2001-03-06 21:43:46 +00:00
|
|
|
|
/*-
|
2002-01-01 19:58:11 +00:00
|
|
|
|
* Copyright (c) 1998,1999,2000,2001,2002 S<EFBFBD>ren Schmidt <sos@FreeBSD.org>
|
2001-03-06 21:43:46 +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,
|
|
|
|
|
* without modification, immediately at the beginning of the file.
|
|
|
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
|
|
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
|
|
|
* documentation and/or other materials provided with the distribution.
|
|
|
|
|
* 3. The name of the author may not be used to endorse or promote products
|
|
|
|
|
* derived from this software without specific prior written permission.
|
|
|
|
|
*
|
|
|
|
|
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``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 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.
|
|
|
|
|
*
|
|
|
|
|
* $FreeBSD$
|
|
|
|
|
*/
|
|
|
|
|
|
|
|
|
|
#include <sys/param.h>
|
|
|
|
|
#include <sys/systm.h>
|
|
|
|
|
#include <sys/kernel.h>
|
|
|
|
|
#include <sys/disk.h>
|
|
|
|
|
#include <sys/module.h>
|
|
|
|
|
#include <sys/bus.h>
|
|
|
|
|
#include <sys/bio.h>
|
|
|
|
|
#include <sys/malloc.h>
|
|
|
|
|
#include <sys/devicestat.h>
|
|
|
|
|
#include <sys/sysctl.h>
|
|
|
|
|
#include <machine/stdarg.h>
|
|
|
|
|
#include <machine/resource.h>
|
|
|
|
|
#include <machine/bus.h>
|
|
|
|
|
#ifdef __alpha__
|
|
|
|
|
#include <machine/md_var.h>
|
|
|
|
|
#endif
|
|
|
|
|
#include <sys/rman.h>
|
|
|
|
|
#include <pci/pcivar.h>
|
|
|
|
|
#include <pci/pcireg.h>
|
|
|
|
|
#include <dev/ata/ata-all.h>
|
|
|
|
|
|
2001-09-25 17:10:39 +00:00
|
|
|
|
/* device structures */
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
struct ata_pci_controller {
|
2001-03-06 21:43:46 +00:00
|
|
|
|
struct resource *bmio;
|
|
|
|
|
int bmaddr;
|
|
|
|
|
struct resource *irq;
|
|
|
|
|
int irqcnt;
|
|
|
|
|
};
|
|
|
|
|
|
2001-09-25 17:10:39 +00:00
|
|
|
|
/* misc defines */
|
2001-12-25 14:44:26 +00:00
|
|
|
|
#define IOMASK 0xfffffffc
|
|
|
|
|
#define GRANDPARENT(dev) device_get_parent(device_get_parent(dev))
|
|
|
|
|
#define ATA_MASTERDEV(dev) ((pci_get_progif(dev) & 0x80) && \
|
|
|
|
|
(pci_get_progif(dev) & 0x05) != 0x05)
|
2001-09-25 17:10:39 +00:00
|
|
|
|
|
2001-03-06 21:43:46 +00:00
|
|
|
|
int
|
2001-09-25 17:10:39 +00:00
|
|
|
|
ata_find_dev(device_t dev, u_int32_t devid, u_int32_t revid)
|
2001-03-06 21:43:46 +00:00
|
|
|
|
{
|
2001-12-02 10:48:52 +00:00
|
|
|
|
device_t *children;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
int nchildren, i;
|
|
|
|
|
|
|
|
|
|
if (device_get_children(device_get_parent(dev), &children, &nchildren))
|
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
|
|
for (i = 0; i < nchildren; i++) {
|
2001-12-14 21:28:49 +00:00
|
|
|
|
if (pci_get_devid(children[i]) == devid &&
|
2001-12-02 10:48:52 +00:00
|
|
|
|
pci_get_revid(children[i]) >= revid) {
|
2001-03-06 21:43:46 +00:00
|
|
|
|
free(children, M_TEMP);
|
|
|
|
|
return 1;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
free(children, M_TEMP);
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
2001-12-25 14:44:26 +00:00
|
|
|
|
static void
|
2002-01-01 19:58:11 +00:00
|
|
|
|
ata_via_southbridge_fixup(device_t dev)
|
2001-09-25 17:10:39 +00:00
|
|
|
|
{
|
2002-01-01 19:58:11 +00:00
|
|
|
|
device_t *children;
|
2001-09-25 17:10:39 +00:00
|
|
|
|
int nchildren, i;
|
|
|
|
|
|
|
|
|
|
if (device_get_children(device_get_parent(dev), &children, &nchildren))
|
|
|
|
|
return;
|
|
|
|
|
|
|
|
|
|
for (i = 0; i < nchildren; i++) {
|
2002-01-01 19:58:11 +00:00
|
|
|
|
if (pci_get_devid(children[i]) == 0x03051106 || /* VIA VT8363 */
|
|
|
|
|
pci_get_devid(children[i]) == 0x03911106 || /* VIA VT8371 */
|
|
|
|
|
pci_get_devid(children[i]) == 0x31021106 || /* VIA VT8662 */
|
|
|
|
|
pci_get_devid(children[i]) == 0x31121106) { /* VIA VT8361 */
|
|
|
|
|
u_int8_t reg76 = pci_read_config(children[i], 0x76, 1);
|
|
|
|
|
|
|
|
|
|
if ((reg76 & 0xf0) != 0xd0) {
|
|
|
|
|
device_printf(dev,
|
|
|
|
|
"Correcting VIA config for southbridge data corruption bug\n");
