cts->valid & CCB_TRANS_TQ_VALID) != 0) {
if ((cts->flags & CCB_TRANS_TAG_ENB) != 0)
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
*tagenable |= devinfo.target_mask;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
else
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
*tagenable &= ~devinfo.target_mask;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
if ((cts->valid & CCB_TRANS_BUS_WIDTH_VALID) != 0) {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahc_validate_width(ahc, &cts->bus_width);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc_set_width(ahc, &devinfo, cts->ccb_h.path,
Correct a bug where the chip could be unpaused in the middle of a bus or device reset error recovery operation.
1999-05-18 03:58:49 +00:00
cts->bus_width, update_type,
/*paused*/FALSE);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if ((cts->valid & CCB_TRANS_SYNC_OFFSET_VALID) == 0) {
if (update_type == AHC_TRANS_USER)
cts->sync_offset = tinfo->user.offset;
else
cts->sync_offset = tinfo->goal.offset;
}
if ((cts->valid & CCB_TRANS_SYNC_RATE_VALID) == 0) {
if (update_type == AHC_TRANS_USER)
cts->sync_period = tinfo->user.period;
else
cts->sync_period = tinfo->goal.period;
}
Add a number of interrelated CAM feature enhancements and bug fixes. NOTE: These changes will require recompilation of any userland applications, like cdrecord, xmcd, etc., that use the CAM passthrough interface. A make world is recommended. camcontrol.[c8]: - We now support two new commands, "tags" and "negotiate". - The tags commands allows users to view the number of tagged openings for a device as well as a number of other related parameters, and it allows users to set tagged openings for a device. - The negotiate command allows users to enable and disable disconnection and tagged queueing, set sync rates, offsets and bus width. Note that not all of those features are available for all controllers. Only the adv, ahc, and ncr drivers fully support all of the features at this point. Some cards do not allow the setting of sync rates, offsets and the like, and some of the drivers don't have any facilities to do so. Some drivers, like the adw driver, only support enabling or disabling sync negotiation, but do not support setting sync rates. - new description in the camcontrol man page of how to format a disk - cleanup of the camcontrol inquiry command - add support in the 'devlist' command for skipping unconfigured devices if -v was not specified on the command line. - make use of the new base_transfer_speed in the path inquiry CCB. - fix CCB bzero cases cam_xpt.c, cam_sim.[ch], cam_ccb.h: - new flags on many CCB function codes to designate whether they're non-immediate, use a user-supplied CCB, and can only be passed from userland programs via the xpt device. Use these flags in the transport layer and pass driver to categorize CCBs. - new flag in the transport layer device matching code for device nodes that indicates whether a device is unconfigured - bump the CAM version from 0x10 to 0x11 - Change the CAM ioctls to use the version as their group code, so we can force users to recompile code even when the CCB size doesn't change. - add + fill in a new value in the path inquiry CCB, base_transfer_speed. Remove a corresponding field from the cam_sim structure, and add code to every SIM to set this field to the proper value. - Fix the set transfer settings code in the transport layer. scsi_cd.c: - make some variables volatile instead of just casting them in various places - fix a race condition in the changer code - attach unless we get a "logical unit not supported" error. This should fix all of the cases where people have devices that return weird errors when they don't have media in the drive. scsi_da.c: - attach unless we get a "logical unit not supported" error scsi_pass.c: - for immediate CCBs, just malloc a CCB to send the user request in. This gets rid of the 'held' count problem in camcontrol tags. scsi_pass.h: - change the CAM ioctls to use the CAM version as their group code. adv driver: - Allow changing the sync rate and offset separately. adw driver - Allow changing the sync rate and offset separately. aha driver: - Don't return CAM_REQ_CMP for SET_TRAN_SETTINGS CCBs. ahc driver: - Allow setting offset and sync rate separately bt driver: - Don't return CAM_REQ_CMP for SET_TRAN_SETTINGS CCBs. NCR driver: - Fix the ultra/ultra 2 negotiation bug - allow setting both the sync rate and offset separately Other HBA drivers: - Put code in to set the base_transfer_speed field for XPT_GET_TRAN_SETTINGS CCBs. Reviewed by: gibbs, mjacob (isp), imp (aha)
1999-05-06 20:16:39 +00:00
if (((cts->valid & CCB_TRANS_SYNC_RATE_VALID) != 0)
|| ((cts->valid & CCB_TRANS_SYNC_OFFSET_VALID) != 0)) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
struct ahc_syncrate *syncrate;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int ppr_options;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
u_int maxsync;
if ((ahc->features & AHC_ULTRA2) != 0)
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
maxsync = AHC_SYNCRATE_DT;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
else if ((ahc->features & AHC_ULTRA) != 0)
maxsync = AHC_SYNCRATE_ULTRA;
else
maxsync = AHC_SYNCRATE_FAST;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ppr_options = 0;
if (cts->sync_period <= 9)
ppr_options = MSG_EXT_PPR_DT_REQ;
Add a number of interrelated CAM feature enhancements and bug fixes. NOTE: These changes will require recompilation of any userland applications, like cdrecord, xmcd, etc., that use the CAM passthrough interface. A make world is recommended. camcontrol.[c8]: - We now support two new commands, "tags" and "negotiate". - The tags commands allows users to view the number of tagged openings for a device as well as a number of other related parameters, and it allows users to set tagged openings for a device. - The negotiate command allows users to enable and disable disconnection and tagged queueing, set sync rates, offsets and bus width. Note that not all of those features are available for all controllers. Only the adv, ahc, and ncr drivers fully support all of the features at this point. Some cards do not allow the setting of sync rates, offsets and the like, and some of the drivers don't have any facilities to do so. Some drivers, like the adw driver, only support enabling or disabling sync negotiation, but do not support setting sync rates. - new description in the camcontrol man page of how to format a disk - cleanup of the camcontrol inquiry command - add support in the 'devlist' command for skipping unconfigured devices if -v was not specified on the command line. - make use of the new base_transfer_speed in the path inquiry CCB. - fix CCB bzero cases cam_xpt.c, cam_sim.[ch], cam_ccb.h: - new flags on many CCB function codes to designate whether they're non-immediate, use a user-supplied CCB, and can only be passed from userland programs via the xpt device. Use these flags in the transport layer and pass driver to categorize CCBs. - new flag in the transport layer device matching code for device nodes that indicates whether a device is unconfigured - bump the CAM version from 0x10 to 0x11 - Change the CAM ioctls to use the version as their group code, so we can force users to recompile code even when the CCB size doesn't change. - add + fill in a new value in the path inquiry CCB, base_transfer_speed. Remove a corresponding field from the cam_sim structure, and add code to every SIM to set this field to the proper value. - Fix the set transfer settings code in the transport layer. scsi_cd.c: - make some variables volatile instead of just casting them in various places - fix a race condition in the changer code - attach unless we get a "logical unit not supported" error. This should fix all of the cases where people have devices that return weird errors when they don't have media in the drive. scsi_da.c: - attach unless we get a "logical unit not supported" error scsi_pass.c: - for immediate CCBs, just malloc a CCB to send the user request in. This gets rid of the 'held' count problem in camcontrol tags. scsi_pass.h: - change the CAM ioctls to use the CAM version as their group code. adv driver: - Allow changing the sync rate and offset separately. adw driver - Allow changing the sync rate and offset separately. aha driver: - Don't return CAM_REQ_CMP for SET_TRAN_SETTINGS CCBs. ahc driver: - Allow setting offset and sync rate separately bt driver: - Don't return CAM_REQ_CMP for SET_TRAN_SETTINGS CCBs. NCR driver: - Fix the ultra/ultra 2 negotiation bug - allow setting both the sync rate and offset separately Other HBA drivers: - Put code in to set the base_transfer_speed field for XPT_GET_TRAN_SETTINGS CCBs. Reviewed by: gibbs, mjacob (isp), imp (aha)
1999-05-06 20:16:39 +00:00
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
syncrate = ahc_find_syncrate(ahc, &cts->sync_period,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
&ppr_options,
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
maxsync);
ahc_validate_offset(ahc, syncrate, &cts->sync_offset,
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
MSG_EXT_WDTR_BUS_8_BIT);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/* We use a period of 0 to represent async */
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (cts->sync_offset == 0) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
cts->sync_period = 0;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ppr_options = 0;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (ppr_options == MSG_EXT_PPR_DT_REQ
&& tinfo->user.transport_version >= 3) {
tinfo->goal.transport_version =
tinfo->user.transport_version;
tinfo->current.transport_version =
tinfo->user.transport_version;
}
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc_set_syncrate(ahc, &devinfo, cts->ccb_h.path,
syncrate, cts->sync_period,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
cts->sync_offset, ppr_options,
update_type, /*paused*/FALSE);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
splx(s);
ccb->ccb_h.status = CAM_REQ_CMP;
xpt_done(ccb);
break;
}
case XPT_GET_TRAN_SETTINGS:
/* Get default/user set transfer settings for the target */
{
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
struct ahc_devinfo devinfo;
struct ccb_trans_settings *cts;
struct ahc_initiator_tinfo *targ_info;
struct tmode_tstate *tstate;
struct ahc_transinfo *tinfo;
int s;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
cts = &ccb->cts;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
ahc_compile_devinfo(&devinfo, SIM_SCSI_ID(ahc, sim),
cts->ccb_h.target_id,
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
cts->ccb_h.target_lun,
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
SIM_CHANNEL(ahc, sim),
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ROLE_UNKNOWN);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
targ_info = ahc_fetch_transinfo(ahc, devinfo.channel,
devinfo.our_scsiid,
devinfo.target, &tstate);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if ((cts->flags & CCB_TRANS_CURRENT_SETTINGS) != 0)
tinfo = &targ_info->current;
else
tinfo = &targ_info->user;
s = splcam();
cts->flags &= ~(CCB_TRANS_DISC_ENB|CCB_TRANS_TAG_ENB);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if ((cts->flags & CCB_TRANS_CURRENT_SETTINGS) != 0) {
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
if ((ahc->user_discenable & devinfo.target_mask) != 0)
cts->flags |= CCB_TRANS_DISC_ENB;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
if ((ahc->user_tagenable & devinfo.target_mask) != 0)
cts->flags |= CCB_TRANS_TAG_ENB;
} else {
if ((tstate->discenable & devinfo.target_mask) != 0)
cts->flags |= CCB_TRANS_DISC_ENB;
if ((tstate->tagenable & devinfo.target_mask) != 0)
cts->flags |= CCB_TRANS_TAG_ENB;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
cts->sync_period = tinfo->period;
cts->sync_offset = tinfo->offset;
cts->bus_width = tinfo->width;
splx(s);
cts->valid = CCB_TRANS_SYNC_RATE_VALID
| CCB_TRANS_SYNC_OFFSET_VALID
| CCB_TRANS_BUS_WIDTH_VALID
| CCB_TRANS_DISC_VALID
| CCB_TRANS_TQ_VALID;
ccb->ccb_h.status = CAM_REQ_CMP;
xpt_done(ccb);
break;
}
case XPT_CALC_GEOMETRY:
{
struct ccb_calc_geometry *ccg;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
uint32_t size_mb;
uint32_t secs_per_cylinder;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
int extended;
ccg = &ccb->ccg;
size_mb = ccg->volume_size
/ ((1024L * 1024L) / ccg->block_size);
extended = SIM_IS_SCSIBUS_B(ahc, sim)
? ahc->flags & AHC_EXTENDED_TRANS_B
: ahc->flags & AHC_EXTENDED_TRANS_A;
if (size_mb > 1024 && extended) {
ccg->heads = 255;
ccg->secs_per_track = 63;
} else {
ccg->heads = 64;
ccg->secs_per_track = 32;
}
secs_per_cylinder = ccg->heads * ccg->secs_per_track;
ccg->cylinders = ccg->volume_size / secs_per_cylinder;
ccb->ccb_h.status = CAM_REQ_CMP;
xpt_done(ccb);
break;
}
case XPT_RESET_BUS: /* Reset the specified SCSI bus */
{
int found;
s = splcam();
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
found = ahc_reset_channel(ahc, SIM_CHANNEL(ahc, sim),
/*initiate reset*/TRUE);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
splx(s);
aic7xxx.c: Correct the BUILD_TCL macro. It was placing the target id in the wrong bits. This was only an issue for adapters that do not perform SCB paging (aha-3940AUW for instance). Don't bother inlining ahc_index_busy_tcl. It is never used in a performance critical path and is a bit chunky. Correct ahc_index_busy_tcl to deal with "busy target tables" embedded in the latter half of 64byte SCBs. Don't initialize the busy target table to its empty state until after we have finished extracting configuration information from chip SRAM. In the common case of using 16 bytes of chip SRAM to do untagged target lookups, we were trashing the last 8 targets configuration data. (actually only target 8 because of the bug in the BUILD_TCL macro). Cram the "bus reset delivered" message back under bootverbose. Fix the cleanup of the SCB busy target table when aborting commands. If the lun is wildcarded, we must loop through all possible luns. aic7xxx.h: Only bother supporting 64 luns right now. It doesn't seem like either this driver or any peripherals will be doing information unit transfers (where the lun number is a 32 bit integer) any time soon. aic7xxx.seq: Fix support for the aic7895. We must flush the data FIFO if performing a manual transfer that is not a multiple of 8 bytes. We were doing this quite regularly for embedded cdbs. Manaually flush the fifo on earlier adapters when dealing with embedded cdbs too. We were stuffing the FIFO with 16 bytes instead, but triggering the flush is more efficient and allows us to remove two instructions from the "copy_to_fifo" routine.
