Hi All,
Wasn't sure whether this post we be more appropriate in the linux-kernel
newsgroup or not.... Hope nobody gets agitated :-)
I have an Adaptec AHA-2940U2W controller which seemingly boots fine (i.e. with
no timeouts) under Slackware kernel 2.0.35 (i.e. with
,aic7xxx-5.1.2-2.0.35.patch) and debian kernel 2.0.36. But with the newer
kernels ( under Slackware - tried 2.2.9 and 2.2.13 so far ) I get this annoying
timeout message (delay seems longer with the 2.2.13 kernel... on the order of
25 seconds):
scsi : aborting command due to timeout : pid 60, scsi0, channel 0, id 1, lun 0 0x28 00
00 70 c2 ce 00 00 02 00
Once the new kernel finally boots, drives seem to mount/read/write at correct
speeds and with no problems.
I tried a few of the boot switches in the README.aic7xxx (e.g.
"aic7xxx=extended", "aic7xxx=no_reset"... to no avail) but this was a whole
other adventuous excursion getting LILO to properly install which makes me
wonder if this LILO problem is somehow related. I have both UDMA33 IDE drives
as well as SCSI Ultra2 (IBM 9LZX [Model: DRVS09V] and QUANTUM Model:
ATLAS 10K 18WLS) drives. Wondering if this is part of the problem, but why no
trouble in the 2.0.35/6 kernels? LILO seems to execute flawlessly under the
debian 2.0.36 kernel.
I had to enter the following lines in the /etc/lilo.conf in order to keep the
machine from hanging while trying to read/write the MBR (at least with the new
kernel):
LINEAR
disk=/dev/hda bios=0x80
disk=/dev/hdb bios=0x81
disk=/dev/hdc bios=0x82
When trying to execute lilo, even with these above lines, I sill get the
warning messages:
Warning: BIOS drive 0x82 may not be accessible
...which I don't get under debian 2.0.36
I get a little more info when the kernel is booted with the "aic7xxx=verbose"
switch:
_} diff non_verbose.dmesg verbose.dmesg
< Detected 350801327 Hz processor.
---
> Detected 350803820 Hz processor.
98a99,102
> (scsi0:0:1:0) Aborting scb 0, flags 0x4
> (scsi0:0:1:0) SCB found in QINFIFO and aborted.
> (scsi0:0:1:0) Aborting scb 0
> (scsi0:-1:-1:-1) 1 commands found and queued for completion.
I have a an FIC PA-2013 MoBo/AMD K6-2 350MHz CPU/128M SDRAM setup with a Matrox
Millenium G200 video card. Slackware 7.0 is the current new kernel distribution.
I've seen some similiar posts in the news groups, but no real answers as to
what is causing this timeout in the newer kernels. Any further debug
suggestions would be much appreciated. Even better would be nice little 2.2.13
patch as a Christmas present :-)
TIA,
--Steve Cahill
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to [EMAIL PROTECTED]