|
|
|
|
|
pci_write_config(children[i], 0x75, 0x80, 1);
|
|
|
|
|
pci_write_config(children[i], 0x76, (reg76 & 0x0f) | 0xd0, 1);
|
|
|
|
|
}
|
2001-09-25 17:10:39 +00:00
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
free(children, M_TEMP);
|
|
|
|
|
}
|
|
|
|
|
|
2001-03-06 21:43:46 +00:00
|
|
|
|
static const char *
|
|
|
|
|
ata_pci_match(device_t dev)
|
|
|
|
|
{
|
|
|
|
|
if (pci_get_class(dev) != PCIC_STORAGE)
|
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
|
|
switch (pci_get_devid(dev)) {
|
|
|
|
|
/* supported chipsets */
|
|
|
|
|
case 0x12308086:
|
|
|
|
|
return "Intel PIIX ATA controller";
|
|
|
|
|
|
|
|
|
|
case 0x70108086:
|
|
|
|
|
return "Intel PIIX3 ATA controller";
|
|
|
|
|
|
|
|
|
|
case 0x71118086:
|
|
|
|
|
case 0x71998086:
|
2001-10-23 09:57:49 +00:00
|
|
|
|
case 0x84ca8086:
|
2001-03-06 21:43:46 +00:00
|
|
|
|
return "Intel PIIX4 ATA33 controller";
|
|
|
|
|
|
|
|
|
|
case 0x24218086:
|
|
|
|
|
return "Intel ICH0 ATA33 controller";
|
|
|
|
|
|
|
|
|
|
case 0x24118086:
|
2001-10-23 09:57:49 +00:00
|
|
|
|
case 0x76018086:
|
2001-03-06 21:43:46 +00:00
|
|
|
|
return "Intel ICH ATA66 controller";
|
|
|
|
|
|
2001-04-16 21:22:34 +00:00
|
|
|
|
case 0x244a8086:
|
2001-03-06 21:43:46 +00:00
|
|
|
|
case 0x244b8086:
|
|
|
|
|
return "Intel ICH2 ATA100 controller";
|
|
|
|
|
|
2001-10-23 09:57:49 +00:00
|
|
|
|
case 0x248a8086:
|
|
|
|
|
case 0x248b8086:
|
|
|
|
|
return "Intel ICH3 ATA100 controller";
|
|
|
|
|
|
2001-03-06 21:43:46 +00:00
|
|
|
|
case 0x522910b9:
|
2001-12-14 21:28:49 +00:00
|
|
|
|
if (pci_get_revid(dev) >= 0xc4)
|
|
|
|
|
return "AcerLabs Aladdin ATA100 controller";
|
|
|
|
|
else if (pci_get_revid(dev) >= 0xc2)
|
|
|
|
|
return "AcerLabs Aladdin ATA66 controller";
|
|
|
|
|
else if (pci_get_revid(dev) >= 0x20)
|
2001-03-06 21:43:46 +00:00
|
|
|
|
return "AcerLabs Aladdin ATA33 controller";
|
2001-12-14 21:28:49 +00:00
|
|
|
|
else
|
|
|
|
|
return "AcerLabs Aladdin ATA controller";
|
2001-03-06 21:43:46 +00:00
|
|
|
|
|
|
|
|
|
case 0x05711106:
|
|
|
|
|
if (ata_find_dev(dev, 0x05861106, 0x02))
|
|
|
|
|
return "VIA 82C586 ATA33 controller";
|
|
|
|
|
if (ata_find_dev(dev, 0x05861106, 0))
|
|
|
|
|
return "VIA 82C586 ATA controller";
|
|
|
|
|
if (ata_find_dev(dev, 0x05961106, 0x12))
|
|
|
|
|
return "VIA 82C596 ATA66 controller";
|
|
|
|
|
if (ata_find_dev(dev, 0x05961106, 0))
|
|
|
|
|
return "VIA 82C596 ATA33 controller";
|
2002-01-28 13:17:10 +00:00
|
|
|
|
if (ata_find_dev(dev, 0x06861106, 0x40))
|
2001-03-06 21:43:46 +00:00
|
|
|
|
return "VIA 82C686 ATA100 controller";
|
2002-01-28 13:17:10 +00:00
|
|
|
|
if (ata_find_dev(dev, 0x06861106, 0x10))
|
2001-03-06 21:43:46 +00:00
|
|
|
|
return "VIA 82C686 ATA66 controller";
|
2002-01-28 13:17:10 +00:00
|
|
|
|
if (ata_find_dev(dev, 0x06861106, 0))
|
|
|
|
|
return "VIA 82C686 ATA33 controller";
|
|
|
|
|
if (ata_find_dev(dev, 0x82311106, 0))
|
|
|
|
|
return "VIA 8231 ATA100 controller";
|
|
|
|
|
if (ata_find_dev(dev, 0x30741106, 0) ||
|
|
|
|
|
ata_find_dev(dev, 0x31091106, 0))
|
|
|
|
|
return "VIA 8233 ATA100 controller";
|
|
|
|
|
if (ata_find_dev(dev, 0x31471106, 0))
|
|
|
|
|
return "VIA 8233 ATA133 controller";
|
2001-03-06 21:43:46 +00:00
|
|
|
|
return "VIA Apollo ATA controller";
|
|
|
|
|
|
|
|
|
|
case 0x55131039:
|
2001-12-02 10:48:52 +00:00
|
|
|
|
if (ata_find_dev(dev, 0x06301039, 0x30) ||
|
|
|
|
|
ata_find_dev(dev, 0x06331039, 0x00) ||
|
|
|
|
|
ata_find_dev(dev, 0x06351039, 0x00) ||
|
2001-12-25 14:44:26 +00:00
|
|
|
|
ata_find_dev(dev, 0x06401039, 0x00) ||
|
2001-12-19 01:53:11 +00:00
|
|
|
|
ata_find_dev(dev, 0x06451039, 0x00) ||
|
2001-12-25 14:44:26 +00:00
|
|
|
|
ata_find_dev(dev, 0x06501039, 0x00) ||
|
2001-12-02 10:48:52 +00:00
|
|
|
|
ata_find_dev(dev, 0x07301039, 0x00) ||
|
|
|
|
|
ata_find_dev(dev, 0x07331039, 0x00) ||
|
2001-12-25 14:44:26 +00:00
|
|
|
|
ata_find_dev(dev, 0x07351039, 0x00) ||
|
|
|
|
|
ata_find_dev(dev, 0x07401039, 0x00) ||
|
|
|
|
|
ata_find_dev(dev, 0x07451039, 0x00) ||
|
|
|
|
|
ata_find_dev(dev, 0x07501039, 0x00))
|
2001-12-02 10:48:52 +00:00