2000-07-24 22:27:40 +00:00
if (bootverbose) {
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
xpt_print_path(SIM_PATH(ahc, sim));
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
printf("SCSI bus reset delivered. "
"%d SCBs aborted.\n", found);
}
ccb->ccb_h.status = CAM_REQ_CMP;
xpt_done(ccb);
break;
}
case XPT_TERM_IO: /* Terminate the I/O process */
/* XXX Implement */
ccb->ccb_h.status = CAM_REQ_INVALID;
xpt_done(ccb);
break;
case XPT_PATH_INQ: /* Path routing inquiry */
{
struct ccb_pathinq *cpi = &ccb->cpi;
cpi->version_num = 1; /* XXX??? */
cpi->hba_inquiry = PI_SDTR_ABLE|PI_TAG_ABLE;
if ((ahc->features & AHC_WIDE) != 0)
cpi->hba_inquiry |= PI_WIDE_16;
if ((ahc->flags & AHC_TARGETMODE) != 0) {
cpi->target_sprt = PIT_PROCESSOR
| PIT_DISCONNECT
| PIT_TERM_IO;
} else {
cpi->target_sprt = 0;
}
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
cpi->hba_misc = (ahc->flags & AHC_INITIATORMODE)
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
? 0 : PIM_NOINITIATOR;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
cpi->hba_eng_cnt = 0;
cpi->max_target = (ahc->features & AHC_WIDE) ? 15 : 7;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
cpi->max_lun = 64;
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
if (SIM_IS_SCSIBUS_B(ahc, sim)) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
cpi->initiator_id = ahc->our_id_b;
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
if ((ahc->flags & AHC_RESET_BUS_B) == 0)
cpi->hba_misc |= PIM_NOBUSRESET;
} else {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
cpi->initiator_id = ahc->our_id;
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
if ((ahc->flags & AHC_RESET_BUS_A) == 0)
cpi->hba_misc |= PIM_NOBUSRESET;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
cpi->bus_id = cam_sim_bus(sim);
Add a number of interrelated CAM feature enhancements and bug fixes. NOTE: These changes will require recompilation of any userland applications, like cdrecord, xmcd, etc., that use the CAM passthrough interface. A make world is recommended. camcontrol.[c8]: - We now support two new commands, "tags" and "negotiate". - The tags commands allows users to view the number of tagged openings for a device as well as a number of other related parameters, and it allows users to set tagged openings for a device. - The negotiate command allows users to enable and disable disconnection and tagged queueing, set sync rates, offsets and bus width. Note that not all of those features are available for all controllers. Only the adv, ahc, and ncr drivers fully support all of the features at this point. Some cards do not allow the setting of sync rates, offsets and the like, and some of the drivers don't have any facilities to do so. Some drivers, like the adw driver, only support enabling or disabling sync negotiation, but do not support setting sync rates. - new description in the camcontrol man page of how to format a disk - cleanup of the camcontrol inquiry command - add support in the 'devlist' command for skipping unconfigured devices if -v was not specified on the command line. - make use of the new base_transfer_speed in the path inquiry CCB. - fix CCB bzero cases cam_xpt.c, cam_sim.[ch], cam_ccb.h: - new flags on many CCB function codes to designate whether they're non-immediate, use a user-supplied CCB, and can only be passed from userland programs via the xpt device. Use these flags in the transport layer and pass driver to categorize CCBs. - new flag in the transport layer device matching code for device nodes that indicates whether a device is unconfigured - bump the CAM version from 0x10 to 0x11 - Change the CAM ioctls to use the version as their group code, so we can force users to recompile code even when the CCB size doesn't change. - add + fill in a new value in the path inquiry CCB, base_transfer_speed. Remove a corresponding field from the cam_sim structure, and add code to every SIM to set this field to the proper value. - Fix the set transfer settings code in the transport layer. scsi_cd.c: - make some variables volatile instead of just casting them in various places - fix a race condition in the changer code - attach unless we get a "logical unit not supported" error. This should fix all of the cases where people have devices that return weird errors when they don't have media in the drive. scsi_da.c: - attach unless we get a "logical unit not supported" error scsi_pass.c: - for immediate CCBs, just malloc a CCB to send the user request in. This gets rid of the 'held' count problem in camcontrol tags. scsi_pass.h: - change the CAM ioctls to use the CAM version as their group code. adv driver: - Allow changing the sync rate and offset separately. adw driver - Allow changing the sync rate and offset separately. aha driver: - Don't return CAM_REQ_CMP for SET_TRAN_SETTINGS CCBs. ahc driver: - Allow setting offset and sync rate separately bt driver: - Don't return CAM_REQ_CMP for SET_TRAN_SETTINGS CCBs. NCR driver: - Fix the ultra/ultra 2 negotiation bug - allow setting both the sync rate and offset separately Other HBA drivers: - Put code in to set the base_transfer_speed field for XPT_GET_TRAN_SETTINGS CCBs. Reviewed by: gibbs, mjacob (isp), imp (aha)
1999-05-06 20:16:39 +00:00
cpi->base_transfer_speed = 3300;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
strncpy(cpi->sim_vid, "FreeBSD", SIM_IDLEN);
strncpy(cpi->hba_vid, "Adaptec", HBA_IDLEN);
strncpy(cpi->dev_name, cam_sim_name(sim), DEV_IDLEN);
cpi->unit_number = cam_sim_unit(sim);
cpi->ccb_h.status = CAM_REQ_CMP;
xpt_done(ccb);
break;
}
default:
ccb->ccb_h.status = CAM_REQ_INVALID;
xpt_done(ccb);
break;
}
}
static void
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahc_async(void *callback_arg, uint32_t code, struct cam_path *path, void *arg)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
struct ahc_softc *ahc;
struct cam_sim *sim;
sim = (struct cam_sim *)callback_arg;
ahc = (struct ahc_softc *)cam_sim_softc(sim);
switch (code) {
case AC_LOST_DEVICE:
{
struct ahc_devinfo devinfo;
Add necessary splcam() protection around target mode setup and teardown routines as well as card access that occurs for the AC_LOST_DEVICE async event.
1999-03-08 22:42:52 +00:00
int s;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
ahc_compile_devinfo(&devinfo, SIM_SCSI_ID(ahc, sim),
xpt_path_target_id(path),
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
xpt_path_lun_id(path),
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
SIM_CHANNEL(ahc, sim),
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ROLE_UNKNOWN);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Revert to async/narrow transfers
* for the next device.
*/
Add necessary splcam() protection around target mode setup and teardown routines as well as card access that occurs for the AC_LOST_DEVICE async event.
1999-03-08 22:42:52 +00:00
s = splcam();
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_set_width(ahc, &devinfo, path, MSG_EXT_WDTR_BUS_8_BIT,
Correct a bug where the chip could be unpaused in the middle of a bus or device reset error recovery operation.
1999-05-18 03:58:49 +00:00
AHC_TRANS_GOAL|AHC_TRANS_CUR,
/*paused*/FALSE);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_set_syncrate(ahc, &devinfo, path, /*syncrate*/NULL,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
/*period*/0, /*offset*/0, /*ppr_options*/0,
Correct a bug where the chip could be unpaused in the middle of a bus or device reset error recovery operation.
1999-05-18 03:58:49 +00:00
AHC_TRANS_GOAL|AHC_TRANS_CUR,
/*paused*/FALSE);
Add necessary splcam() protection around target mode setup and teardown routines as well as card access that occurs for the AC_LOST_DEVICE async event.
1999-03-08 22:42:52 +00:00
splx(s);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
break;
}
default:
break;
}
}
static void
ahc_execute_scb(void *arg, bus_dma_segment_t *dm_segs, int nsegments,
int error)
{
struct scb *scb;
union ccb *ccb;
struct ahc_softc *ahc;
int s;
scb = (struct scb *)arg;
ccb = scb->ccb;
ahc = (struct ahc_softc *)ccb->ccb_h.ccb_ahc_ptr;
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
if (error != 0) {
if (error == EFBIG)
ahcsetccbstatus(scb->ccb, CAM_REQ_TOO_BIG);
else
ahcsetccbstatus(scb->ccb, CAM_REQ_CMP_ERR);
if (nsegments != 0)
bus_dmamap_unload(ahc->buffer_dmat, scb->dmamap);
ahcfreescb(ahc, scb);
xpt_done(ccb);
return;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (nsegments != 0) {
struct ahc_dma_seg *sg;
bus_dma_segment_t *end_seg;
bus_dmasync_op_t op;
end_seg = dm_segs + nsegments;
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
/* Copy the segments into our SG list */
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
sg = scb->sg_list;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
while (dm_segs < end_seg) {
sg->addr = dm_segs->ds_addr;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
/* XXX Add in the 5th byte of the address later. */
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
sg->len = dm_segs->ds_len;
sg++;
dm_segs++;
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
/*
* Note where to find the SG entries in bus space.
* We also set the full residual flag which the
* sequencer will clear as soon as a data transfer
* occurs.
*/
scb->hscb->sgptr = scb->sg_list_phys | SG_FULL_RESID;
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if ((ccb->ccb_h.flags & CAM_DIR_MASK) == CAM_DIR_IN)
op = BUS_DMASYNC_PREREAD;
else
op = BUS_DMASYNC_PREWRITE;
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
bus_dmamap_sync(ahc->buffer_dmat, scb->dmamap, op);
Correct a race condition where a target mode SCB dma could occur before we finished setting the pending phase information. Use the ALL_TARGETS_MASK #define instead of 0xffff for the default ultraenb mask.
1999-02-11 07:07:27 +00:00
if (ccb->ccb_h.func_code == XPT_CONT_TARGET_IO) {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
struct target_data *tdata;
tdata = &scb->hscb->shared_data.tdata;
tdata->target_phases |= DPHASE_PENDING;
if ((ccb->ccb_h.flags & CAM_DIR_MASK) == CAM_DIR_OUT)
tdata->data_phase = P_DATAOUT;
else
tdata->data_phase = P_DATAIN;
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
/*
* If the transfer is of an odd length and in the
* "in" direction (scsi->HostBus), then it may
* trigger a bug in the 'WideODD' feature of
* non-Ultra2 chips. Force the total data-length
* to be even by adding an extra, 1 byte, SG,
* element. We do this even if we are not currently
* negotiated wide as negotiation could occur before
* this command is executed.
*/
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if ((ahc->bugs & AHC_TMODE_WIDEODD_BUG) != 0
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
&& (ccb->csio.dxfer_len & 0x1) != 0
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
&& (ccb->ccb_h.flags & CAM_DIR_MASK) == CAM_DIR_IN) {
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
nsegments++;
if (nsegments > AHC_NSEG) {
ahcsetccbstatus(scb->ccb,
CAM_REQ_TOO_BIG);
bus_dmamap_unload(ahc->buffer_dmat,
scb->dmamap);
ahcfreescb(ahc, scb);
xpt_done(ccb);
return;
}
sg->addr = ahc->dma_bug_buf;
sg->len = 1;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
sg++;
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
}
Correct a race condition where a target mode SCB dma could occur before we finished setting the pending phase information. Use the ALL_TARGETS_MASK #define instead of 0xffff for the default ultraenb mask.
1999-02-11 07:07:27 +00:00
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
sg--;
sg->len |= AHC_DMA_LAST_SEG;
/* Copy the first SG into the "current" data pointer area */
scb->hscb->dataptr = scb->sg_list->addr;
scb->hscb->datacnt = scb->sg_list->len;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
} else {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
scb->hscb->sgptr = SG_LIST_NULL;
scb->hscb->dataptr = 0;
scb->hscb->datacnt = 0;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
scb->sg_count = nsegments;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
s = splcam();
/*
* Last time we need to check if this SCB needs to
* be aborted.
*/
if (ahc_ccb_status(ccb) != CAM_REQ_INPROG) {
if (nsegments != 0)
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
bus_dmamap_unload(ahc->buffer_dmat, scb->dmamap);
ahcfreescb(ahc, scb);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
xpt_done(ccb);
splx(s);
return;
}
LIST_INSERT_HEAD(&ahc->pending_ccbs, &ccb->ccb_h,
sim_links.le);
ccb->ccb_h.status |= CAM_SIM_QUEUED;
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
if (ccb->ccb_h.timeout != CAM_TIME_INFINITY) {
if (ccb->ccb_h.timeout == CAM_TIME_DEFAULT)
ccb->ccb_h.timeout = 5 * 1000;
ccb->ccb_h.timeout_ch =
timeout(ahc_timeout, (caddr_t)scb,
(ccb->ccb_h.timeout * hz) / 1000);
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
/*
* We only allow one untagged transaction
* per target in the initiator role unless
* we are storing a full busy target *lun*
* table in SCB space.
*/
if ((scb->hscb->control & (TARGET_SCB|TAG_ENB)) == 0
&& (ahc->features & AHC_SCB_BTT) == 0) {
struct scb_tailq *untagged_q;
untagged_q = &(ahc->untagged_queues[ccb->ccb_h.target_id]);
TAILQ_INSERT_TAIL(untagged_q, scb, links.tqe);
if (TAILQ_FIRST(untagged_q) != scb) {
splx(s);
return;
}
}
scb->flags |= SCB_ACTIVE;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if ((scb->flags & SCB_TARGET_IMMEDIATE) != 0) {
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
#if 0
printf("Continueing Immediate Command %d:%d\n",
ccb->ccb_h.target_id, ccb->ccb_h.target_lun);
#endif
pause_sequencer(ahc);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if ((ahc->flags & AHC_PAGESCBS) == 0)
ahc_outb(ahc, SCBPTR, scb->hscb->tag);
ahc_outb(ahc, SCB_TAG, scb->hscb->tag);
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc_outb(ahc, RETURN_1, CONT_MSG_LOOP);
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
unpause_sequencer(ahc);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
} else {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahc_queue_scb(ahc, scb);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
splx(s);
}
static void
ahc_poll(struct cam_sim *sim)
{
ahc_intr(cam_sim_softc(sim));
}
static void
ahc_setup_data(struct ahc_softc *ahc, struct ccb_scsiio *csio,
struct scb *scb)
{
struct hardware_scb *hscb;
struct ccb_hdr *ccb_h;
hscb = scb->hscb;
ccb_h = &csio->ccb_h;
if (ccb_h->func_code == XPT_SCSI_IO) {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
hscb->cdb_len = csio->cdb_len;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if ((ccb_h->flags & CAM_CDB_POINTER) != 0) {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (hscb->cdb_len > sizeof(hscb->cdb32)
|| (ccb_h->flags & CAM_CDB_PHYS) != 0) {
ahcsetccbstatus(scb->ccb, CAM_REQ_INVALID);
xpt_done(scb->ccb);
ahcfreescb(ahc, scb);
return;
}
if (hscb->cdb_len > 12) {
memcpy(hscb->cdb32,
csio->cdb_io.cdb_ptr,
hscb->cdb_len);
if ((ahc->flags & AHC_CMD_CHAN) == 0) {
hscb->shared_data.cdb_ptr =
scb->cdb32_busaddr;
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
} else {
memcpy(hscb->shared_data.cdb,
csio->cdb_io.cdb_ptr,
hscb->cdb_len);
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
} else {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (hscb->cdb_len > 12) {
memcpy(hscb->cdb32, csio->cdb_io.cdb_bytes,
hscb->cdb_len);
if ((ahc->flags & AHC_CMD_CHAN) == 0) {
hscb->shared_data.cdb_ptr =
scb->cdb32_busaddr;
}
} else {
memcpy(hscb->shared_data.cdb,
csio->cdb_io.cdb_bytes,
hscb->cdb_len);
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
}
/* Only use S/G if there is a transfer */
if ((ccb_h->flags & CAM_DIR_MASK) != CAM_DIR_NONE) {
if ((ccb_h->flags & CAM_SCATTER_VALID) == 0) {
/* We've been given a pointer to a single buffer */
if ((ccb_h->flags & CAM_DATA_PHYS) == 0) {
int s;
int error;
s = splsoftvm();
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
error = bus_dmamap_load(ahc->buffer_dmat,
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
scb->dmamap,
csio->data_ptr,
csio->dxfer_len,
ahc_execute_scb,
scb, /*flags*/0);
if (error == EINPROGRESS) {
/*
* So as to maintain ordering,
* freeze the controller queue
* until our mapping is
* returned.
*/
xpt_freeze_simq(ahc->sim,
/*count*/1);
scb->ccb->ccb_h.status |=
CAM_RELEASE_SIMQ;
}
splx(s);
} else {
struct bus_dma_segment seg;
/* Pointer to physical buffer */
if (csio->dxfer_len > AHC_MAXTRANSFER_SIZE)
panic("ahc_setup_data - Transfer size "
"larger than can device max");
* Completely rewrite the alpha busspace to hide the implementation from the drivers. * Remove legacy inx/outx support from chipset and replace with macros which call busspace. * Rework pci config accesses to route through the pcib device instead of calling a MD function directly. With these changes it is possible to cleanly support machines which have more than one independantly numbered PCI busses. As a bonus, the new busspace implementation should be measurably faster than the old one.
2000-08-28 21:48:13 +00:00
seg.ds_addr = (intptr_t)csio->data_ptr;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
seg.ds_len = csio->dxfer_len;
ahc_execute_scb(scb, &seg, 1, 0);
}
} else {
struct bus_dma_segment *segs;
if ((ccb_h->flags & CAM_DATA_PHYS) != 0)
panic("ahc_setup_data - Physical segment "
"pointers unsupported");
if ((ccb_h->flags & CAM_SG_LIST_PHYS) == 0)
panic("ahc_setup_data - Virtual segment "
"addresses unsupported");
/* Just use the segments provided */
segs = (struct bus_dma_segment *)csio->data_ptr;
ahc_execute_scb(scb, segs, csio->sglist_cnt, 0);
}
} else {
ahc_execute_scb(scb, NULL, 0, 0);
}
}
static void
ahc_freeze_devq(struct ahc_softc *ahc, struct cam_path *path)
{
int target;
char channel;
int lun;
target = xpt_path_target_id(path);
lun = xpt_path_lun_id(path);
channel = xpt_path_sim(path)->bus_id == 0 ? 'A' : 'B';
ahc_search_qinfifo(ahc, target, channel, lun,
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
/*tag*/SCB_LIST_NULL, ROLE_UNKNOWN,
CAM_REQUEUE_REQ, SEARCH_COMPLETE);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
static void
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
ahcallocscbs(struct ahc_softc *ahc)
{
struct scb_data *scb_data;
struct scb *next_scb;
struct sg_map_node *sg_map;
bus_addr_t physaddr;
struct ahc_dma_seg *segs;
int newcount;
int i;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
scb_data = ahc->scb_data;
if (scb_data->numscbs >= AHC_SCB_MAX)
/* Can't allocate any more */
return;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
next_scb = &scb_data->scbarray[scb_data->numscbs];
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
sg_map = malloc(sizeof(*sg_map), M_DEVBUF, M_NOWAIT);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
if (sg_map == NULL)
return;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
/* Allocate S/G space for the next batch of SCBS */
if (bus_dmamem_alloc(scb_data->sg_dmat, (void **)&sg_map->sg_vaddr,
BUS_DMA_NOWAIT, &sg_map->sg_dmamap) != 0) {
free(sg_map, M_DEVBUF);
return;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
SLIST_INSERT_HEAD(&scb_data->sg_maps, sg_map, links);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
bus_dmamap_load(scb_data->sg_dmat, sg_map->sg_dmamap, sg_map->sg_vaddr,
PAGE_SIZE, ahcdmamapcb, &sg_map->sg_physaddr,
/*flags*/0);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
segs = sg_map->sg_vaddr;
physaddr = sg_map->sg_physaddr;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
newcount = (PAGE_SIZE / (AHC_NSEG * sizeof(struct ahc_dma_seg)));
for (i = 0; scb_data->numscbs < AHC_SCB_MAX && i < newcount; i++) {
int error;
next_scb->sg_list = segs;
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
/*
* The sequencer always starts with the second entry.
* The first entry is embedded in the scb.
*/
next_scb->sg_list_phys = physaddr + sizeof(struct ahc_dma_seg);
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
next_scb->flags = SCB_FREE;
error = bus_dmamap_create(ahc->buffer_dmat, /*flags*/0,
&next_scb->dmamap);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (error != 0)
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
break;
next_scb->hscb = &scb_data->hscbs[scb_data->numscbs];
next_scb->hscb->tag = ahc->scb_data->numscbs;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
next_scb->cdb32_busaddr =
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
ahc_hscb_busaddr(ahc, next_scb->hscb->tag)
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
+ offsetof(struct hardware_scb, cdb32);
SLIST_INSERT_HEAD(&ahc->scb_data->free_scbs,
next_scb, links.sle);
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
segs += AHC_NSEG;
physaddr += (AHC_NSEG * sizeof(struct ahc_dma_seg));
next_scb++;
ahc->scb_data->numscbs++;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
}
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
#ifdef AHC_DUMP_SEQ
static void
ahc_dumpseq(struct ahc_softc* ahc)
{
int i;
int max_prog;
if ((ahc->chip & AHC_BUS_MASK) < AHC_PCI)
max_prog = 448;
else if ((ahc->features & AHC_ULTRA2) != 0)
max_prog = 768;
else
max_prog = 512;
ahc_outb(ahc, SEQCTL, PERRORDIS|FAILDIS|FASTMODE|LOADRAM);
ahc_outb(ahc, SEQADDR0, 0);
ahc_outb(ahc, SEQADDR1, 0);
for (i = 0; i < max_prog; i++) {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
uint8_t ins_bytes[4];
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
ahc_insb(ahc, SEQRAM, ins_bytes, 4);
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
printf("0x%08x\n", ins_bytes[0] << 24
| ins_bytes[1] << 16
| ins_bytes[2] << 8
| ins_bytes[3]);
Simplify my copyright license terms. aic7xxx.c: Add a function for sucking firmware out of the controller prior to reset. Remove some inline bloat from functions that should not have been inlined. During initialization, wait 1ms after the chip reset before touching any registers. You can get machine checks on certain architectures (Atari I think?) without the delay. Return CAM_REQ_CMP for external BDR requests instead of CAM_BDR_SENT. Bump some messages to bootverbose levels above 1. Don't clear any negotiated sync rate if the target rejects a WDTR message. The sync rate is only cleared if the target accepts a WDTR message. Fix a small bug in the mesgin handling code that could cause us to believe that we had recieved a message that was actually received by another target. This could only confuse us in some very rare transmission negotiation scenarios. Remove some unecessary cleanup of residual information after a residual is reported. The sequencer does this when the command is queued now.
1999-12-06 18:23:31 +00:00
}
}
#endif
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static void
ahc_loadseq(struct ahc_softc *ahc)
{
struct patch *cur_patch;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int i;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
int downloaded;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int skip_addr;
uint8_t download_consts[2];
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/* Setup downloadable constant table */
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
download_consts[QOUTFIFO_OFFSET] = 0;
if (ahc->targetcmds != NULL)
download_consts[QOUTFIFO_OFFSET] += 32;
download_consts[QINFIFO_OFFSET] = download_consts[QOUTFIFO_OFFSET] + 1;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
cur_patch = patches;
downloaded = 0;
skip_addr = 0;
ahc_outb(ahc, SEQCTL, PERRORDIS|FAILDIS|FASTMODE|LOADRAM);
ahc_outb(ahc, SEQADDR0, 0);
ahc_outb(ahc, SEQADDR1, 0);
for (i = 0; i < sizeof(seqprog)/4; i++) {
if (ahc_check_patch(ahc, &cur_patch, i, &skip_addr) == 0) {
/*
* Don't download this instruction as it
* is in a patch that was removed.
*/
continue;
}
ahc_download_instr(ahc, i, download_consts);
downloaded++;
}
ahc_outb(ahc, SEQCTL, PERRORDIS|FAILDIS|FASTMODE);
restart_sequencer(ahc);
if (bootverbose)
printf(" %d instructions downloaded\n", downloaded);
}
static int
ahc_check_patch(struct ahc_softc *ahc, struct patch **start_patch,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int start_instr, u_int *skip_addr)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
struct patch *cur_patch;
struct patch *last_patch;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int num_patches;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
num_patches = sizeof(patches)/sizeof(struct patch);
last_patch = &patches[num_patches];
cur_patch = *start_patch;
while (cur_patch < last_patch && start_instr == cur_patch->begin) {
if (cur_patch->patch_func(ahc) == 0) {
/* Start rejecting code */
*skip_addr = start_instr + cur_patch->skip_instr;
cur_patch += cur_patch->skip_patch;
} else {
/* Accepted this patch. Advance to the next
* one and wait for our intruction pointer to
* hit this point.
