APIC errors w/ 2.4.0-test11-pre2

APIC errors w/ 2.4.0-test11-pre2

[Bernhard Rosenkraenzer]

Hi,
after booting a 2.4.0 (any testx-release I've tried so far, including
test11-pre2) on a Dual-Pentium III box, the system works ok, but the
console gets filled with

APIC error on CPU0: 08(08)

every couple of seconds, occasionally some lines in between say

APIC error on CPU0: 08(02)

and

APIC error on CPU0: 02(08)

This doesn't happen with 2.2.x, but I've seen "unexpected IRQ vector 208
on CPU0" in 2.2.18 occasionally (nowhere near as often, maybe once a
month).

This is an ASUS P2-D board (Intel 440BX chipset), 2 Pentium III-700
processors.

The same kernel works perfectly on a similar (but slower) system, Gigabyte
P2B-D board (Intel 440BX chipset), 2 Pentium III-450 processors.

LLaP
bero


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: APIC errors w/ 2.4.0-test11-pre2

[H. Peter Anvin]

Followup to:  <Pine.LNX.4.21.0011101523170.14596-100000@bochum.redhat.de>
By author:    Bernhard Rosenkraenzer <bero@redhat.de>
In newsgroup: linux.dev.kernel
>
> Hi,
> after booting a 2.4.0 (any testx-release I've tried so far, including
> test11-pre2) on a Dual-Pentium III box, the system works ok, but the
> console gets filled with
> 
> APIC error on CPU0: 08(08)
> 
> every couple of seconds, occasionally some lines in between say
> 
> APIC error on CPU0: 08(02)
> 
> and
> 
> APIC error on CPU0: 02(08)
> 
> This doesn't happen with 2.2.x, but I've seen "unexpected IRQ vector 208
> on CPU0" in 2.2.18 occasionally (nowhere near as often, maybe once a
> month).
> 
> This is an ASUS P2-D board (Intel 440BX chipset), 2 Pentium III-700
> processors.
> 
> The same kernel works perfectly on a similar (but slower) system, Gigabyte
> P2B-D board (Intel 440BX chipset), 2 Pentium III-450 processors.
> 

I have seen the same problem on the same motherboard.  It appears to
be a motherboard bug that 2.4 exposes and 2.2 doesn't.

	-hpa


-- 
<hpa@transmeta.com> at work, <hpa@zytor.com> in private!
"Unix gives you enough rope to shoot yourself in the foot."
http://www.zytor.com/~hpa/puzzle.txt
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: APIC errors w/ 2.4.0-test11-pre2

[Rogier Wolff]

H. Peter Anvin wrote:
> Followup to:  <Pine.LNX.4.21.0011101523170.14596-100000@bochum.redhat.de>
> By author:    Bernhard Rosenkraenzer <bero@redhat.de>
> In newsgroup: linux.dev.kernel
> >
> > Hi,
> > after booting a 2.4.0 (any testx-release I've tried so far, including
> > test11-pre2) on a Dual-Pentium III box, the system works ok, but the
> > console gets filled with
> > 
> > APIC error on CPU0: 08(08)
> > 
> > every couple of seconds, occasionally some lines in between say
> > 
> > APIC error on CPU0: 08(02)
> > 
> > and
> > 
> > APIC error on CPU0: 02(08)
> > 
 
> I have seen the same problem on the same motherboard.  It appears to
> be a motherboard bug that 2.4 exposes and 2.2 doesn't.

This PRINT was added in 2.4. 

You're seeing noise on the apic lines. The APICs notice, but every now
and then you may see a lockup due to this. (i.e. if the corruption
does not trigger a parity error, because two bits flipped!)

				Roger. 

-- 
** R.E.Wolff@BitWizard.nl ** http://www.BitWizard.nl/ ** +31-15-2137555 **
*-- BitWizard writes Linux device drivers for any device you may have! --*
*       Common sense is the collection of                                *
******  prejudices acquired by age eighteen.   -- Albert Einstein ********
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: APIC errors w/ 2.4.0-test11-pre2

[H. Peter Anvin]

Rogier Wolff wrote:
> 
> > I have seen the same problem on the same motherboard.  It appears to
> > be a motherboard bug that 2.4 exposes and 2.2 doesn't.
> 
> This PRINT was added in 2.4.
> 
> You're seeing noise on the apic lines. The APICs notice, but every now
> and then you may see a lockup due to this. (i.e. if the corruption
> does not trigger a parity error, because two bits flipped!)
> 

What I was seeing, though, was that 2.4 would lock up with an NMI deadman
after a few days.  2.2 never did.  Even on other hardware I see very high
NMI rates, but no lockups, using 2.4.

	-hpa

-- 
<hpa@transmeta.com> at work, <hpa@zytor.com> in private!
"Unix gives you enough rope to shoot yourself in the foot."
http://www.zytor.com/~hpa/puzzle.txt
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/