|
|
|
|
return "SiS 5591 ATA100 controller";
|
2001-12-25 14:44:26 +00:00
|
|
|
|
else if (ata_find_dev(dev, 0x05301039, 0x00) ||
|
2001-12-02 10:48:52 +00:00
|
|
|
|
ata_find_dev(dev, 0x05401039, 0x00) ||
|
|
|
|
|
ata_find_dev(dev, 0x06201039, 0x00) ||
|
|
|
|
|
ata_find_dev(dev, 0x06301039, 0x00))
|
|
|
|
|
return "SiS 5591 ATA66 controller";
|
2001-12-25 14:44:26 +00:00
|
|
|
|
else
|
|
|
|
|
return "SiS 5591 ATA33 controller";
|
2001-03-06 21:43:46 +00:00
|
|
|
|
|
|
|
|
|
case 0x06491095:
|
|
|
|
|
return "CMD 649 ATA100 controller";
|
|
|
|
|
|
|
|
|
|
case 0x06481095:
|
|
|
|
|
return "CMD 648 ATA66 controller";
|
|
|
|
|
|
|
|
|
|
case 0x06461095:
|
|
|
|
|
return "CMD 646 ATA controller";
|
|
|
|
|
|
|
|
|
|
case 0xc6931080:
|
|
|
|
|
if (pci_get_subclass(dev) == PCIS_STORAGE_IDE)
|
|
|
|
|
return "Cypress 82C693 ATA controller";
|
2001-12-25 14:44:26 +00:00
|
|
|
|
return NULL;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
|
|
|
|
|
case 0x01021078:
|
|
|
|
|
return "Cyrix 5530 ATA33 controller";
|
|
|
|
|
|
|
|
|
|
case 0x74091022:
|
|
|
|
|
return "AMD 756 ATA66 controller";
|
|
|
|
|
|
2001-05-14 18:38:22 +00:00
|
|
|
|
case 0x74111022:
|
|
|
|
|
return "AMD 766 ATA100 controller";
|
|
|
|
|
|
2001-03-06 21:43:46 +00:00
|
|
|
|
case 0x02111166:
|
|
|
|
|
return "ServerWorks ROSB4 ATA33 controller";
|
|
|
|
|
|
|
|
|
|
case 0x4d33105a:
|
|
|
|
|
return "Promise ATA33 controller";
|
|
|
|
|
|
|
|
|
|
case 0x4d38105a:
|
|
|
|
|
return "Promise ATA66 controller";
|
|
|
|
|
|
|
|
|
|
case 0x0d30105a:
|
|
|
|
|
case 0x4d30105a:
|
2001-08-28 13:36:06 +00:00
|
|
|
|
return "Promise ATA100 controller";
|
|
|
|
|
|
2001-06-08 09:51:33 +00:00
|
|
|
|
case 0x4d68105a:
|
2001-12-25 14:44:26 +00:00
|
|
|
|
case 0x6268105a:
|
2002-01-28 13:17:10 +00:00
|
|
|
|
if (pci_get_devid(GRANDPARENT(dev)) == 0x00221011 &&
|
|
|
|
|
pci_get_class(GRANDPARENT(dev)) == PCIC_BRIDGE) {
|
|
|
|
|
static long start = 0, end = 0;
|
|
|
|
|
|
|
|
|
|
/* we belive we are on a TX4, now do our (simple) magic */
|
|
|
|
|
if (pci_get_slot(dev) == 1) {
|
|
|
|
|
bus_get_resource(dev, SYS_RES_IRQ, 0, &start, &end);
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
return "Promise TX4 ATA100 controller (channel 0+1)";
|
2002-01-28 13:17:10 +00:00
|
|
|
|
}
|
|
|
|
|
else if (pci_get_slot(dev) == 2 && start && end) {
|
|
|
|
|
bus_set_resource(dev, SYS_RES_IRQ, 0, start, end);
|
|
|
|
|
start = end = 0;
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
return "Promise TX4 ATA100 controller (channel 2+3)";
|
2002-01-28 13:17:10 +00:00
|
|
|
|
}
|
|
|
|
|
else
|
|
|
|
|
start = end = 0;
|
|
|
|
|
}
|
2001-08-28 13:36:06 +00:00
|
|
|
|
return "Promise TX2 ATA100 controller";
|
2001-03-06 21:43:46 +00:00
|
|
|
|
|
2001-10-23 08:01:47 +00:00
|
|
|
|
case 0x4d69105a:
|
2001-12-25 14:44:26 +00:00
|
|
|
|
return "Promise TX2 ATA133 controller";
|
2001-10-23 08:01:47 +00:00
|
|
|
|
|
2001-03-06 21:43:46 +00:00
|
|
|
|
case 0x00041103:
|
|
|
|
|
switch (pci_get_revid(dev)) {
|
|
|
|
|
case 0x00:
|
|
|
|
|
case 0x01:
|
|
|
|
|
return "HighPoint HPT366 ATA66 controller";
|
|
|
|
|
case 0x02:
|
|
|
|
|
return "HighPoint HPT368 ATA66 controller";
|
|
|
|
|
case 0x03:
|
|
|
|
|
case 0x04:
|
|
|
|
|
return "HighPoint HPT370 ATA100 controller";
|
2001-10-23 09:25:57 +00:00
|
|
|
|
case 0x05:
|
|
|
|
|
return "HighPoint HPT372 ATA133 controller";
|
2001-03-06 21:43:46 +00:00
|
|
|
|
}
|
2001-12-25 14:44:26 +00:00
|
|
|
|
return NULL;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
|
2002-02-11 15:48:04 +00:00
|
|
|
|
case 0x00081103:
|
|
|
|
|
switch (pci_get_revid(dev)) {
|
|
|
|
|
case 0x07:
|
|
|
|
|
return "HighPoint HPT374 ATA133 controller";
|
|
|
|
|
}
|
|
|
|
|
return NULL;
|
|
|
|
|
|
2002-02-12 16:59:28 +00:00
|
|
|
|
case 0x000116ca:
|
|
|
|
|
return "Cenatek Rocket Drive controller";
|
|
|
|
|
|
2001-03-06 21:43:46 +00:00
|
|
|
|
/* unsupported but known chipsets, generic DMA only */
|
|
|
|
|
case 0x10001042:
|
|
|
|
|
case 0x10011042:
|
|
|
|
|
return "RZ 100? ATA controller !WARNING! buggy chip data loss possible";
