[prev in list] [next in list] [prev in thread] [next in thread] 

List:       aic7xxx
Subject:    Re: Freezing with aic7895 and an SMP machine.
From:       Doug Ledford <dledford () redhat ! com>
Date:       1999-09-09 5:03:00
[Download RAW message or body]

sysuh@kigam.re.kr wrote:
> 
> Doug Ledford <dledford@redhat.com> wrote:
> >"Matthew M. DeLoera" wrote:
> >> scsi : aborting command due to timeout : pid 0, scsi0, channel 0, id 0,
> >>        lun 0 Test Unit Ready 00 00 00 00 00
> >> scsi : aborting command due to timeout : pid 0, scsi0, channel 0, id 0,
> >>        lun 0 Test Unit Ready 00 00 00 00 00
>  ...
> >You're obviously compiling the new kernel as an SMP kernel.  Boot the kernel
> >with the option "noapic" and it will work.
> 
> Looks like aic7xxx and SMP are not good combination.

The evidence doesn't support that conclusion.  Good god, were talking about
two machines with totally different problems out of thousands of working
machines, that doesn't support a generalization that SMP and aic7xxx are bad
things.

> I am also getting a few timeout errors on an Adaptec U2W on an SMP.
> --
> sysuh
> 
> Sep  9 10:08:34 yfs kernel: scsi : aborting command due to timeout : pid 5994477,

pid 5994477.  Hmmm...a reset after 5.4 million SCSI operations may not be
optimal (I prefer that they never happen), but it's hardly the end of the
world.  OTOH, the original poster had immediate and permanent timeouts on his
bus (the machine wouldn't boot at all) and the real problem isn't the aic7xxx
driver, it's that his machine had multiple IOAPICs and the default Red Hat
Linux 6.0 kernel doesn't handle it correctly (which has been fixed in later
kernel stuff).  The noapic option gets around that by disabling APIC support
entirely and reverting back to standard PIC mode.

> scsi1, channel 0, id 0, lun 0 Read (10) 00 00 d6 e3 bf 00 00 80 00
> Sep  9 10:08:34 yfs kernel: SCSI host 1 abort (pid 5994477) timed out - resetting
> Sep  9 10:08:34 yfs kernel: SCSI bus is being reset for host 1 channel 0.
> Sep  9 10:08:35 yfs kernel: SCSI host 1 channel 0 reset (pid 5994477) timed out - trying harder
> Sep  9 10:08:35 yfs kernel: SCSI bus is being reset for host 1 channel 0.
> Sep  9 10:08:35 yfs kernel: SCSI host 1 reset (pid 5994477) timed out again -
> Sep  9 10:08:35 yfs kernel: probably an unrecoverable SCSI bus or device hang.

-- 
  Doug Ledford   <dledford@redhat.com>
   Opinions expressed are my own, but
      they should be everybody's.


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


[prev in list] [next in list] [prev in thread] [next in thread]