Re: 2.6.9-rc4-mm1

[Mathieu Segaud <matt@minas-morgul.org>]

William Wolf <wwolf@vt.edu> disait dernièrement que :

> Dmitry Torokhov wrote:
>> On Thursday 14 October 2004 03:56 pm, William Wolf wrote:
>>
>>>Hey, I just tried -rc4-mm1 on my amd64 laptop, and my keyboard fails to
>>> work, I don't even think it is being recognized.
>> Could you try booting with i8042.noacpi and if it helps mailing me
>> your
>> /proc/acpi/dsdt?
>> Thanks!
>>
>
>
> Thanks a bunch, this fixed the keyboard problem.  What did this do 
> exactly?  Anyway, now I'm having more issues with this kernel.
>
> I'm getting a lot of scheduling while atomic, and not tainted messages. 
>    There are no such messages right after boot or after log in, it only 
> happens when i try to launch certain applications.  gaim, thunderbird, 
> and firefox are some that i tried, all just dying and dumping these 
> types of messages:
>
>
>
>
> Unable to handle kernel paging request at 00000000000c1644 RIP:
> <ffffffff8013564d>{profile_hit+45}
> PML4 16557067 PGD 16b76067 PMD 0
> Oops: 0002 [5] PREEMPT
> CPU 0
> Modules linked in: usbcore snd_intel8x0 snd_ac97_codec snd_pcm snd_timer 
> snd snd_page_alloc evdev sis900 sg sd_mod scsi_mod
> Pid: 7329, comm: firefox-bin Not tainted 2.6.9-rc4-mm1
> RIP: 0010:[<ffffffff8013564d>] <ffffffff8013564d>{profile_hit+45}
> RSP: 0018:0000010016a6df20  EFLAGS: 00010082
> RAX: 0000000000000000 RBX: 0000010016a7f860 RCX: 0000000000000000
> RDX: 0000010016a7f998 RSI: 00000000000c1644 RDI: 0000000000000002
> RBP: 0000010016a6df68 R08: 0000000000000001 R09: 0000000000000001
> R10: 0000002a95d67000 R11: 0000000000000206 R12: 0000000000000000
> R13: 00000000ffffffea R14: 0000007fbfffefb4 R15: ffffffff804543e0
> FS:  000000000051e4a0(005b) GS:ffffffff80497840(0000) knlGS:0000000000000000
> CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
> CR2: 00000000000c1644 CR3: 0000000000101000 CR4: 00000000000006e0
> Process firefox-bin (pid: 7329, threadinfo 0000010016a6c000, task 
> 0000010016a7f860)
> Stack: ffffffff801312f6 000000000055c570 0000000000000006 00000000bffff680
>        0000002a95e7c820 0000000000004000 0000000000000000 0000007fbfffefb4
>        0000002a97775e18 0000010016a6df78
> Call Trace:<ffffffff801312f6>{setscheduler+214} 
> <ffffffff801314a9>{sys_sched_setscheduler+9}
>        <ffffffff80110496>{system_call+126}

known issue
try doing this:
cd /path/to/your/kernel/source
wget ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.9-rc4/2.6.9-rc4-mm1/broken-out/optimize-profile-path-slightly.patch
patch -1 -R -i optimize-profile-path-slightly.patch

and the oops above should disappear.

for the other logs, I don't know where they come from, right now.
They may come from the "bad state" the kernel is left in by the previous
oops...

[snip]


> Unable to handle kernel paging request at 00000000000c1644 RIP:
> <ffffffff8013564d>{profile_hit+45}
> PML4 1675a067 PGD 167ae067 PMD 0
> Oops: 0002 [6] PREEMPT
> CPU 0
> Modules linked in: usbcore snd_intel8x0 snd_ac97_codec snd_pcm snd_timer 
> snd snd_page_alloc evdev sis900 sg sd_mod scsi_mod
> Pid: 7353, comm: mozilla-xremote Not tainted 2.6.9-rc4-mm1
> RIP: 0010:[<ffffffff8013564d>] <ffffffff8013564d>{profile_hit+45}
> RSP: 0018:00000100167d7f20  EFLAGS: 00010082
> RAX: 0000000000000000 RBX: 00000100165091d0 RCX: 0000000000000000
> RDX: 0000010016509308 RSI: 00000000000c1644 RDI: 0000000000000002
> RBP: 00000100167d7f68 R08: 0000000000000001 R09: 0000000000000001
> R10: 0000002a958744a0 R11: 0000000000000206 R12: 0000000000000000
> R13: 00000000ffffffea R14: 0000007fbffff4c4 R15: ffffffff804543e0
> FS:  00000000005034a0(005b) GS:ffffffff80497840(0000) knlGS:0000000000000000
> CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
> CR2: 00000000000c1644 CR3: 0000000000101000 CR4: 00000000000006e0
> Process mozilla-xremote (pid: 7353, threadinfo 00000100167d6000, task 
> 00000100165091d0)
> Stack: ffffffff801312f6 0000000000504770 0000000000000002 0000000000402360
>        0000002a95ad8820 0000000000004000 0000000000000000 0000007fbffff4c4
>        0000002a964dfe18 00000100167d7f78
> Call Trace:<ffffffff801312f6>{setscheduler+214} 
> <ffffffff801314a9>{sys_sched_setscheduler+9}
>        <ffffffff80110496>{system_call+126}

this one should be fixed, as the first one

>
> Code: ff 06 c3 48 83 ec 08 39 3d 86 49 32 00 75 54 48 83 3d 64 49
> RIP <ffffffff8013564d>{profile_hit+45} RSP <00000100167d7f20>
> CR2: 00000000000c1644
>  <6>note: mozilla-xremote[7353] exited with preempt_count 2
> scheduling while atomic: mozilla-xremote/0x04000002/7353
>
> Call Trace:<ffffffff803358e7>{schedule+119} 
> <ffffffff80130e34>{__wake_up+84}
>        <ffffffff803364cf>{cond_resched+63} 
> <ffffffff801633af>{unmap_vmas+1519}
>        <ffffffff80168316>{exit_mmap+118} <ffffffff801322a0>{mmput+48}
>        <ffffffff801369f3>{do_exit+483} 
> <ffffffff80259dc7>{do_unblank_screen+119}
>        <ffffffff801212df>{do_page_fault+1263} 
> <ffffffff80164afd>{do_no_page+1341}
>        <ffffffff80164cc6>{handle_mm_fault+326} 
> <ffffffff80110d49>{error_exit+0}
>        <ffffffff8013564d>{profile_hit+45} 
> <ffffffff801312f6>{setscheduler+214}
>        <ffffffff801314a9>{sys_sched_setscheduler+9} 
> <ffffffff80110496>{system_call+126}

as for the second kernel message, this may be a "consequence" of the
previous one

[snip]

I guess that all 'scheduling while atomic' messages were consequences of
the oopses occured just before them.

-- 
All the authors of the linux kernel? That would be more difficult than
herding Debian developers.

	- Martijn van Oosterhout on linux-kernel