|
|
|
|
|
|
|
|
|
|
case 0x06401095:
|
|
|
|
|
return "CMD 640 ATA controller !WARNING! buggy chip data loss possible";
|
|
|
|
|
|
|
|
|
|
/* unknown chipsets, try generic DMA if it seems possible */
|
|
|
|
|
default:
|
|
|
|
|
if (pci_get_class(dev) == PCIC_STORAGE &&
|
|
|
|
|
(pci_get_subclass(dev) == PCIS_STORAGE_IDE))
|
|
|
|
|
return "Generic PCI ATA controller";
|
|
|
|
|
}
|
|
|
|
|
return NULL;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int
|
|
|
|
|
ata_pci_probe(device_t dev)
|
|
|
|
|
{
|
|
|
|
|
const char *desc = ata_pci_match(dev);
|
|
|
|
|
|
|
|
|
|
if (desc) {
|
|
|
|
|
device_set_desc(dev, desc);
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
else
|
|
|
|
|
return ENXIO;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int
|
|
|
|
|
ata_pci_add_child(device_t dev, int unit)
|
|
|
|
|
{
|
|
|
|
|
device_t child;
|
|
|
|
|
|
|
|
|
|
/* check if this is located at one of the std addresses */
|
|
|
|
|
if (ATA_MASTERDEV(dev)) {
|
|
|
|
|
if (!(child = device_add_child(dev, "ata", unit)))
|
|
|
|
|
return ENOMEM;
|
|
|
|
|
}
|
|
|
|
|
else {
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
if (!(child =
|
|
|
|
|
device_add_child(dev, "ata",
|
|
|
|
|
devclass_find_free_unit(ata_devclass, 2))))
|
2001-03-06 21:43:46 +00:00
|
|
|
|
return ENOMEM;
|
|
|
|
|
}
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int
|
|
|
|
|
ata_pci_attach(device_t dev)
|
|
|
|
|
{
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
struct ata_pci_controller *controller = device_get_softc(dev);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
u_int8_t class, subclass;
|
|
|
|
|
u_int32_t type, cmd;
|
|
|
|
|
int rid;
|
|
|
|
|
|
|
|
|
|
/* set up vendor-specific stuff */
|
|
|
|
|
type = pci_get_devid(dev);
|
|
|
|
|
class = pci_get_class(dev);
|
|
|
|
|
subclass = pci_get_subclass(dev);
|
|
|
|
|
cmd = pci_read_config(dev, PCIR_COMMAND, 4);
|
|
|
|
|
|
2001-03-19 13:31:58 +00:00
|
|
|
|
if (!(cmd & PCIM_CMD_PORTEN)) {
|
|
|
|
|
device_printf(dev, "ATA channel disabled by BIOS\n");
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
2001-03-06 21:43:46 +00:00
|
|
|
|
/* is busmastering supported ? */
|
|
|
|
|
if ((cmd & (PCIM_CMD_PORTEN | PCIM_CMD_BUSMASTEREN)) ==
|
|
|
|
|
(PCIM_CMD_PORTEN | PCIM_CMD_BUSMASTEREN)) {
|
|
|
|
|
|
|
|
|
|
/* is there a valid port range to connect to ? */
|
|
|
|
|
rid = 0x20;
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
controller->bmio = bus_alloc_resource(dev, SYS_RES_IOPORT, &rid,
|
|
|
|
|
0, ~0, 1, RF_ACTIVE);
|
|
|
|
|
if (!controller->bmio)
|
2001-03-06 21:43:46 +00:00
|
|
|
|
device_printf(dev, "Busmastering DMA not configured\n");
|
|
|
|
|
}
|
|
|
|
|
else
|
|
|
|
|
device_printf(dev, "Busmastering DMA not supported\n");
|
|
|
|
|
|
|
|
|
|
/* do extra chipset specific setups */
|
|
|
|
|
switch (type) {
|
|
|
|
|
case 0x522910b9: /* Aladdin need to activate the ATAPI FIFO */
|
|
|
|
|
pci_write_config(dev, 0x53,
|
|
|
|
|
(pci_read_config(dev, 0x53, 1) & ~0x01) | 0x02, 1);
|
|
|
|
|
break;
|
|
|
|
|
|
2001-08-28 13:36:06 +00:00
|
|
|
|
case 0x4d38105a: /* Promise 66 & 100 (before TX2) need the clock changed */
|
2001-03-06 21:43:46 +00:00
|
|
|
|
case 0x4d30105a:
|
|
|
|
|
case 0x0d30105a:
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
ATA_OUTB(controller->bmio, 0x11, ATA_INB(controller->bmio, 0x11)|0x0a);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
/* FALLTHROUGH */
|
|
|
|
|
|
2001-08-28 13:36:06 +00:00
|
|
|
|
case 0x4d33105a: /* Promise (before TX2) need burst mode turned on */
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
ATA_OUTB(controller->bmio, 0x1f, ATA_INB(controller->bmio, 0x1f)|0x01);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
break;
|
|
|
|
|
|
2001-12-25 14:44:26 +00:00
|
|
|
|
case 0x00041103: /* HighPoint HPT 366/368/370/372 */
|
2002-02-11 15:48:04 +00:00
|
|
|
|
case 0x00081103: /* HighPoint HPT 374 */
|
2001-03-06 21:43:46 +00:00
|
|
|
|
switch (pci_get_revid(dev)) {
|
2001-10-23 09:25:57 +00:00
|
|
|
|
case 0x00: /* HPT 366 */
|
2001-03-06 21:43:46 +00:00
