talexander
/
freebsd_amp_hwpstate
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

>From BETA_1_1 branch: 

----------------------------
revision 1.20.2.2
date: 1994/05/03 05:16:50;  author: rgrimes;  state: Exp;  lines: +39 -16
Do a board reset if the AHA_INQUIRE command fails.  This is needed for the
DTC3290.  Change the messages about the BT545 to be generic to non adaptec
boards, since it appears that most vendors do not implement the AHA_INQUIRE
command in thier compatibility mode.
----------------------------
This commit is contained in:
Rodney W. Grimes 1994-05-03 05:44:53 +00:00
parent 61f969e858
commit 8568be7ca9
1 changed files with 39 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

55
sys/i386/isa/aha1542.c
View File

@ -12,7 +12,7 @@
* on the understanding that TFS is not responsible for the correct
* functioning of this software in any circumstances.
*
* $Id: aha1542.c,v 1.23 1994/03/23 09:15:26 davidg Exp $
* $Id: aha1542.c,v 1.24 1994/04/21 22:14:43 wollman Exp $
*/
/*
@ -848,40 +848,63 @@ aha_init(unit)
for (i = AHA_RESET_TIMEOUT; i; i--) {
sts = inb(AHA_CTRL_STAT_PORT);
if (sts == (AHA_IDLE | AHA_INIT))
if (sts == (AHA_IDLE | AHA_INIT)) {
break;
}
DELAY(1000); /* calibrated in msec */
}
#ifdef AHADEBUG
printf("aha_init: AHA_RESET_TIMEOUT went to %d\n", i);
#endif /* AHADEBUG */
if (i == 0) {
#ifdef AHADEBUG
if (aha_debug)
printf("aha_init: No answer from adaptec board\n");
printf("aha_init: No answer from board\n");
#endif /*AHADEBUG */
return (ENXIO);
}
/*
* Assume we have a board at this stage, do an adapter inquire
* to find out what type of controller it is
* to find out what type of controller it is. If the AHA_INQUIRE
* command fails, blatter about it, nuke the boardid so the 1542C
* stuff gets skipped over, and reset the board again.
*/
aha_cmd(unit, 0, sizeof(inquire), 1 ,&inquire, AHA_INQUIRE);
if(aha_cmd(unit, 0, sizeof(inquire), 1 ,&inquire, AHA_INQUIRE)) {
/*
* Blah.. not a real adaptec board!!!
* Seems that the Buslogic 545S and the DTC3290 both get
* this wrong.
*/
printf ("aha%d: not a REAL adaptec board, may cause warnings\n",
unit);
inquire.boardid = 0;
outb(AHA_CTRL_STAT_PORT, AHA_HRST | AHA_SRST);
for (i = AHA_RESET_TIMEOUT; i; i--) {
sts = inb(AHA_CTRL_STAT_PORT);
if (sts == (AHA_IDLE | AHA_INIT)) {
break;
}
DELAY(1000); /* calibrated in msec */
}
#ifdef AHADEBUG
printf("aha_init2: AHA_RESET_TIMEOUT went to %d\n", i);
#endif /* AHADEBUG */
if (i == 0) {
#ifdef AHADEBUG
if (aha_debug)
printf("aha_init2: No answer from board\n");
#endif /*AHADEBUG */
return (ENXIO);
}
}
#ifdef AHADEBUG
printf("aha%d: inquire %x, %x, %x, %x\n",
unit,
inquire.boardid, inquire.spec_opts,
inquire.revision_1, inquire.revision_2);
#endif /* AHADEBUG */
/*
* XXX The Buslogic 545S gets the AHA_INQUIRE command wrong,
* they only return one byte which causes us to print an error,
* so if the boardid comes back as 0x20, tell the user why they
* get the "cmd/data port empty" message
*/
if (inquire.boardid == 0x20) {
/* looks like a Buslogic 545 */
printf ("aha%d: above cmd/data port empty do to Buslogic 545\n",
unit);
}
/*
* If we are a 1542C or 1542CF disable the extended bios so that the
* mailbox interface is unlocked.