Traffic from brian martin <brian@cats.ucsc.edu>:
> I'm still having problems with the newer kernels ( > 2.1.93 ) not being
> ablt to detect my on-board scsi. I was secretly hoping that 2.1.96 would
> be the answer, but it doesn't seem to be that way. I've got an integrated
> scsi controller with an ncr53c7xx chipset. I've been sucessfully running
> kernels since somewhere around 2.1.59 with the same setup, but some recent
> change has made it so on bootup it says
>
> SCSI: 0 hosts
I have an Adaptec 2940U (not the more popular 2940UW) and have acheived
the same results on 2.1.96 and 2.1.97. The last attempted working kernel
for me was 2.1.91. I have not tried 2.1.93 or 2.1.94.
> It's an SMP system, and I've tried both UP and SMP kernels with no
> difference.
I have attempted these same alterations. In addition, I have attempted
the complete de-modularization of the kernel (I usually keep the tape
and cdrom devices as modules) and I have turned the verbosity of the
aic7xxx debug up to include PROBE2.
I should probably turn that up to 0xffff and try again.
No messages are ever issued from the aic7xxx driver proper. At this time,
it seems to me the kernel is never giving the aic7xxx a chance to probe.
For the sharing of common information, the working PROBE output of the
aic7xxx on 2.1.91:
aic7xxx: <Adaptec AHA-2940A Ultra SCSI host adapter> at PCI 9
aic7xxx: BIOS enabled, IO Port 0xe800, IO Mem 0xfebff000, IRQ 15, Revision C
aic7xxx: Single Channel, SCSI ID 7, 3/8 SCBs, QFull 8, QMask 0xf
> I find it hard to believe that noone else has had this problem, but even
> so, can anyone give me any direction as to how to debug the issue myself?
> I've not done any kernel hacking per se, but I can hold my own in C and am
> interested in running the newest kernels.
The solution to this problem is not immediately critical to my existence.
If no one tells me not to, I'll start my own (misguided) work on debugging
the problem specific to my configuration. Due to circumstances regarding
my transition of employment, I probably won't be able to share the results
of my efforts, if any, for a substantial time period.
I have been able to find reference to this same symptom on the linux-scsi
archive:
http://www.linuxhq.com/lnxlists/linux-scsi/ls_9804_03/msg00004.html
Brian Martin should compare notes with the above gentleman since you share
the same hardware platform.
No later reference on the linux-kernel archive, and no stated fixes.
-- David Hankins, "If you don't do it right the first time, Network Mercenary you'll just have to do it again." CURRENTLY FOR HIRE, http://www.mercenary.net/ -- J.T. Hankins- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu