* 2.5.34-mm2 kernel BUG at sched.c:944! only with CONFIG_PREEMPT=y
@ 2002-09-12 14:14 Steven Cole
2002-09-12 19:46 ` Steven Cole
2002-09-13 20:41 ` Arador
0 siblings, 2 replies; 4+ messages in thread
From: Steven Cole @ 2002-09-12 14:14 UTC (permalink / raw)
To: Andrew Morton; +Cc: Linux Kernel, linux-mm
I got the following BUG at sched.c:944! with 2.5.34-mm2 and PREEMPT on.
This was repeatable.
With no PREEMPT, 2.5.34-mm2 booted and is running fine. Some other
options used: SMP, HUGETLB_PAGE, HIGHPTE, HIGHMEM4G.
System is dual p3, scsi, 1GB.
Steven
ksymoops 2.4.4 on i686 2.5.34. Options used
-v vmlinux (specified)
-K (specified)
-L (specified)
-O (specified)
-m System.map (specified)
kernel BUG at sched.c:944!
invalid operand: 0000
CPU: 0
EIP: 0060:[<c01176ff>] Not tainted
Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00010206
eax: c02d4000 ebx: c02d4000 ecx: 00000000 edx: 00000000
esi: 0009b800 edi: c0105000 ebp: c02d5c8 esp: c02d5fa8
ds: 068 es: 0068 ss: 0068
Stack: c01072c4 00000060 00000286 00000000 00000000 c02d4000 0009b800 c0105000
c02d5fd4 c0117ad6 00000000 0008e000 c010504b c02d68c2 c02ba3a0 00000000
c027e980 0003fff0 0003fff0 c033e660 00000002 c01001b1
Call Trace: [<c01072c4>] [<c0105000>] [<c0117ad6>] [<c010504b>]
Code: 0f 0b b0 03 5f 52 28 c0 b9 00 e0 ff ff 21 e1 ff 41 10 9b 01
>>EIP; c01176ff <schedule+1f/3c0> <=====
Trace; c01072c4 <kernel_thread_helper+0/c>
Trace; c0105000 <_stext+0/0>
Trace; c0117ad6 <preempt_schedule+36/50>
Trace; c010504b <rest_init+4b/50>
Code; c01176ff <schedule+1f/3c0>
00000000 <_EIP>:
Code; c01176ff <schedule+1f/3c0> <=====
0: 0f 0b ud2a <=====
Code; c0117701 <schedule+21/3c0>
2: b0 03 mov $0x3,%al
Code; c0117703 <schedule+23/3c0>
4: 5f pop %edi
Code; c0117704 <schedule+24/3c0>
5: 52 push %edx
Code; c0117705 <schedule+25/3c0>
6: 28 c0 sub %al,%al
Code; c0117707 <schedule+27/3c0>
8: b9 00 e0 ff ff mov $0xffffe000,%ecx
Code; c011770c <schedule+2c/3c0>
d: 21 e1 and %esp,%ecx
Code; c011770e <schedule+2e/3c0>
f: ff 41 10 incl 0x10(%ecx)
Code; c0117711 <schedule+31/3c0>
12: 9b fwait
Code; c0117712 <schedule+32/3c0>
13: 01 00 add %eax,(%eax)
<0>Kernel panic: Attempted to kill the idle task!
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: 2.5.34-mm2 kernel BUG at sched.c:944! only with CONFIG_PREEMPT=y
2002-09-12 14:14 2.5.34-mm2 kernel BUG at sched.c:944! only with CONFIG_PREEMPT=y Steven Cole
@ 2002-09-12 19:46 ` Steven Cole
2002-09-13 20:41 ` Arador
1 sibling, 0 replies; 4+ messages in thread
From: Steven Cole @ 2002-09-12 19:46 UTC (permalink / raw)
To: Andrew Morton; +Cc: Linux Kernel
On Thu, 2002-09-12 at 08:14, Steven Cole wrote:
> I got the following BUG at sched.c:944! with 2.5.34-mm2 and PREEMPT on.
> This was repeatable.
>
> With no PREEMPT, 2.5.34-mm2 booted and is running fine. Some other
> options used: SMP, HUGETLB_PAGE, HIGHPTE, HIGHMEM4G.
The above also occurred with no SMP.
I backed out Changeset 1.606 which did this in kernel/sched.c:
- if (unlikely(in_interrupt()))
+ if (unlikely(in_atomic()))
and 2.5.34-mm2 was able to boot with CONFIG_PREEMPT=y.
Obviously, this is not a fix of any kind, but shows that the scheduler
may be being called with a preemption lock.
Steven
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: 2.5.34-mm2 kernel BUG at sched.c:944! only with CONFIG_PREEMPT=y
2002-09-12 14:14 2.5.34-mm2 kernel BUG at sched.c:944! only with CONFIG_PREEMPT=y Steven Cole
2002-09-12 19:46 ` Steven Cole
@ 2002-09-13 20:41 ` Arador
2002-09-13 21:20 ` Steven Cole
1 sibling, 1 reply; 4+ messages in thread
From: Arador @ 2002-09-13 20:41 UTC (permalink / raw)
To: Steven Cole; +Cc: akpm, linux-kernel, linux-mm
On 12 Sep 2002 08:14:01 -0600
Steven Cole <elenstev@mesatop.com> escribió:
> I got the following BUG at sched.c:944! with 2.5.34-mm2 and PREEMPT on.