|
|
|
|
case 0x01:
|
|
|
|
|
/* turn off interrupt prediction */
|
|
|
|
|
pci_write_config(dev, 0x51,
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
(pci_read_config(dev, 0x51, 1) & ~0x80), 1);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
break;
|
|
|
|
|
|
2001-10-23 09:25:57 +00:00
|
|
|
|
case 0x02: /* HPT 368 */
|
|
|
|
|
case 0x03: /* HPT 370 */
|
2001-03-06 21:43:46 +00:00
|
|
|
|
case 0x04:
|
2001-10-23 09:25:57 +00:00
|
|
|
|
case 0x05: /* HPT 372 */
|
2002-02-11 15:48:04 +00:00
|
|
|
|
case 0x07: /* HPT 374 */
|
2001-03-06 21:43:46 +00:00
|
|
|
|
/* turn off interrupt prediction */
|
|
|
|
|
pci_write_config(dev, 0x51,
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
(pci_read_config(dev, 0x51, 1) & ~0x03), 1);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
pci_write_config(dev, 0x55,
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
(pci_read_config(dev, 0x55, 1) & ~0x03), 1);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
/* turn on interrupts */
|
|
|
|
|
pci_write_config(dev, 0x5a,
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
(pci_read_config(dev, 0x5a, 1) & ~0x10), 1);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
}
|
|
|
|
|
break;
|
|
|
|
|
|
2001-09-25 17:10:39 +00:00
|
|
|
|
case 0x05711106: /* VIA 82C586, '596, '686 default setup */
|
2002-01-28 13:17:10 +00:00
|
|
|
|
/* prepare for ATA-66 on the 82C686a and 82C596b */
|
|
|
|
|
if ((ata_find_dev(dev, 0x06861106, 0x10) &&
|
2001-09-25 17:10:39 +00:00
|
|
|
|
!ata_find_dev(dev, 0x06861106, 0x40)) ||
|
|
|
|
|
ata_find_dev(dev, 0x05961106, 0x12))
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
pci_write_config(dev, 0x50, 0x030b030b, 4);
|
2001-09-25 17:10:39 +00:00
|
|
|
|
|
2002-01-01 19:58:11 +00:00
|
|
|
|
/* the southbridge might need the data corruption fix */
|
|
|
|
|
if (ata_find_dev(dev, 0x06861106, 0x40) ||
|
|
|
|
|
ata_find_dev(dev, 0x82311106, 0x10))
|
|
|
|
|
ata_via_southbridge_fixup(dev);
|
2001-09-25 17:10:39 +00:00
|
|
|
|
/* FALLTHROUGH */
|
|
|
|
|
|
|
|
|
|
case 0x74091022: /* AMD 756 default setup */
|
|
|
|
|
case 0x74111022: /* AMD 766 default setup */
|
2001-03-06 21:43:46 +00:00
|
|
|
|
|
|
|
|
|
/* set prefetch, postwrite */
|
|
|
|
|
pci_write_config(dev, 0x41, pci_read_config(dev, 0x41, 1) | 0xf0, 1);
|
|
|
|
|
|
|
|
|
|
/* set fifo configuration half'n'half */
|
|
|
|
|
pci_write_config(dev, 0x43,
|
|
|
|
|
(pci_read_config(dev, 0x43, 1) & 0x90) | 0x2a, 1);
|
|
|
|
|
|
|
|
|
|
/* set status register read retry */
|
|
|
|
|
pci_write_config(dev, 0x44, pci_read_config(dev, 0x44, 1) | 0x08, 1);
|
|
|
|
|
|
|
|
|
|
/* set DMA read & end-of-sector fifo flush */
|
|
|
|
|
pci_write_config(dev, 0x46,
|
|
|
|
|
(pci_read_config(dev, 0x46, 1) & 0x0c) | 0xf0, 1);
|
|
|
|
|
|
|
|
|
|
/* set sector size */
|
|
|
|
|
pci_write_config(dev, 0x60, DEV_BSIZE, 2);
|
|
|
|
|
pci_write_config(dev, 0x68, DEV_BSIZE, 2);
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
case 0x10001042: /* RZ 100? known bad, no DMA */
|
|
|
|
|
case 0x10011042:
|
|
|
|
|
case 0x06401095: /* CMD 640 known bad, no DMA */
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
controller->bmio = NULL;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
device_printf(dev, "Busmastering DMA disabled\n");
|
|
|
|
|
}
|
|
|
|
|
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
if (controller->bmio) {
|
|
|
|
|
controller->bmaddr = rman_get_start(controller->bmio);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
BUS_RELEASE_RESOURCE(device_get_parent(dev), dev,
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
SYS_RES_IOPORT, rid, controller->bmio);
|
|
|
|
|
controller->bmio = NULL;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* the Cypress chip is a mess, it contains two ATA functions, but
|
|
|
|
|
* both channels are visible on the first one.
|
|
|
|
|
* simply ignore the second function for now, as the right
|
|
|
|
|
* solution (ignoring the second channel on the first function)
|
|
|
|
|
* doesn't work with the crappy ATA interrupt setup on the alpha.
