HELP!

Michael Poole poole at graviton.res.cmu.edu
Thu Dec 10 08:23:27 PST 1998


Kis Zsolt <zsolt at bird.szfki.kfki.hu> writes:

> Hi,
> I am a new and very disappointed owner of an ASUS P2L97DS mainboard which 
> has a built in AIC7880 SCSI controller chip. There is a Teac CD writer 
> and an IBM UW harddrave on the bus. The termination is O.K. The bus hangs 
> frequently, and in the /var/log/messages file I found:
> 
> Dec 10 17:02:45 merle kernel: scsi : aborting command due to timeout : 
> pid 10588, scsi0, channel 0, i
> d 12, lun 0 Write (10) 00 00 49 e0 7b 00 00 02 00 
> Dec 10 17:02:45 merle kernel: scsi : aborting command due to timeout : 
> pid 10589, scsi0, channel 0, i
> d 12, lun 0 Write (10) 00 00 49 e8 37 00 00 04 00 
> Dec 10 17:02:45 merle kernel: scsi : aborting command due to timeout : 
> pid 10590, scsi0, channel 0, i
> d 12, lun 0 Write (10) 00 00 4a 1b 31 00 00 02 00 
> Dec 10 17:02:47 merle kernel: SCSI host 0 abort (pid 10588) timed out - 
> resetting
> Dec 10 17:02:47 merle kernel: SCSI bus is being reset for host 0 channel 0.
> Dec 10 17:02:48 merle kernel: (scsi0:0:12:0) Synchronous at 40.0 
> Mbyte/sec, offset 8.
> 
> Does anibody know what can be done?

    I had similar problems on my computer, with the same motherboard.
I have a Matsushita (Panasonic) CD-R, a Toshiba CD-ROM, and a few
Seagate UW hard drives.  Whenever ultra negotiation is turned on in the
SCSI BIOS, I get that sort of reset every 15 seconds or so -- since each
reset takes a few seconds, this makes the system mostly unusable.  I
thought it was just a glitch in the CD-R firmware, but if the same problem
occurs with other types of CD-Rs, perhaps it's in the motherboard -- it's
not a Linux-specific problem, since the same thing would happen under
NT 5 beta 1 and Windows 95.  My current "solution" is to disable the ultra
negotiation (with the associated halving of transfer rate to my UW drives).

-- Michael

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



More information about the aic7xxx mailing list