*/
cur_patch++;
}
}
*start_patch = cur_patch;
if (start_instr < *skip_addr)
/* Still skipping */
return (0);
return (1);
}
static void
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahc_download_instr(struct ahc_softc *ahc, u_int instrptr, uint8_t *dconsts)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
union ins_formats instr;
struct ins_format1 *fmt1_ins;
struct ins_format3 *fmt3_ins;
u_int opcode;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/* Structure copy */
instr = *(union ins_formats*)&seqprog[instrptr * 4];
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
#if BYTE_ORDER == BIG_ENDIAN
opcode = instr.format.bytes[0];
instr.format.bytes[0] = instr.format.bytes[3];
instr.format.bytes[3] = opcode;
opcode = instr.format.bytes[1];
instr.format.bytes[1] = instr.format.bytes[2];
instr.format.bytes[2] = opcode;
#endif
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
fmt1_ins = &instr.format1;
fmt3_ins = NULL;
/* Pull the opcode */
opcode = instr.format1.opcode;
switch (opcode) {
case AIC_OP_JMP:
case AIC_OP_JC:
case AIC_OP_JNC:
case AIC_OP_CALL:
case AIC_OP_JNE:
case AIC_OP_JNZ:
case AIC_OP_JE:
case AIC_OP_JZ:
{
struct patch *cur_patch;
int address_offset;
u_int address;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
u_int skip_addr;
u_int i;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
fmt3_ins = &instr.format3;
address_offset = 0;
address = fmt3_ins->address;
cur_patch = patches;
skip_addr = 0;
for (i = 0; i < address;) {
ahc_check_patch(ahc, &cur_patch, i, &skip_addr);
if (skip_addr > i) {
int end_addr;
end_addr = MIN(address, skip_addr);
address_offset += end_addr - i;
i = skip_addr;
} else {
i++;
}
}
address -= address_offset;
fmt3_ins->address = address;
/* FALLTHROUGH */
}
case AIC_OP_OR:
case AIC_OP_AND:
case AIC_OP_XOR:
case AIC_OP_ADD:
case AIC_OP_ADC:
case AIC_OP_BMOV:
if (fmt1_ins->parity != 0) {
fmt1_ins->immediate = dconsts[fmt1_ins->immediate];
}
fmt1_ins->parity = 0;
/* FALLTHROUGH */
case AIC_OP_ROL:
if ((ahc->features & AHC_ULTRA2) != 0) {
int i, count;
/* Calculate odd parity for the instruction */
for (i = 0, count = 0; i < 31; i++) {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
uint32_t mask;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
mask = 0x01 << i;
if ((instr.integer & mask) != 0)
count++;
}
if ((count & 0x01) == 0)
instr.format1.parity = 1;
} else {
/* Compress the instruction for older sequencers */
if (fmt3_ins != NULL) {
instr.integer =
fmt3_ins->immediate
| (fmt3_ins->source << 8)
| (fmt3_ins->address << 16)
| (fmt3_ins->opcode << 25);
} else {
instr.integer =
fmt1_ins->immediate
| (fmt1_ins->source << 8)
| (fmt1_ins->destination << 16)
| (fmt1_ins->ret << 24)
| (fmt1_ins->opcode << 25);
}
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
#if BYTE_ORDER == BIG_ENDIAN
opcode = instr.format.bytes[0];
instr.format.bytes[0] = instr.format.bytes[3];
instr.format.bytes[3] = opcode;
opcode = instr.format.bytes[1];
instr.format.bytes[1] = instr.format.bytes[2];
instr.format.bytes[2] = opcode;
#endif
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_outsb(ahc, SEQRAM, instr.bytes, 4);
break;
default:
panic("Unknown opcode encountered in seq program");
break;
}
}
static void
ahc_set_recoveryscb(struct ahc_softc *ahc, struct scb *scb) {
if ((scb->flags & SCB_RECOVERY_SCB) == 0) {
struct ccb_hdr *ccbh;
scb->flags |= SCB_RECOVERY_SCB;
/*
* Take all queued, but not sent SCBs out of the equation.
* Also ensure that no new CCBs are queued to us while we
* try to fix this problem.
*/
if ((scb->ccb->ccb_h.status & CAM_RELEASE_SIMQ) == 0) {
xpt_freeze_simq(ahc->sim, /*count*/1);
scb->ccb->ccb_h.status |= CAM_RELEASE_SIMQ;
}
/*
* Go through all of our pending SCBs and remove
* any scheduled timeouts for them. We will reschedule
* them after we've successfully fixed this problem.
*/
ccbh = ahc->pending_ccbs.lh_first;
while (ccbh != NULL) {
struct scb *pending_scb;
pending_scb = (struct scb *)ccbh->ccb_scb_ptr;
untimeout(ahc_timeout, pending_scb, ccbh->timeout_ch);
ccbh = ccbh->sim_links.le.le_next;
}
}
}
static void
ahc_timeout(void *arg)
{
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
struct scb *scb;
struct ahc_softc *ahc;
int s, found;
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
u_int last_phase;
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
int target;
int lun;
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
int i;
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
char channel;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
scb = (struct scb *)arg;
ahc = (struct ahc_softc *)scb->ccb->ccb_h.ccb_ahc_ptr;
s = splcam();
/*
* Ensure that the card doesn't do anything
* behind our back. Also make sure that we
* didn't "just" miss an interrupt that would
* affect this timeout.
*/
do {
ahc_intr(ahc);
pause_sequencer(ahc);
} while (ahc_inb(ahc, INTSTAT) & INT_PEND);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
xpt_print_path(scb->ccb->ccb_h.path);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if ((scb->flags & SCB_ACTIVE) == 0) {
/* Previous timeout took care of me already */
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
printf("Timedout SCB %d handled by another timeout\n",
scb->hscb->tag);
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
unpause_sequencer(ahc);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
splx(s);
return;
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
target = SCB_GET_TARGET(ahc, scb);
channel = SCB_GET_CHANNEL(ahc, scb);
lun = SCB_GET_LUN(scb);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
printf("SCB 0x%x - timed out ", scb->hscb->tag);
/*
* Take a snapshot of the bus state and print out
* some information so we can track down driver bugs.
*/
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
last_phase = ahc_inb(ahc, LASTPHASE);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
for (i = 0; i < num_phases; i++) {
if (last_phase == phase_table[i].phase)
break;
}
printf("%s", phase_table[i].phasemsg);
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
printf(", SEQADDR == 0x%x\n",
ahc_inb(ahc, SEQADDR0) | (ahc_inb(ahc, SEQADDR1) << 8));
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
#if 0
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
printf("SSTAT1 == 0x%x\n", ahc_inb(ahc, SSTAT1));
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
printf("SSTAT3 == 0x%x\n", ahc_inb(ahc, SSTAT3));
printf("SCSIPHASE == 0x%x\n", ahc_inb(ahc, SCSIPHASE));
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
printf("SCSIRATE == 0x%x\n", ahc_inb(ahc, SCSIRATE));
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
printf("SCSIOFFSET == 0x%x\n", ahc_inb(ahc, SCSIOFFSET));
printf("SEQ_FLAGS == 0x%x\n", ahc_inb(ahc, SEQ_FLAGS));
printf("SCB_DATAPTR == 0x%x\n", ahc_inb(ahc, SCB_DATAPTR)
| ahc_inb(ahc, SCB_DATAPTR + 1) << 8
| ahc_inb(ahc, SCB_DATAPTR + 2) << 16
| ahc_inb(ahc, SCB_DATAPTR + 3) << 24);
printf("SCB_DATACNT == 0x%x\n", ahc_inb(ahc, SCB_DATACNT)
| ahc_inb(ahc, SCB_DATACNT + 1) << 8
| ahc_inb(ahc, SCB_DATACNT + 2) << 16);
printf("SCB_SGCOUNT == 0x%x\n", ahc_inb(ahc, SCB_SGCOUNT));
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
printf("CCSCBCTL == 0x%x\n", ahc_inb(ahc, CCSCBCTL));
printf("CCSCBCNT == 0x%x\n", ahc_inb(ahc, CCSCBCNT));
printf("DFCNTRL == 0x%x\n", ahc_inb(ahc, DFCNTRL));
printf("DFSTATUS == 0x%x\n", ahc_inb(ahc, DFSTATUS));
printf("CCHCNT == 0x%x\n", ahc_inb(ahc, CCHCNT));
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
if (scb->sg_count > 0) {
for (i = 0; i < scb->sg_count; i++) {
printf("sg[%d] - Addr 0x%x : Length %d\n",
i,
scb->sg_list[i].addr,
scb->sg_list[i].len);
}
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
#endif
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
if (scb->flags & (SCB_DEVICE_RESET|SCB_ABORT)) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Been down this road before.
* Do a full bus reset.
*/
bus_reset:
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
ahcsetccbstatus(scb->ccb, CAM_CMD_TIMEOUT);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
found = ahc_reset_channel(ahc, channel, /*Initiate Reset*/TRUE);
printf("%s: Issued Channel %c Bus Reset. "
"%d SCBs aborted\n", ahc_name(ahc), channel, found);
} else {
/*
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
* If we are a target, transition to bus free and report
* the timeout.
*
* The target/initiator that is holding up the bus may not
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
* be the same as the one that triggered this timeout
* (different commands have different timeout lengths).
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
* If the bus is idle and we are actiing as the initiator
* for this request, queue a BDR message to the timed out
* target. Otherwise, if the timed out transaction is
* active:
* Initiator transaction:
* Stuff the message buffer with a BDR message and assert
* ATN in the hopes that the target will let go of the bus
* and go to the mesgout phase. If this fails, we'll
* get another timeout 2 seconds later which will attempt
* a bus reset.
*
* Target transaction:
* Transition to BUS FREE and report the error.
* It's good to be the target!
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
u_int active_scb_index;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
active_scb_index = ahc_inb(ahc, SCB_TAG);
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
if (last_phase != P_BUSFREE
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
&& (active_scb_index < ahc->scb_data->numscbs)) {
struct scb *active_scb;
/*
* If the active SCB is not from our device,
* assume that another device is hogging the bus
* and wait for it's timeout to expire before
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
* taking additional action.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
active_scb = &ahc->scb_data->scbarray[active_scb_index];
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (active_scb->hscb->scsiid != scb->hscb->scsiid
|| active_scb->hscb->lun != scb->hscb->lun) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
struct ccb_hdr *ccbh;
u_int newtimeout;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
xpt_print_path(scb->ccb->ccb_h.path);
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
printf("Other SCB Timeout%s",
(scb->flags & SCB_OTHERTCL_TIMEOUT) != 0
? " again\n" : "\n");
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
scb->flags |= SCB_OTHERTCL_TIMEOUT;
newtimeout = MAX(active_scb->ccb->ccb_h.timeout,
scb->ccb->ccb_h.timeout);
ccbh = &scb->ccb->ccb_h;
scb->ccb->ccb_h.timeout_ch =
timeout(ahc_timeout, scb,
(newtimeout * hz) / 1000);
splx(s);
return;
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
}
/* It's us */
if ((scb->hscb->control & TARGET_SCB) != 0) {
/*
* Send back any queued up transactions
* and properly record the error condition.
*/
ahc_freeze_devq(ahc, scb->ccb->ccb_h.path);
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
ahcsetccbstatus(scb->ccb, CAM_CMD_TIMEOUT);
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
ahc_freeze_ccb(scb->ccb);
ahc_done(ahc, scb);
/* Will clear us from the bus */
restart_sequencer(ahc);
return;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_set_recoveryscb(ahc, active_scb);
ahc_outb(ahc, MSG_OUT, MSG_BUS_DEV_RESET);
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
ahc_outb(ahc, SCSISIGO, last_phase|ATNO);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
xpt_print_path(active_scb->ccb->ccb_h.path);
printf("BDR message in message buffer\n");
active_scb->flags |= SCB_DEVICE_RESET;
active_scb->ccb->ccb_h.timeout_ch =
timeout(ahc_timeout, (caddr_t)active_scb, 2 * hz);
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
unpause_sequencer(ahc);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
} else {
int disconnected;
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
/* XXX Shouldn't panic. Just punt instead */
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
if ((scb->hscb->control & TARGET_SCB) != 0)
panic("Timed-out target SCB but bus idle");
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
if (last_phase != P_BUSFREE
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
&& (ahc_inb(ahc, SSTAT0) & TARGET) != 0) {
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
/* XXX What happened to the SCB? */
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
/* Hung target selection. Goto busfree */
printf("%s: Hung target selection\n",
ahc_name(ahc));
restart_sequencer(ahc);
return;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (ahc_search_qinfifo(ahc, target, channel, lun,
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
scb->hscb->tag, ROLE_INITIATOR,
/*status*/0, SEARCH_COUNT) > 0) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
disconnected = FALSE;
} else {
disconnected = TRUE;
}
if (disconnected) {
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
u_int active_scb;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_set_recoveryscb(ahc, scb);
/*
* Simply set the MK_MESSAGE control bit.
*/
scb->hscb->control |= MK_MESSAGE;
scb->flags |= SCB_QUEUED_MSG
| SCB_DEVICE_RESET;
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
/*
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
* Mark the cached copy of this SCB in the
* disconnected list too, so that a reconnect
* at this point causes a BDR or abort.
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
*/
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
active_scb = ahc_inb(ahc, SCBPTR);
if (ahc_search_disc_list(ahc, target,
channel, lun,
scb->hscb->tag,
/*stop_on_first*/TRUE,
/*remove*/FALSE,
/*save_state*/FALSE)) {
u_int scb_control;
scb_control = ahc_inb(ahc, SCB_CONTROL);
scb_control |= MK_MESSAGE;
ahc_outb(ahc, SCB_CONTROL, scb_control);
}
ahc_outb(ahc, SCBPTR, active_scb);
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Actually re-queue this SCB in case we can
* select the device before it reconnects.
* Clear out any entries in the QINFIFO first
* so we are the next SCB for this target
* to run.
*/
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahc_search_qinfifo(ahc,
SCB_GET_TARGET(ahc, scb),
channel, SCB_GET_LUN(scb),
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
SCB_LIST_NULL,
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
ROLE_INITIATOR,
Correct an "argument reversal" bug that could cause commands requed from the input fifo to be returned as successful and frozen. Most, if not all, peripheral drivers do not check the qfrozen bit for successfully completed commands, so the result would not only be lost commands, but devices locked out from receiving commands. This was a bad bug that crept in two or three months ago during some target mode work.