|
|
|
|
|
*/
|
|
|
|
|
if (pci_get_devid(dev) == 0xc6931080 && pci_get_function(dev) > 1)
|
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
|
|
ata_pci_add_child(dev, 0);
|
|
|
|
|
|
|
|
|
|
if (ATA_MASTERDEV(dev) || pci_read_config(dev, 0x18, 4) & IOMASK)
|
|
|
|
|
ata_pci_add_child(dev, 1);
|
|
|
|
|
|
|
|
|
|
return bus_generic_attach(dev);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
ata_pci_intr(struct ata_channel *ch)
|
2001-03-06 21:43:46 +00:00
|
|
|
|
{
|
|
|
|
|
u_int8_t dmastat;
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* since we might share the IRQ with another device, and in some
|
|
|
|
|
* cases with our twin channel, we only want to process interrupts
|
|
|
|
|
* that we know this channel generated.
|
|
|
|
|
*/
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
switch (ch->chiptype) {
|
|
|
|
|
case 0x00041103: /* HighPoint HPT366/368/370/372 */
|
2002-02-11 15:48:04 +00:00
|
|
|
|
case 0x00081103: /* HighPoint HPT374 */
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
if (((dmastat = ata_dmastatus(ch)) &
|
2001-03-06 21:43:46 +00:00
|
|
|
|
(ATA_BMSTAT_ACTIVE | ATA_BMSTAT_INTERRUPT)) != ATA_BMSTAT_INTERRUPT)
|
|
|
|
|
return 1;
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
ATA_OUTB(ch->r_bmio, ATA_BMSTAT_PORT, dmastat | ATA_BMSTAT_INTERRUPT);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
DELAY(1);
|
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
|
|
case 0x06481095: /* CMD 648 */
|
|
|
|
|
case 0x06491095: /* CMD 649 */
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
if (!(pci_read_config(device_get_parent(ch->dev), 0x71, 1) &
|
|
|
|
|
(ch->unit ? 0x08 : 0x04)))
|
2001-03-06 21:43:46 +00:00
|
|
|
|
return 1;
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
case 0x4d33105a: /* Promise Ultra/Fasttrak 33 */
|
|
|
|
|
case 0x4d38105a: /* Promise Ultra/Fasttrak 66 */
|
|
|
|
|
case 0x4d30105a: /* Promise Ultra/Fasttrak 100 */
|
|
|
|
|
case 0x0d30105a: /* Promise OEM ATA100 */
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
if (!(ATA_INL(ch->r_bmio, (ch->unit ? 0x14 : 0x1c)) &
|
|
|
|
|
(ch->unit ? 0x00004000 : 0x00000400)))
|
2001-03-06 21:43:46 +00:00
|
|
|
|
return 1;
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
break;
|
2001-10-03 08:27:29 +00:00
|
|
|
|
|
|
|
|
|
case 0x4d68105a: /* Promise TX2 ATA100 */
|
2001-12-25 14:44:26 +00:00
|
|
|
|
case 0x6268105a: /* Promise TX2 ATA100 */
|
|
|
|
|
case 0x4d69105a: /* Promise TX2 ATA133 */
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
ATA_OUTB(ch->r_bmio, ATA_BMDEVSPEC_0, 0x0b);
|
|
|
|
|
if (!(ATA_INB(ch->r_bmio, ATA_BMDEVSPEC_1) & 0x20))
|
2001-10-03 08:27:29 +00:00
|
|
|
|
return 1;
|
|
|
|
|
break;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
}
|
|
|
|
|
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
if (ch->flags & ATA_DMA_ACTIVE) {
|
|
|
|
|
if (!((dmastat = ata_dmastatus(ch)) & ATA_BMSTAT_INTERRUPT))
|
2001-03-06 21:43:46 +00:00
|
|
|
|
return 1;
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
ATA_OUTB(ch->r_bmio, ATA_BMSTAT_PORT, dmastat | ATA_BMSTAT_INTERRUPT);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
DELAY(1);
|
|
|
|
|
}
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int
|
|
|
|
|
ata_pci_print_child(device_t dev, device_t child)
|
|
|
|
|
{
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
struct ata_channel *ch = device_get_softc(child);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
int retval = 0;
|
|
|
|
|
|
|
|
|
|
retval += bus_print_child_header(dev, child);
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
retval += printf(": at 0x%lx", rman_get_start(ch->r_io));
|
2001-03-06 21:43:46 +00:00
|
|
|
|
|
|
|
|
|
if (ATA_MASTERDEV(dev))
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
retval += printf(" irq %d", 14 + ch->unit);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
|
|
|
|
|
retval += bus_print_child_footer(dev, child);
|
|
|
|
|
|
|
|
|
|
return retval;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static struct resource *
|
|
|
|
|
ata_pci_alloc_resource(device_t dev, device_t child, int type, int *rid,
|
|
|
|
|
u_long start, u_long end, u_long count, u_int flags)
|
|
|
|
|
{
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
struct ata_pci_controller *controller = device_get_softc(dev);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
struct resource *res = NULL;
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
int unit = ((struct ata_channel *)device_get_softc(child))->unit;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
int myrid;
|
|
|
|
|
|
|
|
|
|
if (type == SYS_RES_IOPORT) {
|
|
|
|
|
switch (*rid) {
|
|
|
|
|
case ATA_IOADDR_RID:
|
|
|
|
|
if (ATA_MASTERDEV(dev)) {
|
|
|
|
|
myrid = 0;
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
start = (unit ? ATA_SECONDARY : ATA_PRIMARY);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
end = start + ATA_IOSIZE - 1;
|
|
|
|
|
count = ATA_IOSIZE;
|
|
|
|
|
res = BUS_ALLOC_RESOURCE(device_get_parent(dev), child,
|
|
|
|
|
SYS_RES_IOPORT, &myrid,
|
|
|
|
|
start, end, count, flags);
|
|
|
|
|
}
|
|
|
|
|
else {
