linux 2.4.8 vs 2.4.7 AHA2840A/42A "aic7xxx_abort return 8194" [kop@meme.com]

[Karl O . Pinc]

On 2001.09.13 15:58 Justin T. Gibbs wrote:
> >I've tried again using 2.4.9 and am still getting what seem to be the
> >same errors.

> 
> >> [root at arthur /root]# cat /proc/scsi/scsi
> >> Attached devices: 
> >> Host: scsi0 Channel: 00 Id: 00 Lun: 00
> >>   Vendor: QUANTUM  Model: FIREBALL ST4300S Rev: 0F0D
> >>   Type:   Direct-Access                    ANSI SCSI revision: 02
> 
> Known bad firmware.  Try upgrading to 0F0J if you can find it somewhere.
> Quantum used to have it on their ftp site, but sadly this seems to be
> gone now that Maxtor has acquired their disk drive division.

Upgraded the firmware and no longer have problems with this drive.  Thanks!

Called Maxtor and they sent me to the RMA department who wouldn't give
me an RMA because the drive is out of warrenty.  I found instructions
at http://www.stclairsoft.com/Sleeper/quantum_fix.html and a link to
an Apple updater utility (
ftp://ftp.info.apple.com/Apple_Support_Area/Apple_Software_Updates/English-North_American/Macintosh/Utilities/Firmware_Utilities/2-3-4-6_GB_Firmware_Util.img.bin)
which only works when run from a certain vintage PPC Mac and managed
to get the firmware upgraded.

> >> Host: scsi0 Channel: 00 Id: 04 Lun: 00
> >>   Vendor: IMPRIMIS Model: 94601-15         Rev: 4202
> >>   Type:   Direct-Access                    ANSI SCSI revision: 01 CCS
> >> Host: scsi0 Channel: 00 Id: 05 Lun: 00
> >>   Vendor: CONNER   Model: CFP1060S 1.05GB  Rev: 245F
> >>   Type:   Direct-Access                    ANSI SCSI revision: 02
> 
> I can't recall if it was the 1060 or the 1080 that had so many problems
> with tagged queuing, but it was one of the two.

Its the 1060, which is _supposed_ to be fixed if the firmware's >= 203C
(http://www.geocrawler.com/mail/msg.php3?msg_id=785749&list=160), but
245F is so far past 203C that I wonder if there's not another set of
firmware for the drives with the 80 pin hot-swapable interface, of
which mine is one.  (BTW, a firmware upgrade archive is available at
ftp://ftp.relcom.ru/pub/msdos/hardware/scsi/barunix.exe) I'm
especially suspicious as the Conner instructions identify the drives
to be updated by identifiers printed on the drive, rev's 9WA1.62,
9WA1.66, 9WA1.68 are all supposed to be upgraded.  My printed
identifier is 9WA2.66R1, which is suspiciously close enough to 9WA1.66
that the difference could be due to the hot-swap interface.

After more than a couple of hours of load, I get some ABORT errors on
the Conner 1060.  I'm going to try limiting the queue depth to 8 for
the drive and if that doesn't work I'll try disabling tagged queueing
for the drive.  I haven't tried upgrading the firmware.  I don't know
that I could find the DOS driver for the Adaptec card to do so, and
I'd like to wait until Monday and call tech support at Segate to look
into just what firmware revisions need a fix.

> >> scsi0:0:0:0: Attempting to queue an ABORT message
> >> scsi0:0:0:0: Command found on device queue
> >> aic7xxx_abort_returns 8194
> 
> Just this snippet doesn't tell much of anything.  You need to run with
> aic7xxx=verbose and record *all* of the messages displayed by the
> driver.  You may need to use a serial console to do this.

Do you suspect more than buggy drive firmware that you'd like me to do
this?  My plan is to wait until I think the drives work before setting
this up.

Thanks for the help.

Karl <kop at meme.com>

To Unsubscribe: send mail to majordomo at FreeBSD.org
with "unsubscribe aic7xxx" in the body of the message