1999-12-20 21:32:32 +00:00
CAM_REQUEUE_REQ,
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
SEARCH_COMPLETE);
xpt_print_path(scb->ccb->ccb_h.path);
printf("Queuing a BDR SCB\n");
ahc->qinfifo[ahc->qinfifonext++] =
scb->hscb->tag;
if ((ahc->features & AHC_QUEUE_REGS) != 0) {
ahc_outb(ahc, HNSCB_QOFF,
ahc->qinfifonext);
} else {
ahc_outb(ahc, KERNEL_QINPOS,
ahc->qinfifonext);
}
scb->ccb->ccb_h.timeout_ch =
timeout(ahc_timeout, (caddr_t)scb, 2 * hz);
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
unpause_sequencer(ahc);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
} else {
/* Go "immediatly" to the bus reset */
/* This shouldn't happen */
ahc_set_recoveryscb(ahc, scb);
xpt_print_path(scb->ccb->ccb_h.path);
printf("SCB %d: Immediate reset. "
"Flags = 0x%x\n", scb->hscb->tag,
scb->flags);
goto bus_reset;
}
}
}
splx(s);
}
static int
ahc_search_qinfifo(struct ahc_softc *ahc, int target, char channel,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
int lun, u_int tag, role_t role, uint32_t status,
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_search_action action)
{
struct scb *scbp;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
uint8_t qinpos;
uint8_t qintail;
uint8_t next, prev;
uint8_t curscbptr;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
int found;
qinpos = ahc_inb(ahc, QINPOS);
qintail = ahc->qinfifonext;
found = 0;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (action == SEARCH_COMPLETE) {
/*
* Don't attempt to run any queued untagged transactions
* until we are done with the abort process.
*/
ahc_freeze_untagged_queues(ahc);
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Start with an empty queue. Entries that are not chosen
* for removal will be re-added to the queue as we go.
*/
ahc->qinfifonext = qinpos;
while (qinpos != qintail) {
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
scbp = &ahc->scb_data->scbarray[ahc->qinfifo[qinpos]];
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (ahc_match_scb(ahc, scbp, target, channel, lun, tag, role)) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
* We found an scb that needs to be acted on.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
found++;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
switch (action) {
case SEARCH_COMPLETE:
if (ahc_ccb_status(scbp->ccb) == CAM_REQ_INPROG)
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
ahcsetccbstatus(scbp->ccb, status);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_freeze_ccb(scbp->ccb);
ahc_done(ahc, scbp);
break;
case SEARCH_COUNT:
ahc->qinfifo[ahc->qinfifonext++] =
scbp->hscb->tag;
break;
case SEARCH_REMOVE:
break;
}
} else {
ahc->qinfifo[ahc->qinfifonext++] = scbp->hscb->tag;
}
qinpos++;
}
if ((ahc->features & AHC_QUEUE_REGS) != 0) {
ahc_outb(ahc, HNSCB_QOFF, ahc->qinfifonext);
} else {
ahc_outb(ahc, KERNEL_QINPOS, ahc->qinfifonext);
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
/*
* Search waiting for selection list.
*/
curscbptr = ahc_inb(ahc, SCBPTR);
next = ahc_inb(ahc, WAITING_SCBH); /* Start at head of list. */
prev = SCB_LIST_NULL;
while (next != SCB_LIST_NULL) {
uint8_t scb_index;
ahc_outb(ahc, SCBPTR, next);
scb_index = ahc_inb(ahc, SCB_TAG);
if (scb_index >= ahc->scb_data->numscbs) {
panic("Waiting List inconsistency. "
"SCB index == %d, yet numscbs == %d.",
scb_index, ahc->scb_data->numscbs);
}
scbp = &ahc->scb_data->scbarray[scb_index];
if (ahc_match_scb(ahc, scbp, target, channel,
lun, SCB_LIST_NULL, role)) {
/*
* We found an scb that needs to be acted on.
*/
found++;
switch (action) {
case SEARCH_REMOVE:
next = ahc_rem_wscb(ahc, next, prev);
break;
case SEARCH_COMPLETE:
next = ahc_rem_wscb(ahc, next, prev);
if (ahc_ccb_status(scbp->ccb) == CAM_REQ_INPROG)
ahcsetccbstatus(scbp->ccb, status);
ahc_freeze_ccb(scbp->ccb);
ahc_done(ahc, scbp);
break;
case SEARCH_COUNT:
prev = next;
next = ahc_inb(ahc, SCB_NEXT);
break;
}
} else {
prev = next;
next = ahc_inb(ahc, SCB_NEXT);
}
}
ahc_outb(ahc, SCBPTR, curscbptr);
if (action == SEARCH_COMPLETE)
ahc_release_untagged_queues(ahc);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
return (found);
}
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
static void
ahc_abort_ccb(struct ahc_softc *ahc, struct cam_sim *sim, union ccb *ccb)
{
union ccb *abort_ccb;
abort_ccb = ccb->cab.abort_ccb;
switch (abort_ccb->ccb_h.func_code) {
case XPT_ACCEPT_TARGET_IO:
case XPT_IMMED_NOTIFY:
case XPT_CONT_TARGET_IO:
{
struct tmode_tstate *tstate;
struct tmode_lstate *lstate;
struct ccb_hdr_slist *list;
cam_status status;
status = ahc_find_tmode_devs(ahc, sim, abort_ccb, &tstate,
&lstate, TRUE);
if (status != CAM_REQ_CMP) {
ccb->ccb_h.status = status;
break;
}
if (abort_ccb->ccb_h.func_code == XPT_ACCEPT_TARGET_IO)
list = &lstate->accept_tios;
else if (abort_ccb->ccb_h.func_code == XPT_IMMED_NOTIFY)
list = &lstate->immed_notifies;
else
list = NULL;
if (list != NULL) {
struct ccb_hdr *curelm;
int found;
curelm = SLIST_FIRST(list);
found = 0;
if (curelm == &abort_ccb->ccb_h) {
found = 1;
SLIST_REMOVE_HEAD(list, sim_links.sle);
} else {
while(curelm != NULL) {
struct ccb_hdr *nextelm;
nextelm =
SLIST_NEXT(curelm, sim_links.sle);
if (nextelm == &abort_ccb->ccb_h) {
found = 1;
SLIST_NEXT(curelm,
sim_links.sle) =
SLIST_NEXT(nextelm,
sim_links.sle);
break;
}
curelm = nextelm;
}
}
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
if (found) {
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
abort_ccb->ccb_h.status = CAM_REQ_ABORTED;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
xpt_done(abort_ccb);
ccb->ccb_h.status = CAM_REQ_CMP;
} else {
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
printf("Not found\n");
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
ccb->ccb_h.status = CAM_PATH_INVALID;
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
}
Perform a save data pointers operation if a data transfer was performed in target mode, but we are not completing the command. Use a template of allowed bus arbitration phases to selectively and dynamically enable/disable initiator or target (re)selection. Properly handle timeouts for target role transactions - just go to the bus free state and report the error to the peripheral driver. Checkpoint support for the XPT_ABORT_CCB function code. This currently handles the accept tio and immediate notify ccb types, but does not handle the continue target I/O or SCSI I/O ccb types. This is enough to handle dynamic target enable/disable events. Clean up the SCSI reset code so that we perform at most 1 SCSI bus reset at initialization, the reset requested by the XPT layer.
1998-12-15 08:22:42 +00:00
break;
}
/* FALLTHROUGH */
}
case XPT_SCSI_IO:
/* XXX Fully implement the hard ones */
ccb->ccb_h.status = CAM_UA_ABORT;
break;
default:
ccb->ccb_h.status = CAM_REQ_INVALID;
break;
}
xpt_done(ccb);
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Abort all SCBs that match the given description (target/channel/lun/tag),
* setting their status to the passed in status if the status has not already
* been modified from CAM_REQ_INPROG. This routine assumes that the sequencer
* is paused before it is called.
*/
static int
ahc_abort_scbs(struct ahc_softc *ahc, int target, char channel,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
int lun, u_int tag, role_t role, uint32_t status)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
struct scb *scbp;
u_int active_scb;
int i;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
int maxtarget;
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
int found;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
/*
* Don't attempt to run any queued untagged transactions
* until we are done with the abort process.
*/
ahc_freeze_untagged_queues(ahc);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/* restore this when we're done */
active_scb = ahc_inb(ahc, SCBPTR);
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
found = ahc_search_qinfifo(ahc, target, channel, lun, SCB_LIST_NULL,
Correct an "argument reversal" bug that could cause commands requed from the input fifo to be returned as successful and frozen. Most, if not all, peripheral drivers do not check the qfrozen bit for successfully completed commands, so the result would not only be lost commands, but devices locked out from receiving commands. This was a bad bug that crept in two or three months ago during some target mode work.
1999-12-20 21:32:32 +00:00
role, CAM_REQUEUE_REQ, SEARCH_COMPLETE);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
* Clean out the busy target table for any untagged commands.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
i = 0;
maxtarget = 16;
if (target != CAM_TARGET_WILDCARD) {
i = target;
maxtarget = target + 1;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
for (;i < maxtarget; i++) {
u_int scbid;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
scbid = ahc_index_busy_tcl(ahc, BUILD_TCL(i << 4, 0),
/*unbusy*/FALSE);
scbp = &ahc->scb_data->scbarray[scbid];
if (scbid < ahc->scb_data->numscbs
aic7xxx.c: Correct the BUILD_TCL macro. It was placing the target id in the wrong bits. This was only an issue for adapters that do not perform SCB paging (aha-3940AUW for instance). Don't bother inlining ahc_index_busy_tcl. It is never used in a performance critical path and is a bit chunky. Correct ahc_index_busy_tcl to deal with "busy target tables" embedded in the latter half of 64byte SCBs. Don't initialize the busy target table to its empty state until after we have finished extracting configuration information from chip SRAM. In the common case of using 16 bytes of chip SRAM to do untagged target lookups, we were trashing the last 8 targets configuration data. (actually only target 8 because of the bug in the BUILD_TCL macro). Cram the "bus reset delivered" message back under bootverbose. Fix the cleanup of the SCB busy target table when aborting commands. If the lun is wildcarded, we must loop through all possible luns. aic7xxx.h: Only bother supporting 64 luns right now. It doesn't seem like either this driver or any peripherals will be doing information unit transfers (where the lun number is a 32 bit integer) any time soon. aic7xxx.seq: Fix support for the aic7895. We must flush the data FIFO if performing a manual transfer that is not a multiple of 8 bytes. We were doing this quite regularly for embedded cdbs. Manaually flush the fifo on earlier adapters when dealing with embedded cdbs too. We were stuffing the FIFO with 16 bytes instead, but triggering the flush is more efficient and allows us to remove two instructions from the "copy_to_fifo" routine.
2000-07-24 22:27:40 +00:00
&& ahc_match_scb(ahc, scbp, target, channel, lun, tag, role)) {
u_int minlun;
u_int maxlun;
if (lun == CAM_LUN_WILDCARD) {
/*
* Unless we are using an SCB based
* busy targets table, there is only
* one table entry for all luns of
* a target.
*/
minlun = 0;
maxlun = 1;
if ((ahc->flags & AHC_SCB_BTT) != 0)
maxlun = AHC_NUM_LUNS;
} else {
minlun = lun;
maxlun = lun + 1;
}
while (minlun < maxlun) {
ahc_index_busy_tcl(ahc, BUILD_TCL(i << 4,
minlun), /*unbusy*/TRUE);
minlun++;
}
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Go through the disconnected list and remove any entries we
* have queued for completion, 0'ing their control byte too.
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
* We save the active SCB and restore it ourselves, so there
* is no reason for this search to restore it too.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
ahc_search_disc_list(ahc, target, channel, lun, tag,
/*stop_on_first*/FALSE, /*remove*/TRUE,
/*save_state*/FALSE);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Go through the hardware SCB array looking for commands that
* were active but not on any list.
*/
for(i = 0; i < ahc->scb_data->maxhscbs; i++) {
Remove unused variables and functions. In the rejected message handler, it was possible for us to confuse a rejected SDTR or WDTR for a rejected tag.
1998-09-20 23:30:14 +00:00
u_int scbid;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_outb(ahc, SCBPTR, i);
scbid = ahc_inb(ahc, SCB_TAG);
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
scbp = &ahc->scb_data->scbarray[scbid];
if (scbid < ahc->scb_data->numscbs
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
&& ahc_match_scb(ahc, scbp, target, channel, lun, tag, role))
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
ahc_add_curscb_to_free_list(ahc);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Go through the pending CCB list and look for
* commands for this target that are still active.
* These are other tagged commands that were
* disconnected when the reset occured.