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
myrid = 0x10 + 8 * unit;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
res = BUS_ALLOC_RESOURCE(device_get_parent(dev), dev,
|
|
|
|
|
SYS_RES_IOPORT, &myrid,
|
|
|
|
|
start, end, count, flags);
|
|
|
|
|
}
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
case ATA_ALTADDR_RID:
|
|
|
|
|
if (ATA_MASTERDEV(dev)) {
|
|
|
|
|
myrid = 0;
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
start = (unit ? ATA_SECONDARY : ATA_PRIMARY) + ATA_ALTOFFSET;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
end = start + ATA_ALTIOSIZE - 1;
|
|
|
|
|
count = ATA_ALTIOSIZE;
|
|
|
|
|
res = BUS_ALLOC_RESOURCE(device_get_parent(dev), child,
|
|
|
|
|
SYS_RES_IOPORT, &myrid,
|
|
|
|
|
start, end, count, flags);
|
|
|
|
|
}
|
|
|
|
|
else {
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
myrid = 0x14 + 8 * unit;
|
|
|
|
|
res = BUS_ALLOC_RESOURCE(device_get_parent(dev), dev,
|
2001-03-06 21:43:46 +00:00
|
|
|
|
SYS_RES_IOPORT, &myrid,
|
|
|
|
|
start, end, count, flags);
|
|
|
|
|
if (res) {
|
|
|
|
|
start = rman_get_start(res) + 2;
|
2001-03-19 08:02:42 +00:00
|
|
|
|
end = rman_get_start(res) + ATA_ALTIOSIZE - 1;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
count = ATA_ALTIOSIZE;
|
|
|
|
|
BUS_RELEASE_RESOURCE(device_get_parent(dev), dev,
|
|
|
|
|
SYS_RES_IOPORT, myrid, res);
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
res = BUS_ALLOC_RESOURCE(device_get_parent(dev), dev,
|
2001-03-06 21:43:46 +00:00
|
|
|
|
SYS_RES_IOPORT, &myrid,
|
|
|
|
|
start, end, count, flags);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
case ATA_BMADDR_RID:
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
if (controller->bmaddr) {
|
2001-03-06 21:43:46 +00:00
|
|
|
|
myrid = 0x20;
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
start = (unit == 0 ?
|
|
|
|
|
controller->bmaddr : controller->bmaddr+ATA_BMIOSIZE);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
end = start + ATA_BMIOSIZE - 1;
|
|
|
|
|
count = ATA_BMIOSIZE;
|
|
|
|
|
res = BUS_ALLOC_RESOURCE(device_get_parent(dev), child,
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
SYS_RES_IOPORT, &myrid,
|
|
|
|
|
start, end, count, flags);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
return res;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (type == SYS_RES_IRQ && *rid == ATA_IRQ_RID) {
|
|
|
|
|
if (ATA_MASTERDEV(dev)) {
|
|
|
|
|
#ifdef __alpha__
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
return alpha_platform_alloc_ide_intr(unit);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
#else
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
int irq = (unit == 0 ? 14 : 15);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
|
|
|
|
|
return BUS_ALLOC_RESOURCE(device_get_parent(dev), child,
|
2001-08-21 11:35:47 +00:00
|
|
|
|
SYS_RES_IRQ, rid, irq, irq, 1, flags);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
#endif
|
|
|
|
|
}
|
|
|
|
|
else {
|
2001-03-19 08:02:42 +00:00
|
|
|
|
/* primary and secondary channels share interrupt, keep track */
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
if (!controller->irq)
|
|
|
|
|
controller->irq = BUS_ALLOC_RESOURCE(device_get_parent(dev),
|
|
|
|
|
dev, SYS_RES_IRQ,
|
|
|
|
|
rid, 0, ~0, 1, flags);
|
|
|
|
|
controller->irqcnt++;
|
|
|
|
|
return controller->irq;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int
|
|
|
|
|
ata_pci_release_resource(device_t dev, device_t child, int type, int rid,
|
|
|
|
|
struct resource *r)
|
|
|
|
|
{
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
struct ata_pci_controller *controller = device_get_softc(dev);
|
|
|
|
|
int unit = ((struct ata_channel *)device_get_softc(child))->unit;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
|
|
|
|
|
if (type == SYS_RES_IOPORT) {
|
|
|
|
|
switch (rid) {
|
|
|
|
|
case ATA_IOADDR_RID:
|
|
|
|
|
if (ATA_MASTERDEV(dev))
|
2001-03-19 08:02:42 +00:00
|
|
|
|
return BUS_RELEASE_RESOURCE(device_get_parent(dev), child,
|
|
|
|
|
SYS_RES_IOPORT, 0x0, r);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
else
|
2001-03-19 08:02:42 +00:00
|
|
|
|
return BUS_RELEASE_RESOURCE(device_get_parent(dev), dev,
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
SYS_RES_IOPORT, 0x10 + 8 * unit, r);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
case ATA_ALTADDR_RID:
|
|
|
|
|
if (ATA_MASTERDEV(dev))
|
2001-03-19 08:02:42 +00:00
|
|
|
|
return BUS_RELEASE_RESOURCE(device_get_parent(dev), child,
|
|
|
|
|
SYS_RES_IOPORT, 0x0, r);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
else
|
2001-03-19 08:02:42 +00:00
|
|
|
|
return BUS_RELEASE_RESOURCE(device_get_parent(dev), dev,
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
SYS_RES_IOPORT, 0x14 + 8 * unit, r);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
case ATA_BMADDR_RID:
|
2001-03-19 08:02:42 +00:00
|
|
|
|
return BUS_RELEASE_RESOURCE(device_get_parent(dev), child,
|
|
|
|
|
SYS_RES_IOPORT, 0x20, r);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
default:
|
|
|
|
|
return ENOENT;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
if (type == SYS_RES_IRQ) {
|
|