> This was repeatable.
Same for me:
POSIX conformance testing by UNIFIX
Kernel BUG at sched.c:944!
preempt, no smp, HUGETLB_PAGE
a cyrix 6x86MX 233+ with 32 MB of ram...
>
> With no PREEMPT, 2.5.34-mm2 booted and is running fine. Some other
> options used: SMP, HUGETLB_PAGE, HIGHPTE, HIGHMEM4G.
>
> System is dual p3, scsi, 1GB.
>
> Steven
>
> ksymoops 2.4.4 on i686 2.5.34. Options used
> -v vmlinux (specified)
> -K (specified)
> -L (specified)
> -O (specified)
> -m System.map (specified)
>
> kernel BUG at sched.c:944!
> invalid operand: 0000
> CPU: 0
> EIP: 0060:[<c01176ff>] Not tainted
> Using defaults from ksymoops -t elf32-i386 -a i386
> EFLAGS: 00010206
> eax: c02d4000 ebx: c02d4000 ecx: 00000000 edx: 00000000
> esi: 0009b800 edi: c0105000 ebp: c02d5c8 esp: c02d5fa8
> ds: 068 es: 0068 ss: 0068
> Stack: c01072c4 00000060 00000286 00000000 00000000 c02d4000 0009b800 c0105000
> c02d5fd4 c0117ad6 00000000 0008e000 c010504b c02d68c2 c02ba3a0 00000000
> c027e980 0003fff0 0003fff0 c033e660 00000002 c01001b1
> Call Trace: [<c01072c4>] [<c0105000>] [<c0117ad6>] [<c010504b>]
> Code: 0f 0b b0 03 5f 52 28 c0 b9 00 e0 ff ff 21 e1 ff 41 10 9b 01
>
> >>EIP; c01176ff <schedule+1f/3c0> <=====
> Trace; c01072c4 <kernel_thread_helper+0/c>
> Trace; c0105000 <_stext+0/0>
> Trace; c0117ad6 <preempt_schedule+36/50>
> Trace; c010504b <rest_init+4b/50>
> Code; c01176ff <schedule+1f/3c0>
> 00000000 <_EIP>:
> Code; c01176ff <schedule+1f/3c0> <=====
> 0: 0f 0b ud2a <=====
> Code; c0117701 <schedule+21/3c0>
> 2: b0 03 mov $0x3,%al
> Code; c0117703 <schedule+23/3c0>
> 4: 5f pop %edi
> Code; c0117704 <schedule+24/3c0>
> 5: 52 push %edx
> Code; c0117705 <schedule+25/3c0>
> 6: 28 c0 sub %al,%al
> Code; c0117707 <schedule+27/3c0>
> 8: b9 00 e0 ff ff mov $0xffffe000,%ecx
> Code; c011770c <schedule+2c/3c0>
> d: 21 e1 and %esp,%ecx
> Code; c011770e <schedule+2e/3c0>
> f: ff 41 10 incl 0x10(%ecx)
> Code; c0117711 <schedule+31/3c0>
> 12: 9b fwait
> Code; c0117712 <schedule+32/3c0>
> 13: 01 00 add %eax,(%eax)
>
> <0>Kernel panic: Attempted to kill the idle task!
>
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: 2.5.34-mm2 kernel BUG at sched.c:944! only with CONFIG_PREEMPT=y
2002-09-13 20:41 ` Arador
@ 2002-09-13 21:20 ` Steven Cole
0 siblings, 0 replies; 4+ messages in thread
From: Steven Cole @ 2002-09-13 21:20 UTC (permalink / raw)
To: Arador; +Cc: Andrew Morton, linux-kernel, linux-mm
On Fri, 2002-09-13 at 14:41, Arador wrote:
> On 12 Sep 2002 08:14:01 -0600
> Steven Cole <elenstev@mesatop.com> escribió:
>
> > I got the following BUG at sched.c:944! with 2.5.34-mm2 and PREEMPT on.
> > This was repeatable.
>
> Same for me:
> POSIX conformance testing by UNIFIX
> Kernel BUG at sched.c:944!
>
If you just want a quick workaround (not a real fix), you
can change in_atomic back to in_interrupt on line 933 of kernel/sched.c.
Robert Love has posted another patch for this, so you might want to
try that out. Here is a link to that post:
http://marc.theaimsgroup.com/?l=linux-kernel&m=103190275327089&w=2
You may need to change KERN_ERROR to KERN_ERR in that patch.
Steven
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2002-09-13 21:18 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2002-09-12 14:14 2.5.34-mm2 kernel BUG at sched.c:944! only with CONFIG_PREEMPT=y Steven Cole
2002-09-12 19:46 ` Steven Cole
2002-09-13 20:41 ` Arador
2002-09-13 21:20 ` Steven Cole
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).