*/
{
struct ccb_hdr *ccb_h;
ccb_h = ahc->pending_ccbs.lh_first;
while (ccb_h != NULL) {
scbp = (struct scb *)ccb_h->ccb_scb_ptr;
ccb_h = ccb_h->sim_links.le.le_next;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (ahc_match_scb(ahc, scbp, target, channel,
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
lun, tag, role)) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (ahc_ccb_status(scbp->ccb) == CAM_REQ_INPROG)
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
ahcsetccbstatus(scbp->ccb, status);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_freeze_ccb(scbp->ccb);
ahc_done(ahc, scbp);
found++;
}
}
}
ahc_outb(ahc, SCBPTR, active_scb);
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahc_release_untagged_queues(ahc);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
return found;
}
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
static int
ahc_search_disc_list(struct ahc_softc *ahc, int target, char channel,
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
int lun, u_int tag, int stop_on_first, int remove,
int save_state)
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
{
struct scb *scbp;
u_int next;
u_int prev;
u_int count;
u_int active_scb;
count = 0;
next = ahc_inb(ahc, DISCONNECTED_SCBH);
prev = SCB_LIST_NULL;
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
if (save_state) {
/* restore this when we're done */
active_scb = ahc_inb(ahc, SCBPTR);
} else
/* Silence compiler */
active_scb = SCB_LIST_NULL;
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
while (next != SCB_LIST_NULL) {
u_int scb_index;
ahc_outb(ahc, SCBPTR, next);
scb_index = ahc_inb(ahc, SCB_TAG);
if (scb_index >= ahc->scb_data->numscbs) {
panic("Disconnected List inconsistency. "
"SCB index == %d, yet numscbs == %d.",
scb_index, ahc->scb_data->numscbs);
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (next == prev) {
panic("Disconnected List Loop. "
"cur SCBPTR == %x, prev SCBPTR == %x.",
next, prev);
}
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
scbp = &ahc->scb_data->scbarray[scb_index];
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (ahc_match_scb(ahc, scbp, target, channel, lun,
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
tag, ROLE_INITIATOR)) {
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
count++;
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
if (remove) {
next =
ahc_rem_scb_from_disc_list(ahc, prev, next);
} else {
prev = next;
next = ahc_inb(ahc, SCB_NEXT);
}
if (stop_on_first)
break;
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
} else {
prev = next;
next = ahc_inb(ahc, SCB_NEXT);
}
}
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
if (save_state)
ahc_outb(ahc, SCBPTR, active_scb);
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
return (count);
}
static u_int
ahc_rem_scb_from_disc_list(struct ahc_softc *ahc, u_int prev, u_int scbptr)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
u_int next;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_outb(ahc, SCBPTR, scbptr);
next = ahc_inb(ahc, SCB_NEXT);
ahc_outb(ahc, SCB_CONTROL, 0);
ahc_add_curscb_to_free_list(ahc);
if (prev != SCB_LIST_NULL) {
ahc_outb(ahc, SCBPTR, prev);
ahc_outb(ahc, SCB_NEXT, next);
} else
ahc_outb(ahc, DISCONNECTED_SCBH, next);
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
return (next);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
static void
ahc_add_curscb_to_free_list(struct ahc_softc *ahc)
{
/* Invalidate the tag so that ahc_find_scb doesn't think it's active */
ahc_outb(ahc, SCB_TAG, SCB_LIST_NULL);
ahc_outb(ahc, SCB_NEXT, ahc_inb(ahc, FREE_SCBH));
ahc_outb(ahc, FREE_SCBH, ahc_inb(ahc, SCBPTR));
}
/*
* Manipulate the waiting for selection list and return the
* scb that follows the one that we remove.
*/
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
static u_int
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahc_rem_wscb(struct ahc_softc *ahc, u_int scbpos, u_int prev)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
u_int curscb, next;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Select the SCB we want to abort and
* pull the next pointer out of it.
*/
curscb = ahc_inb(ahc, SCBPTR);
ahc_outb(ahc, SCBPTR, scbpos);
next = ahc_inb(ahc, SCB_NEXT);
/* Clear the necessary fields */
ahc_outb(ahc, SCB_CONTROL, 0);
ahc_add_curscb_to_free_list(ahc);
/* update the waiting list */
if (prev == SCB_LIST_NULL) {
/* First in the list */
ahc_outb(ahc, WAITING_SCBH, next);
/*
* Ensure we aren't attempting to perform
* selection for this entry.
*/
ahc_outb(ahc, SCSISEQ, (ahc_inb(ahc, SCSISEQ) & ~ENSELO));
} else {
/*
* Select the scb that pointed to us
* and update its next pointer.
*/
ahc_outb(ahc, SCBPTR, prev);
ahc_outb(ahc, SCB_NEXT, next);
}
/*
* Point us back at the original scb position.
*/
ahc_outb(ahc, SCBPTR, curscb);
return next;
}
static void
ahc_clear_intstat(struct ahc_softc *ahc)
{
/* Clear any interrupt conditions this may have caused */
ahc_outb(ahc, CLRSINT0, CLRSELDO|CLRSELDI|CLRSELINGO);
ahc_outb(ahc, CLRSINT1, CLRSELTIMEO|CLRATNO|CLRSCSIRSTI
|CLRBUSFREE|CLRSCSIPERR|CLRPHASECHG|
CLRREQINIT);
ahc_outb(ahc, CLRINT, CLRSCSIINT);
}
static void
ahc_reset_current_bus(struct ahc_softc *ahc)
{
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
uint8_t scsiseq;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_outb(ahc, SIMODE1, ahc_inb(ahc, SIMODE1) & ~ENSCSIRST);
scsiseq = ahc_inb(ahc, SCSISEQ);
ahc_outb(ahc, SCSISEQ, scsiseq | SCSIRSTO);
DELAY(AHC_BUSRESET_DELAY);
/* Turn off the bus reset */
ahc_outb(ahc, SCSISEQ, scsiseq & ~SCSIRSTO);
ahc_clear_intstat(ahc);
/* Re-enable reset interrupts */
ahc_outb(ahc, SIMODE1, ahc_inb(ahc, SIMODE1) | ENSCSIRST);
}
static int
ahc_reset_channel(struct ahc_softc *ahc, char channel, int initiate_reset)
{
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
struct cam_path *path;
u_int initiator, target, max_scsiid;
u_int sblkctl;
u_int our_id;
int found;
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
int restart_needed;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
char cur_channel;
ahc->pending_device = NULL;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
pause_sequencer(ahc);
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
/*
* Run our command complete fifos to ensure that we perform
* completion processing on any commands that 'completed'
* before the reset occurred.
*/
ahc_run_qoutfifo(ahc);
if ((ahc->flags & AHC_TARGETMODE) != 0) {
Update copyrights to Y2K. 93cx6.c: Make the SRAM dump output a little prettier. aic7xxx.c: Store all SG entries into our SG array in kernel space. This makes data-overrun and other error reporting more useful as we can dump all SG entries. In the past, we only stored the SG entries that the sequencer might need to access, which meant we skipped the first element that is embedded into the SCB. Add a table of chip strings and replace ugly switch statements with table lookups. Add a table with bus phase strings and message reponses to parity errors in those phases. Use the table to pretty print bus phase messages as well as collapse another switch statement. Fix a bug in target mode that could cause us to unpause the sequencer early in bus reset processing. Add the 80MHz/DT mode into our syncrate table. This rate is not yet used or enabled. Correct some comments, clean up some code... aic7xxx.h: Add U160 controller feature information. Add some more bit fields for various SEEPROM formats. aic7xxx.reg: Add U160 register and register bit definitions. aic7xxx.seq: Make phasemis state tracking more straight forward. This avoids the consumption of SINDEX which is a very useful register. For the U160 chips, you must use the 'mov' instruction to update DFCNTRL. Using 'or' to set the PRELOADED bit is completely ineffective. At the end of the command phase, wair for our ACK signal to de-assert before disabling the SCSI dma engine. For slow devices, this avoids clearing the ACK before the other end has had a chance to see it and lower REQ.
2000-01-07 23:08:20 +00:00
ahc_run_tqinfifo(ahc, /*paused*/TRUE);
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
}
/*
* Reset the bus if we are initiating this reset
*/
sblkctl = ahc_inb(ahc, SBLKCTL);
cur_channel = 'A';
if ((ahc->features & AHC_TWIN) != 0
&& ((sblkctl & SELBUSB) != 0))
cur_channel = 'B';
if (cur_channel != channel) {
/* Case 1: Command for another bus is active
* Stealthily reset the other bus without
* upsetting the current bus.
*/
ahc_outb(ahc, SBLKCTL, sblkctl ^ SELBUSB);
ahc_outb(ahc, SIMODE1, ahc_inb(ahc, SIMODE1) & ~ENBUSFREE);
ahc_outb(ahc, SCSISEQ,
ahc_inb(ahc, SCSISEQ) & (ENSELI|ENRSELI|ENAUTOATNP));
if (initiate_reset)
ahc_reset_current_bus(ahc);
ahc_clear_intstat(ahc);
ahc_outb(ahc, SBLKCTL, sblkctl);
restart_needed = FALSE;
} else {
/* Case 2: A command from this bus is active or we're idle */
ahc_clear_msg_state(ahc);
ahc_outb(ahc, SIMODE1, ahc_inb(ahc, SIMODE1) & ~ENBUSFREE);
ahc_outb(ahc, SCSISEQ,
ahc_inb(ahc, SCSISEQ) & (ENSELI|ENRSELI|ENAUTOATNP));
if (initiate_reset)
ahc_reset_current_bus(ahc);
ahc_clear_intstat(ahc);
/*
* Since we are going to restart the sequencer, avoid
* a race in the sequencer that could cause corruption
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
* of our Q pointers by starting over from index 1.
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
*/
ahc->qoutfifonext = 0;
if ((ahc->features & AHC_QUEUE_REGS) != 0)
ahc_outb(ahc, SDSCB_QOFF, 0);
else
ahc_outb(ahc, QOUTPOS, 0);
if ((ahc->flags & AHC_TARGETMODE) != 0) {
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
ahc->tqinfifonext = 1;
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
ahc_outb(ahc, KERNEL_TQINPOS, ahc->tqinfifonext - 1);
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
ahc_outb(ahc, TQINPOS, ahc->tqinfifonext);
if ((ahc->features & AHC_HS_MAILBOX) != 0) {
u_int hs_mailbox;
hs_mailbox = ahc_inb(ahc, HS_MAILBOX);
hs_mailbox &= ~HOST_TQINPOS;
ahc_outb(ahc, HS_MAILBOX, hs_mailbox);
}
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
}
restart_needed = TRUE;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Clean up all the state information for the
* pending transactions on this bus.
*/
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
found = ahc_abort_scbs(ahc, CAM_TARGET_WILDCARD, channel,
CAM_LUN_WILDCARD, SCB_LIST_NULL,
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
ROLE_UNKNOWN, CAM_SCSI_BUS_RESET);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
if (channel == 'B') {
path = ahc->path_b;
our_id = ahc->our_id_b;
} else {
path = ahc->path;
our_id = ahc->our_id;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
max_scsiid = (ahc->features & AHC_WIDE) ? 15 : 7;
/*
* Send an immediate notify ccb to all target more peripheral
* drivers affected by this action.
*/
for (target = 0; target <= max_scsiid; target++) {
struct tmode_tstate* tstate;
u_int lun;
tstate = ahc->enabled_targets[target];
if (tstate == NULL)
continue;
for (lun = 0; lun <= 7; lun++) {
struct tmode_lstate* lstate;
lstate = tstate->enabled_luns[lun];
if (lstate == NULL)
continue;
ahc_queue_lstate_event(ahc, lstate, CAM_TARGET_WILDCARD,
EVENT_TYPE_BUS_RESET, /*arg*/0);
ahc_send_lstate_events(ahc, lstate);
}
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/* Notify the XPT that a bus reset occurred */
xpt_async(AC_BUS_RESET, path, NULL);
/*
* Revert to async/narrow transfers until we renegotiate.
*/
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
for (target = 0; target <= max_scsiid; target++) {
if (ahc->enabled_targets[target] == NULL)
continue;
for (initiator = 0; initiator <= max_scsiid; initiator++) {
struct ahc_devinfo devinfo;
ahc_compile_devinfo(&devinfo, target, initiator,
CAM_LUN_WILDCARD,
channel, ROLE_UNKNOWN);
ahc_set_width(ahc, &devinfo, path,
MSG_EXT_WDTR_BUS_8_BIT,
Correct a bug where the chip could be unpaused in the middle of a bus or device reset error recovery operation.
