Still SCSI problems in 2.0.33 (AHA2940)

[Ulrich Windl]

Hello,

I had reported SCSI driver problems for AHA2940 during backup to 
tape. Yesterday I had installed Linux 2.0.33 and I still had an 
error, but the details were better: The system did not crash, the 
application got an I/O error (or EOF), and I could capture the 
messages in syslog.

I tried to reproduce the problem, but could not. Maybe it's an 
initialization problem.

22:00:03: scsi0: Referenced SCB 255 not valid during SELTO.
22:00:03:         SCSISEQ = 0x5a SEQADDR = 0x4 SSTAT0 = 0x15 SSTAT1 = 0x8a
22:00:23: scsi : aborting command due to timeout : pid 12149, scsi0, channel 0, id 0, lun 0 Read (6) 0e 00 98 04 00 
22:00:23: (scsi0:0:0:0) Aborting scb 2, flags 0x1
22:00:23: scsi : aborting command due to timeout : pid 12150, scsi0, channel 0, id 0, lun 0 Read (6) 14 a3 90 04 00 
22:00:24: SCSI host 0 abort (pid 12149) timed out - resetting
22:00:24: SCSI bus is being reset for host 0 channel 0.
22:00:24: (scsi0:0:0:0) Reset called, scb 2, flags 0x41
22:00:24: (scsi0:0:-1:-1) Channel reset
22:00:24: (scsi0:0:0:0) Synchronous at 10.0MHz, offset 15.    # my harddisk
22:00:25: (scsi0:0:2:0) Synchronous at 5.0MHz, offset 8.      # my DDS drive

Adaptec AIC7xxx driver version: 4.1.1/3.2.1                   # /proc/scsi/aic7xxx/0
Compile Options:
  AIC7XXX_RESET_DELAY    : 5
  AIC7XXX_TAGGED_QUEUEING: Enabled
  AIC7XXX_PAGE_ENABLE    : Enabled
  AIC7XXX_PROC_STATS     : Enabled

Adapter Configuration:
           SCSI Adapter: Adaptec AHA-294X SCSI host adapter
                         (AIC-787x chipset)
               Host Bus: Single
                Base IO: 0xe000
         Base IO Memory: 0xf9800000
                    IRQ: 11
                   SCBs: Used 9, HW 16, Page 255
             Interrupts: 81978
          Serial EEPROM: True
   Extended Translation: Enabled
         SCSI Bus Reset: Enabled
             Ultra SCSI: Disabled
Disconnect Enable Flags: 0xff

Statistics:
CHAN#A (TGT 0 LUN 0):
nxfers 32412 (31096 read;1316 written)
blks(512) rd=1214273; blks(512) wr=7290
        < 512 512-1K   1-2K   2-4K   4-8K  8-16K 16-32K 32-64K 64-128K >128K
 Reads:     1      4     43  17033   2132   1975    732   9146     30      0
Writes:     0      0      9   1043    217     39      6      2      0      0

CHAN#A (TGT 2 LUN 0):
nxfers 49524 (49524 read;0 written)
blks(512) rd=990240; blks(512) wr=0
        < 512 512-1K   1-2K   2-4K   4-8K  8-16K 16-32K 32-64K 64-128K >128K
 Reads:    12      0      0      0      0  49512      0      0      0      0
Writes:     0      0      0      0      0      0      0      0      0      0


# Still the SCSI statistics for ID2 (TAPE) are wrong: I read and wrote from and
# to tape, but only reads are displayed (see above).

Regards,
Ulrich