Kernel BUG!

[Axel Thimm]

On Mon, Sep 27, 2004 at 10:41:34AM +0200, GianPiero Puccioni wrote:
> Hi,
> 
> I have ten similar machines and time to time after days of uptime I find
> one of them frozen and there is  nothing to do but switch it off and
> on to restart it. ALl are dual Xeon with FC2 kernel  2.6.8-1.521smp
> and Nvidia card with Nvidia driver 6111.
>  
> A very similar thing happened to another one a week ago but it just
> rebooted. I don't know if this could be the problem that sometimes
> stopped the others but it's a problem anyway. By the way, the random
> freezes happened when I was using the Xorg nvidia driver before 6111 so
> I don't think it's that.
> 
> Any thought?

The BUG below is clearly triggered by the nvidia modules, which cannot
be debugged by Fedora developers.

If you believe you can reproduce this bug on a machine w/o any
external modules, i.e. using only xorg's nv driver (also comment the
nvidia kernel module out of /etc/modprobe.conf, your kernel needs to
report that it is not tainted), then please bugzilla and post a
message here.

> This morning one of them had X frozen and I found this in the log 
>  
> Sep 24 12:56:35 septem kernel: ------------[ cut here ]------------
> Sep 24 12:56:35 septem kernel: kernel BUG at mm/page_alloc.c:792!
> Sep 24 12:56:35 septem kernel: invalid operand: 0000 [#1]
> Sep 24 12:56:35 septem kernel: SMP 
> Sep 24 12:56:35 septem kernel: Modules linked in: snd_pcm_oss snd_mixer_oss snd_intel8x0 snd_ac97_codec snd_pcm snd_timer snd_page_alloc gameport snd_mpu401_uart snd_rawmidi snd_seq_device snd soundcore nvidia(U) nfs lockd parport_pc lp parport autofs4 sunrpc e1000 floppy sg scsi_mod microcode dm_mod uhci_hcd ehci_hcd button battery asus_acpi ac ext3 jbd
> Sep 24 12:56:35 septem kernel: CPU:    2
> Sep 24 12:56:35 septem kernel: EIP:    0060:[<02140502>]    Tainted: P  
> Sep 24 12:56:35 septem kernel: EFLAGS: 00013256   (2.6.8-1.521smp) 
> Sep 24 12:56:35 septem kernel: EIP is at __free_pages+0x20/0x4c
> Sep 24 12:56:35 septem kernel: eax: 00000000   ebx: 00000000   ecx: 03640240   edx: 03640240
> Sep 24 12:56:35 septem kernel: esi: 39710cf0   edi: 00000010   ebp: 00000000   esp: 39710ce8
> Sep 24 12:56:35 septem kernel: ds: 007b   es: 007b   ss: 0068
> Sep 24 12:56:35 septem kernel: Process X (pid: 2344, threadinfo=39710000 task=37e998b0)
> Sep 24 12:56:35 septem kernel: Stack: 0368a938 0211a5a9 03368378 0360c258 03b4d100 00000010 03b4d100 00000010 
> Sep 24 12:56:35 septem kernel:        022080c6 00000010 00000010 39710d60 39710ed8 43238d1c 434f3e20 00000010 
> Sep 24 12:56:35 septem kernel:        39710d60 41705bc0 00000010 00000041 41706ec0 434f3e20 432375fe 39710e80 
> Sep 24 12:56:35 septem kernel: Call Trace:
> Sep 24 12:56:35 septem kernel:  [<0211a5a9>] global_flush_tlb+0xe9/0xf3
> Sep 24 12:56:35 septem kernel:  [<022080c6>] agp_allocate_memory+0x8c/0x93
> Sep 24 12:56:35 septem kernel:  [<43238d1c>] KernAllocAGPPages+0x6c/0x167 [nvidia]
> Sep 24 12:56:35 septem kernel:  [<432375fe>] nv_alloc_pages+0xb9/0x310 [nvidia]
> Sep 24 12:56:35 septem kernel:  [<43096a5b>] _nv004819rm+0x4b/0x88 [nvidia]
> Sep 24 12:56:35 septem kernel:  [<43078882>] _nv001629rm+0x6e/0x78 [nvidia]
> Sep 24 12:56:35 septem kernel:  [<43078841>] _nv001629rm+0x2d/0x78 [nvidia]
> Sep 24 12:56:35 septem kernel:  [<4305b8d2>] _nv006380rm+0x16/0x3c [nvidia]
> Sep 24 12:56:35 septem kernel:  [<4306d1ca>] _nv001734rm+0x36/0xe0 [nvidia]
> Sep 24 12:56:35 septem kernel:  [<43062799>] _nv001226rm+0x121/0x138 [nvidia]
> Sep 24 12:56:35 septem kernel:  [<4305ba3e>] _nv006349rm+0xaa/0xd8 [nvidia]
> Sep 24 12:56:35 septem kernel:  [<4307fe9d>] _nv002132rm+0x31/0x48 [nvidia]
> Sep 24 12:56:35 septem kernel:  [<43062552>] _nv001231rm+0x1ce/0x2f4 [nvidia]
> Sep 24 12:56:35 septem kernel:  [<4306dbe6>] _nv001659rm+0x26/0x2c [nvidia]
> Sep 24 12:56:35 septem kernel:  [<4307872e>] _nv001633rm+0x12/0x18 [nvidia]
> Sep 24 12:56:35 septem kernel:  [<4305f0fa>] _nv003453rm+0x86/0xd8 [nvidia]
> Sep 24 12:56:35 septem kernel:  [<4307e1c5>] rm_change_res_mode+0x69/0x8c [nvidia]
> Sep 24 12:56:35 septem kernel:  [<4307eefb>] _nv001139rm+0x16b/0x4b8 [nvidia]
> Sep 24 12:56:35 septem kernel:  [<02158793>] rw_vm+0x2df/0x331
> Sep 24 12:56:35 septem kernel:  [<4307e147>] rm_ioctl+0x23/0x38 [nvidia]
> Sep 24 12:56:35 septem kernel:  [<43236514>] nv_kern_ioctl+0x31c/0x367 [nvidia]
> Sep 24 12:56:35 septem kernel:  [<0216d078>] sys_ioctl+0x23d/0x2a0
> Sep 24 12:56:35 septem kernel:  [<0215a97b>] filp_close+0x59/0x5f
> Sep 24 12:56:35 septem kernel:  [<0215aa21>] sys_close+0xa0/0xd3
> Sep 24 12:56:35 septem kernel: Code: 0f 0b 18 03 01 f6 2e 02 f0 83 41 04 ff 0f 98 c0 84 c0 74 16 
> Sep 24 12:57:50 septem kdm: :0[31429]: Cannot connect to :0, giving up
> Sep 24 12:57:50 septem kdm[2192]: Display :0 cannot be opened
> Sep 24 12:57:50 septem kdm[2192]: Unable to fire up local display :0; disabling.
> 
> 
> 
> Ciao,
>   GianPiero

-- 
Axel.Thimm at ATrpms.net
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/fedora-list/attachments/20040927/32531238/attachment-0001.sig>