1999-05-18 03:58:49 +00:00
AHC_TRANS_CUR, /*paused*/TRUE);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
ahc_set_syncrate(ahc, &devinfo, path,
/*syncrate*/NULL, /*period*/0,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
/*offset*/0, /*ppr_options*/0,
AHC_TRANS_CUR, /*paused*/TRUE);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
if (restart_needed)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
restart_sequencer(ahc);
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
else
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
unpause_sequencer(ahc);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
return found;
}
static int
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahc_match_scb(struct ahc_softc *ahc, struct scb *scb, int target,
char channel, int lun, u_int tag, role_t role)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
int targ = SCB_GET_TARGET(ahc, scb);
char chan = SCB_GET_CHANNEL(ahc, scb);
int slun = SCB_GET_LUN(scb);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
int match;
match = ((chan == channel) || (channel == ALL_CHANNELS));
if (match != 0)
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
match = ((targ == target) || (target == CAM_TARGET_WILDCARD));
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
if (match != 0)
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
match = ((lun == slun) || (lun == CAM_LUN_WILDCARD));
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
if (match != 0) {
int group;
group = XPT_FC_GROUP(scb->ccb->ccb_h.func_code);
if (role == ROLE_INITIATOR) {
match = (group == XPT_FC_GROUP_COMMON)
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
&& ((tag == scb->hscb->tag)
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
|| (tag == SCB_LIST_NULL));
} else if (role == ROLE_TARGET) {
match = (group == XPT_FC_GROUP_TMODE)
&& ((tag == scb->ccb->csio.tag_id)
|| (tag == SCB_LIST_NULL));
}
}
Kill the "unpause_always" argument to unpause_sequencer(). The reasons for optimizing the unpause operation no-longer exist, and this is much safer. When restarting the sequencer, reconstitute the free SCB list on the card. This deals with a single instruction gap between marking the SCB as free and actually getting it onto the free list. Reduce the number of transfer negotiations that occur. In the past, we renegotiated after every reported check condition status. This ensures that we catch devices that have unexpectidly reset. In this situation, the target will always report the check condition before performing a data-phase. The new behavior is to renegotiate for any check-condition where the residual matches the orginal data-length of the command (including 0 length transffers). This avoids renegotiations during things like variable tape block reads, where the check condition is reported only to indicate the residual of the read. Revamp the parity error detection logic. We now properly report and handle injected parity errors in all phases. The old code used to hang on message-in parity errors. Correct the reporting of selection timeout errors to the XPT. When a selection timeout occurs, only the currently selecting command is flagged with SELTO status instead of aborting all currently active commands to that target. Fix flipped arguments in ahc_match_scb and in some of the callers of this routine. I wish that gcc allowed you to request warnings for enums passed as ints. Make ahc_find_msg generically handle all message types. Work around the target mode data-in wideodd bug in all non-U2 chips. We can now do sync-wide target mode transfers in target mode across the hole product line. Use lastphase exclusively for handling timeouts. The current phase doesn't take the bus free state into account. Fix a bug in the timeout handler that could cause corruption of the disconnected list. When sending an embedded cdb to a target, ensure that we start on a quad word boundary in the data-fifo. It seems that unaligned stores do not work correctly.
2000-02-09 21:25:00 +00:00
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
return match;
}
static void
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
ahc_construct_sdtr(struct ahc_softc *ahc, u_int period, u_int offset)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc->msgout_buf[ahc->msgout_index++] = MSG_EXTENDED;
ahc->msgout_buf[ahc->msgout_index++] = MSG_EXT_SDTR_LEN;
ahc->msgout_buf[ahc->msgout_index++] = MSG_EXT_SDTR;
ahc->msgout_buf[ahc->msgout_index++] = period;
ahc->msgout_buf[ahc->msgout_index++] = offset;
ahc->msgout_len += 5;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
static void
Correct a bug in the timeout handler. When queuing a BDR response to a timeout, we must remove the pending SCB from the disconnected list or risk list corruption when our BDR request using the same SCB is placed on the waiting list. Eradicate some silly uses of u_int8_t that just serve to slow the code down.
1998-09-20 05:06:10 +00:00
ahc_construct_wdtr(struct ahc_softc *ahc, u_int bus_width)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ahc->msgout_buf[ahc->msgout_index++] = MSG_EXTENDED;
ahc->msgout_buf[ahc->msgout_index++] = MSG_EXT_WDTR_LEN;
ahc->msgout_buf[ahc->msgout_index++] = MSG_EXT_WDTR;
ahc->msgout_buf[ahc->msgout_index++] = bus_width;
ahc->msgout_len += 4;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
static void
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahc_construct_ppr(struct ahc_softc *ahc, u_int period, u_int offset,
u_int bus_width, u_int ppr_options)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
ahc->msgout_buf[ahc->msgout_index++] = MSG_EXTENDED;
ahc->msgout_buf[ahc->msgout_index++] = MSG_EXT_PPR_LEN;
ahc->msgout_buf[ahc->msgout_index++] = MSG_EXT_PPR;
ahc->msgout_buf[ahc->msgout_index++] = period;
ahc->msgout_buf[ahc->msgout_index++] = 0;
ahc->msgout_buf[ahc->msgout_index++] = offset;
ahc->msgout_buf[ahc->msgout_index++] = bus_width;
ahc->msgout_buf[ahc->msgout_index++] = ppr_options;
ahc->msgout_len += 8;
}
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
static void
ahc_calc_residual(struct scb *scb)
{
struct hardware_scb *hscb;
struct status_pkt *spkt;
uint32_t resid;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
Properly handle the case where the residual is 0, but, as the target didn't bother to send a saved data pointers after the last transfer, is not recorded in sgptr. This was only a problem if the target reported non-zero status as we always check the residual in that case.
2000-07-25 20:40:34 +00:00
* 5 cases.
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
* 1) No residual.
* SG_RESID_VALID clear in sgptr.
* 2) Transferless command
* 3) Never performed any transfers.
* sgptr has SG_FULL_RESID set.
Properly handle the case where the residual is 0, but, as the target didn't bother to send a saved data pointers after the last transfer, is not recorded in sgptr. This was only a problem if the target reported non-zero status as we always check the residual in that case.
2000-07-25 20:40:34 +00:00
* 4) No residual but target did not
* save data pointers after the
* last transfer, so sgptr was
* never updated.
* 5) We have a partial residual.
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
* Use residual_sgptr to determine
* where we are.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
hscb = scb->hscb;
if ((hscb->sgptr & SG_RESID_VALID) == 0)
Properly handle the case where the residual is 0, but, as the target didn't bother to send a saved data pointers after the last transfer, is not recorded in sgptr. This was only a problem if the target reported non-zero status as we always check the residual in that case.
2000-07-25 20:40:34 +00:00
/* Case 1 */
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
return;
hscb->sgptr &= ~SG_RESID_VALID;
if ((hscb->sgptr & SG_LIST_NULL) != 0)
Properly handle the case where the residual is 0, but, as the target didn't bother to send a saved data pointers after the last transfer, is not recorded in sgptr. This was only a problem if the target reported non-zero status as we always check the residual in that case.
2000-07-25 20:40:34 +00:00
/* Case 2 */
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
return;
spkt = &hscb->shared_data.status;
if ((hscb->sgptr & SG_FULL_RESID) != 0)
Properly handle the case where the residual is 0, but, as the target didn't bother to send a saved data pointers after the last transfer, is not recorded in sgptr. This was only a problem if the target reported non-zero status as we always check the residual in that case.
2000-07-25 20:40:34 +00:00
/* Case 3 */
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
resid = scb->ccb->csio.dxfer_len;
Properly handle the case where the residual is 0, but, as the target didn't bother to send a saved data pointers after the last transfer, is not recorded in sgptr. This was only a problem if the target reported non-zero status as we always check the residual in that case.
2000-07-25 20:40:34 +00:00
else if ((spkt->residual_sg_ptr & SG_LIST_NULL) != 0)
/* Case 4 */
return;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
else if ((spkt->residual_sg_ptr & ~SG_PTR_MASK) != 0)
panic("Bogus resid sgptr value 0x%x\n", spkt->residual_sg_ptr);
else {
struct ahc_dma_seg *sg;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Remainder of the SG where the transfer
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
* stopped.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
resid = spkt->residual_datacnt & AHC_SG_LEN_MASK;
sg = ahc_sg_bus_to_virt(scb,
spkt->residual_sg_ptr & SG_PTR_MASK);
/* The residual sg_ptr always points to the next sg */
sg--;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/*
* Add up the contents of all residual
* SG segments that are after the SG where
* the transfer stopped.
*/
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
while ((sg->len & AHC_DMA_LAST_SEG) == 0) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
sg++;
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
resid += sg->len & AHC_SG_LEN_MASK;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
}
if ((scb->flags & SCB_SENSE) == 0) {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
scb->ccb->csio.resid = resid;
} else {
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
scb->ccb->csio.sense_resid = resid;
}
o Correct the offsets into the syncrate table for paritcular negotiation features (DT, ULTRA2, ULTRA, FAST). The offsets where not properly updated when the DT entry was added and so the driver could attempt to negotiate a speed faster than that supported by the target device or even requested by the user via SCSI-Select settings. * o Update the target mode incoming command queue kernel index value ever 128 commands instead of 32. This means that the kernel will always try to keep its index (as seen on the card - the kernel may actually have cleared more space) 128 commands ahead of where the sequencer is adding entries. o Use the HS_MAILBOX register instead of the KERNEL_TQINPOS location in SRAM to indicate the kernel's target queue possition on Ultra2 cards. This avoids the "pause bug" on these cards and also turns out to be much more efficient. o When enabling or disabling a particular target id for target mode, make sure that the taret id in the SCSIID register does not reference an ID that is not to receive target selections. This is only an issue on chips that support the multiple target id feature where the value in SCSIID will still affect selection behavior regardless of the values in the target id bit field registers. o Remove some target mode debugging printfs. o Make sure that the sense length reported in ATIO commands is always zero. This driver does not, yet, report HBA generated sense information for accepted commands. o Honor the CAM_TIME_INFINITY and CAM_TIME_DEFAULT values for the CCB timeout field. o Make the driver compile with AHC_DEBUG again. * Noticed by: Andrew Gallatin<gallatin@cs.duke.edu>
2000-03-18 22:28:20 +00:00
#ifdef AHC_DEBUG
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
if (ahc_debug & AHC_SHOWMISC) {
xpt_print_path(scb->ccb->ccb_h.path);
printf("Handled Residual of %d bytes\n", resid);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
#endif
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
}
static void
ahc_update_pending_syncrates(struct ahc_softc *ahc)
{
struct ccb_hdr *ccbh;
int pending_ccb_count;
int i;
u_int saved_scbptr;
/*
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
* Traverse the pending SCB list and ensure that all of the
* SCBs there have the proper settings.
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
*/
ccbh = LIST_FIRST(&ahc->pending_ccbs);
pending_ccb_count = 0;
while (ccbh != NULL) {
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
struct ahc_devinfo devinfo;
union ccb *ccb;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
struct scb *pending_scb;
struct hardware_scb *pending_hscb;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
struct ahc_initiator_tinfo *tinfo;
struct tmode_tstate *tstate;
u_int our_id, remote_id;
ccb = (union ccb*)ccbh;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
pending_scb = (struct scb *)ccbh->ccb_scb_ptr;
pending_hscb = pending_scb->hscb;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
if (ccbh->func_code == XPT_CONT_TARGET_IO) {
our_id = ccb->ccb_h.target_id;
remote_id = ccb->ctio.init_id;
} else {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
our_id = SCB_IS_SCSIBUS_B(ahc, pending_scb)
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
? ahc->our_id_b : ahc->our_id;
remote_id = ccb->ccb_h.target_id;
}
ahc_compile_devinfo(&devinfo, our_id, remote_id,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
SCB_GET_LUN(pending_scb),
SCB_GET_CHANNEL(ahc, pending_scb),
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ROLE_UNKNOWN);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
tinfo = ahc_fetch_transinfo(ahc, devinfo.channel,
our_id, remote_id, &tstate);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
pending_hscb->control &= ~ULTRAENB;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
if ((tstate->ultraenb & devinfo.target_mask) != 0)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
pending_hscb->control |= ULTRAENB;
pending_hscb->scsirate = tinfo->scsirate;
pending_hscb->scsioffset = tinfo->current.offset;
pending_ccb_count++;
ccbh = LIST_NEXT(ccbh, sim_links.le);
}
if (pending_ccb_count == 0)
return;
saved_scbptr = ahc_inb(ahc, SCBPTR);
/* Ensure that the hscbs down on the card match the new information */
for (i = 0; i < ahc->scb_data->maxhscbs; i++) {
u_int scb_tag;
ahc_outb(ahc, SCBPTR, i);
scb_tag = ahc_inb(ahc, SCB_TAG);
if (scb_tag != SCB_LIST_NULL) {
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
struct ahc_devinfo devinfo;
union ccb *ccb;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
struct scb *pending_scb;
struct hardware_scb *pending_hscb;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
struct ahc_initiator_tinfo *tinfo;
struct tmode_tstate *tstate;
u_int our_id, remote_id;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
u_int control;
Complete conversion to bus dma. This driver now works on the alpha. aicasm_symbol.c: Correct an unaligned access problem. You can't rely on DB to store your data in an aligned fashion.
1999-04-23 23:27:31 +00:00
pending_scb = &ahc->scb_data->scbarray[scb_tag];
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
if (pending_scb->flags == SCB_FREE)
continue;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
pending_hscb = pending_scb->hscb;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
ccb = pending_scb->ccb;
if (ccb->ccb_h.func_code == XPT_CONT_TARGET_IO) {
our_id = ccb->ccb_h.target_id;
remote_id = ccb->ctio.init_id;
} else {
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
our_id = SCB_IS_SCSIBUS_B(ahc, pending_scb)
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
? ahc->our_id_b : ahc->our_id;
remote_id = ccb->ccb_h.target_id;
}
ahc_compile_devinfo(&devinfo, our_id, remote_id,
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
SCB_GET_LUN(pending_scb),
SCB_GET_CHANNEL(ahc, pending_scb),
Revamp the way that exceptional message handling is performed so that it is more robust and common code can be used for both the target and iniator roles. The mechanism for tracking negotiation state has also been simplified. Add support for sync/wide negotiation in target mode and fix many of the target mode bugs running at higher speeds uncovered. Make a first stab at getting all of the bus skew delays correct. Sync+Wide dataout transfers still cause problems, but this may be an initiator problem. Ensure that we exit BITBUCKET mode if the controller is restarted. Add support for target mode only firmware downloads. This has been tested on the aic7880, but should mean that we can perform target mode on any aic7xxx controller. Mixed mode (initiator and target roles in the same firmware load) is currently only supported on the aic7890, but with optimization, may fit on chips with less instruction space.
1998-12-10 04:14:50 +00:00
ROLE_UNKNOWN);
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
tinfo = ahc_fetch_transinfo(ahc, devinfo.channel,
our_id, remote_id, &tstate);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
control = ahc_inb(ahc, SCB_CONTROL);
control &= ~ULTRAENB;
Keep track of negotiated transfer parameters for each initiator<->target connection. Clean up support for devices featuring the multiple target SCSI ID feature. On aic7890/91/96/97 chips, we can now assume the target role on multiple target ids simultaneously. Although these chips also have sufficient instruction space to hold to support the initiator and target role at the same time, the initiator role is currently disabled as it will conflict (chip design restriction) with the multi-tid feature. I'll probably add a nob to enable the initiator (there-by disabling multi-tid) some time in the future. Return queue full or busy, depending on the tagged nature of the incoming request, if our command input queue fills up in host memeory. Deal with accept target I/O resource shortages. If we get an underrun on a transaction that wasn't supposed to transmit any data, don't attempt to print out the S/G list. The code would run until hitting a non-present page. (oops)
1999-03-05 23:35:48 +00:00
if ((tstate->ultraenb & devinfo.target_mask) != 0)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
control |= ULTRAENB;
ahc_outb(ahc, SCB_CONTROL, control);
ahc_outb(ahc, SCB_SCSIRATE, tinfo->scsirate);
ahc_outb(ahc, SCB_SCSIOFFSET, tinfo->current.offset);
}
}
ahc_outb(ahc, SCBPTR, saved_scbptr);
}
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
#if UNUSED
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static void
ahc_dump_targcmd(struct target_cmd *cmd)
{
o Convert to <inttypes.h> style fixed sized types to facilitate porting to other systems. o Normalize copyright text. o Clean up probe code function interfaces by passing around a single structure of common arguments instead of passing "too many" args in each function call. o Add support for the AAA-131 as a SCSI adapter. o Add support for the AHA-4944 courtesy of "Matthew N. Dodd" <winter@jurai.net o Correct manual termination support for PCI cards. The bit definitions for manual termination control in the SEEPROM were incorrect. o Add support for extracting NVRAM information from SCB 2 for BIOSen that use this mechanism to pass this data to OS drivers. o Properly set the STPWLEVEL bit in PCI config space based on the setting in an SEEPROM. o Go back to useing 32byte SCBs for all controllers. The current firmware allows us to embed 12byte cdbs on all controllers in a 32byte SCB, and larger cdbs are rarely used, so it is a better use of this space to offer more SCBs (32). o Add support for U160 transfers. o Add an idle loop executed during data transfers that prefetches S/G segments on controllers that have a secondary DMA engine (aic789X). o Improve the performance of reselections by avoiding an extra one byte DMA in the case of an SCB lookup miss for the reselecting target. We now keep a 16byte "untagged target" array on the card for dealing with untagged reselections. If the controller has external SCB ram and can support 64byte SCBs, then we use an "untagged target/lun" array to maximize concurrency. Without external SCB ram, the controller is limited to one untagged transaction per target, auto-request sense operations excluded. o Correct the setup of the STPWEN bit in SXFRCTL1. This control line is tri-stated until set to one, so set it to one and then set it to the desired value. o Add tagged queuing support to our target role implementation. o Handle the common cases of the ignore wide residue message in firmware. o Add preliminary support for 39bit addressing. o Add support for assembling on big-endian machines. Big-endian support is not complete in the driver. o Correctly remove SCBs in the waiting for selection queue when freezing a device queue. o Now that we understand more about the autoflush bug on the aic7890, only use the workaround on devices that need it. o Add a workaround for the "aic7890 hangs the system when you attempt to pause it" problem. We can now pause the aic7890 safely regardless of what instruction it is executing.
2000-07-18 20:12:14 +00:00
uint8_t *byte;
uint8_t *last_byte;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
int i;
Change the delivery mechanism for incoming target commands. We now use a 256 entry ring buffer of descriptersfor this purpose. This allows the use of a simple 8bit counter in the sequencer code for tracking start location. Entries in the ring buffer now contain a "cmd_valid" byte at their tail. As an entry is serviced, this byte is cleared by the kernel and set by the sequencer during its dma of a new entry. Since this byte is the last portion of the command touched during a dma, the kernel can use this byte to ensure the command it processes is completely valid. The new command format requires a fixed sized DMA from the controller to deliver which allowed for additional simplification of the sequencer code. The hack that required 1 SCB slot to be stolen for incoming command delivery notification is also gone.
1998-11-23 01:33:47 +00:00
byte = &cmd->initiator_channel;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
/* Debugging info for received commands */
Change the delivery mechanism for incoming target commands. We now use a 256 entry ring buffer of descriptersfor this purpose. This allows the use of a simple 8bit counter in the sequencer code for tracking start location. Entries in the ring buffer now contain a "cmd_valid" byte at their tail. As an entry is serviced, this byte is cleared by the kernel and set by the sequencer during its dma of a new entry. Since this byte is the last portion of the command touched during a dma, the kernel can use this byte to ensure the command it processes is completely valid. The new command format requires a fixed sized DMA from the controller to deliver which allowed for additional simplification of the sequencer code. The hack that required 1 SCB slot to be stolen for incoming command delivery notification is also gone.
1998-11-23 01:33:47 +00:00
last_byte = &cmd[1].initiator_channel;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
i = 0;
while (byte < last_byte) {
if (i == 0)
printf("\t");
printf("%#x", *byte++);
i++;
if (i == 8) {
printf("\n");
i = 0;
} else {
printf(", ");
}
}
}
Fix a few problems with handling rejected transfer negotiation messages. Use the host message loop for any unknown message types instead of performing a reject message in the sequencer. Pass reject messages to the host message loop too which frees up a sequencer interrupt type slot. Default to issuing a bus reset if initiator mode is enabled. It seems that the reset scsi bus bit is not defined in the same location for all aic78xx BIOSes, so attempting to honor this setting will have to wait until I get more information on how to detect it. Nuke some unused variables.
1998-12-17 00:06:52 +00:00
#endif
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
static void
Implement a new generic mechanism for attaching handler functions to events, in order to pave the way for removing a number of the ad-hoc implementations currently in use. Retire the at_shutdown family of functions and replace them with new event handler lists. Rework kern_shutdown.c to take greater advantage of the use of event handlers. Reviewed by: green
1999-08-21 06:24:40 +00:00
ahc_shutdown(void *arg, int howto)
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
{
struct ahc_softc *ahc;
int i;
Preserve termination settings across the card reset in our shutdown hook.
1999-04-19 21:28:15 +00:00
u_int sxfrctl1_a, sxfrctl1_b;
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc = (struct ahc_softc *)arg;
Preserve termination settings across the card reset in our shutdown hook.
1999-04-19 21:28:15 +00:00
pause_sequencer(ahc);
/*
* Preserve the value of the SXFRCTL1 register for all channels.
* It contains settings that affect termination and we don't want
* to disturb the integrity of the bus during shutdown in case
* we are in a multi-initiator setup.
*/
sxfrctl1_b = 0;
if ((ahc->features & AHC_TWIN) != 0) {
u_int sblkctl;
sblkctl = ahc_inb(ahc, SBLKCTL);
ahc_outb(ahc, SBLKCTL, sblkctl | SELBUSB);
sxfrctl1_b = ahc_inb(ahc, SXFRCTL1);
ahc_outb(ahc, SBLKCTL, sblkctl & ~SELBUSB);
}
sxfrctl1_a = ahc_inb(ahc, SXFRCTL1);
/* This will reset most registers to 0, but not all */
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_reset(ahc);
Preserve termination settings across the card reset in our shutdown hook.
1999-04-19 21:28:15 +00:00
if ((ahc->features & AHC_TWIN) != 0) {
u_int sblkctl;
sblkctl = ahc_inb(ahc, SBLKCTL);
ahc_outb(ahc, SBLKCTL, sblkctl | SELBUSB);
ahc_outb(ahc, SXFRCTL1, sxfrctl1_b);
ahc_outb(ahc, SBLKCTL, sblkctl & ~SELBUSB);
}
ahc_outb(ahc, SXFRCTL1, sxfrctl1_a);
Massive overhaul of the aic7xxx driver: - Convert to CAM - Use a new DMA based queuing and paging scheme - Add preliminary target mode support - Add support for the aic789X chips - Take advantage of external SRAM on more controllers. - Numerous bug fixes and performance improvements.
1998-09-15 07:24:17 +00:00
ahc_outb(ahc, SCSISEQ, 0);
ahc_outb(ahc, SXFRCTL0, 0);
ahc_outb(ahc, DSPCISTATUS, 0);
for (i = TARG_SCSIRATE; i < HA_274_BIOSCTRL; i++)
ahc_outb(ahc, i, 0);
}
Add support for issuing immediate notify event ccbs for bus resets, bdr messages, abort messages, and abort tag messages. Fix a bug in how default transfer negotiations are handled if the user had disabled initial bus resets. Support multi-targetid on the aic7895C.
1999-08-16 22:49:29 +00:00
/*
* Add a target mode event to this lun's queue
*/
static void
ahc_queue_lstate_event(struct ahc_softc *ahc, struct tmode_lstate *lstate,
u_int initiator_id, u_int event_type, u_int event_arg)
{
struct ahc_tmode_event *event;
int pending;
xpt_freeze_devq(lstate->path, /*count*/1);
if (lstate->event_w_idx >= lstate->event_r_idx)
pending = lstate->event_w_idx - lstate->event_r_idx;
else
pending = AHC_TMODE_EVENT_BUFFER_SIZE + 1
- (lstate->event_r_idx - lstate->event_w_idx);
if (event_type == EVENT_TYPE_BUS_RESET
|| event_type == MSG_BUS_DEV_RESET) {
/*
* Any earlier events are irrelevant, so reset our buffer.
* This has the effect of allowing us to deal with reset
* floods (an external device holding down the reset line)
* without losing the event that is really interesting.
*/
lstate->event_r_idx = 0;
lstate->event_w_idx = 0;
xpt_release_devq(lstate->path, pending, /*runqueue*/FALSE);
}
if (pending == AHC_TMODE_EVENT_BUFFER_SIZE) {
xpt_print_path(lstate->path);
printf("immediate event %x:%x lost\n",
lstate->event_buffer[lstate->event_r_idx].event_type,
lstate->event_buffer[lstate->event_r_idx].event_arg);
lstate->event_r_idx++;
if (lstate->event_r_idx == AHC_TMODE_EVENT_BUFFER_SIZE)
lstate->event_r_idx = 0;
xpt_release_devq(lstate->path, /*count*/1, /*runqueue*/FALSE);
}
event = &lstate->event_buffer[lstate->event_w_idx];
event->initiator_id = initiator_id;
event->event_type = event_type;
event->event_arg = event_arg;
lstate->event_w_idx++;
if (lstate->event_w_idx == AHC_TMODE_EVENT_BUFFER_SIZE)
lstate->event_w_idx = 0;
}
/*
* Send any target mode events queued up waiting
* for immediate notify resources.
*/
static void
ahc_send_lstate_events(struct ahc_softc *ahc, struct tmode_lstate *lstate)
{
struct ccb_hdr *ccbh;
struct ccb_immed_notify *inot;
while (lstate->event_r_idx != lstate->event_w_idx
&& (ccbh = SLIST_FIRST(&lstate->immed_notifies)) != NULL) {
struct ahc_tmode_event *event;
event = &lstate->event_buffer[lstate->event_r_idx];
SLIST_REMOVE_HEAD(&lstate->immed_notifies, sim_links.sle);
inot = (struct ccb_immed_notify *)ccbh;
switch (event->event_type) {
case EVENT_TYPE_BUS_RESET:
ccbh->status = CAM_SCSI_BUS_RESET|CAM_DEV_QFRZN;
break;
default:
ccbh->status = CAM_MESSAGE_RECV|CAM_DEV_QFRZN;
inot->message_args[0] = event->event_type;
inot->message_args[1] = event->event_arg;
break;
}
inot->initiator_id = event->initiator_id;
inot->sense_len = 0;
xpt_done((union ccb *)inot);
lstate->event_r_idx++;
if (lstate->event_r_idx == AHC_TMODE_EVENT_BUFFER_SIZE)
lstate->event_r_idx = 0;
}
}
Reference in New Issue Copy Permalink