Hello David!
I have exactly the same problem of Meelis on my Sun Ultra Enterprise 3000.
Here's the output of my /proc/interrupts :
berta-test:~# cat /proc/interrupts
CPU10 CPU14
0: 8346 7272 <NULL> timer
2: 7 0 sun4u zs
4: 0 0 sun4u SYSIO_UE
5: 0 0 sun4u SYSIO_CE
6: 0 0 sun4u SYSIO_SBERR
13: 0 0 sun4u SYSIO_UE
14: 0 0 sun4u SYSIO_CE
15: 0 0 sun4u SYSIO_SBERR
19: 1676 0 sun4u eth0
20: 1604 11293 sun4u ESP
22: 0 0 sun4u SYSIO_UE
23: 0 0 sun4u SYSIO_CE
24: 0 0 sun4u SYSIO_SBERR
31: 0 0 sun4u SYSIO_UE
32: 0 0 sun4u SYSIO_CE
33: 0 0 sun4u SYSIO_SBERR
38: 15 0 sun4u ESP
39: 0 7 sun4u ESP
berta-test:~#
Hope it can help to solve the problem!
Benoit-Pierre
From: David Miller <davem@xxxxxxxxxxxxx>
> From: David Miller <davem@xxxxxxxxxxxxx>
> Date: Wed, 04 Mar 2009 03:35:14 -0800 (PST)
>
> > From: Meelis Roos <mroos@xxxxxxxx>
> > Date: Wed, 4 Mar 2009 13:22:21 +0200 (EET)
> >
> > > > I successfully installed Debian Lenny with 2.6.26 kernel, later compiled
> > > > yesterdays 2.6.29-rc6+git. Both kernels have the same problem - on
> > > > shutdown, scsi disk goes offline like in console log below. This is
> > > > repeatable and happens every shutdown so far.
> > >
> > > Testing by hand shows that 'ifconfig eth0 down' on eth0 (SBus hme) is
> > > the command that kills esp.
> >
> > Thanks for figuring out this clue. I'll see if there is anything
> > obvious in that area.
>
> Meelis... if ESP is aborting commands this smells like
> interrupt problems. I suspect that when HME is brought
> down, this accidently disables ESP's interrupt.
>
> Do any of your hme chips share interrupts with esp chips?
>
> What does /proc/interrupts look like on this system when
> everything is up and running?
>
> Thanks!
>
--
To unsubscribe from this list: send the line "unsubscribe sparclinux" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