|
|
|
if (rid != ATA_IRQ_RID)
|
|
|
|
|
return ENOENT;
|
|
|
|
|
|
|
|
|
|
if (ATA_MASTERDEV(dev)) {
|
|
|
|
|
#ifdef __alpha__
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
return alpha_platform_release_ide_intr(unit, r);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
#else
|
2001-03-19 08:02:42 +00:00
|
|
|
|
return BUS_RELEASE_RESOURCE(device_get_parent(dev), child,
|
|
|
|
|
SYS_RES_IRQ, rid, r);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
#endif
|
|
|
|
|
}
|
|
|
|
|
else {
|
2001-03-19 08:02:42 +00:00
|
|
|
|
/* primary and secondary channels share interrupt, keep track */
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
if (--controller->irqcnt)
|
2001-03-06 21:43:46 +00:00
|
|
|
|
return 0;
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
controller->irq = 0;
|
2001-03-19 08:02:42 +00:00
|
|
|
|
return BUS_RELEASE_RESOURCE(device_get_parent(dev), dev,
|
|
|
|
|
SYS_RES_IRQ, rid, r);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
return EINVAL;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int
|
|
|
|
|
ata_pci_setup_intr(device_t dev, device_t child, struct resource *irq,
|
|
|
|
|
int flags, driver_intr_t *intr, void *arg,
|
|
|
|
|
void **cookiep)
|
|
|
|
|
{
|
|
|
|
|
if (ATA_MASTERDEV(dev)) {
|
|
|
|
|
#ifdef __alpha__
|
|
|
|
|
return alpha_platform_setup_ide_intr(child, irq, intr, arg, cookiep);
|
|
|
|
|
#else
|
|
|
|
|
return BUS_SETUP_INTR(device_get_parent(dev), child, irq,
|
|
|
|
|
flags, intr, arg, cookiep);
|
|
|
|
|
#endif
|
|
|
|
|
}
|
|
|
|
|
else
|
|
|
|
|
return BUS_SETUP_INTR(device_get_parent(dev), dev, irq,
|
|
|
|
|
flags, intr, arg, cookiep);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int
|
|
|
|
|
ata_pci_teardown_intr(device_t dev, device_t child, struct resource *irq,
|
|
|
|
|
void *cookie)
|
|
|
|
|
{
|
|
|
|
|
if (ATA_MASTERDEV(dev)) {
|
|
|
|
|
#ifdef __alpha__
|
|
|
|
|
return alpha_platform_teardown_ide_intr(child, irq, cookie);
|
|
|
|
|
#else
|
|
|
|
|
return BUS_TEARDOWN_INTR(device_get_parent(dev), child, irq, cookie);
|
|
|
|
|
#endif
|
|
|
|
|
}
|
|
|
|
|
else
|
|
|
|
|
return BUS_TEARDOWN_INTR(device_get_parent(dev), dev, irq, cookie);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static device_method_t ata_pci_methods[] = {
|
|
|
|
|
/* device interface */
|
|
|
|
|
DEVMETHOD(device_probe, ata_pci_probe),
|
|
|
|
|
DEVMETHOD(device_attach, ata_pci_attach),
|
|
|
|
|
DEVMETHOD(device_shutdown, bus_generic_shutdown),
|
|
|
|
|
DEVMETHOD(device_suspend, bus_generic_suspend),
|
|
|
|
|
DEVMETHOD(device_resume, bus_generic_resume),
|
|
|
|
|
|
|
|
|
|
/* bus methods */
|
|
|
|
|
DEVMETHOD(bus_print_child, ata_pci_print_child),
|
|
|
|
|
DEVMETHOD(bus_alloc_resource, ata_pci_alloc_resource),
|
|
|
|
|
DEVMETHOD(bus_release_resource, ata_pci_release_resource),
|
|
|
|
|
DEVMETHOD(bus_activate_resource, bus_generic_activate_resource),
|
|
|
|
|
DEVMETHOD(bus_deactivate_resource, bus_generic_deactivate_resource),
|
|
|
|
|
DEVMETHOD(bus_setup_intr, ata_pci_setup_intr),
|
|
|
|
|
DEVMETHOD(bus_teardown_intr, ata_pci_teardown_intr),
|
|
|
|
|
{ 0, 0 }
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
static driver_t ata_pci_driver = {
|
|
|
|
|
"atapci",
|
|
|
|
|
ata_pci_methods,
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
sizeof(struct ata_pci_controller),
|
2001-03-06 21:43:46 +00:00
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
static devclass_t ata_pci_devclass;
|
|
|
|
|
|
|
|
|
|
DRIVER_MODULE(atapci, pci, ata_pci_driver, ata_pci_devclass, 0, 0);
|
|
|
|
|
|
|
|
|
|
static int
|
|
|
|
|
ata_pcisub_probe(device_t dev)
|
|
|
|
|
{
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
struct ata_channel *ch = device_get_softc(dev);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
device_t *children;
|
|
|
|
|
int count, i;
|
|
|
|
|
|
|
|
|
|
/* find channel number on this controller */
|
|
|
|
|
device_get_children(device_get_parent(dev), &children, &count);
|
|
|
|
|
for (i = 0; i < count; i++) {
|
|
|
|
|
if (children[i] == dev)
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
ch->unit = i;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
}
|
|
|
|
|
free(children, M_TEMP);
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
ch->chiptype = pci_get_devid(device_get_parent(dev));
|
|
|
|
|
ch->intr_func = ata_pci_intr;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
return ata_probe(dev);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static device_method_t ata_pcisub_methods[] = {
|
|
|
|
|
/* device interface */
|
|
|
|
|
DEVMETHOD(device_probe, ata_pcisub_probe),
|
|
|
|
|
DEVMETHOD(device_attach, ata_attach),
|
|
|
|
|
DEVMETHOD(device_detach, ata_detach),
|
|
|
|
|
DEVMETHOD(device_resume, ata_resume),
|
|
|
|
|
{ 0, 0 }
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
static driver_t ata_pcisub_driver = {
|
|
|
|
|
"ata",
|
|
|
|
|
ata_pcisub_methods,
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
sizeof(struct ata_channel),
|
2001-03-06 21:43:46 +00:00
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
DRIVER_MODULE(ata, atapci, ata_pcisub_driver, ata_devclass, 0, 0);
|