Re: Kernel Oops in CVS with NI PCI-6052E

On 2004/07/15 09:21, Eran Tromer wrote:

> With the current Comedi and Comedilib CVS snapshot, compilation and
> installation go smoothly. However, comedi_test segfaults and causes a
> kernel oops. 
> Call Trace:
>   [<02140096>] get_user_size+0x2e/0x55
>   [<16a71ff6>] do_insn_ioctl+0x86/0xd0 [comedi]
>   [<0214f7f6>] sys_ioctl+0x1f2/0x224
>   [<02142236>] sys_write+0x2c/0x42

Update: running qtscope also consistently causes an oops, at a different
code point.

Oops: 0000 [#1]
Modules linked in: ni_pcimio comedi_fc mite 8255 comedi sch_ingress
cls_u32 cls_fw sch_sfq sch_htb ppp_deflate zlib_deflate ppp_async
ip_conntrack_ftp ipt_length ipt_mark ipt_MARK iptable_mangle
ipt_REDIRECT ipt_MASQUERADE iptable_nat ipt_TCPMSS ipt_owner snd_pcm_oss
snd_mixer_oss binfmt_misc ipt_state ip_conntrack ipt_REJECT ipt_LOG
ipt_limit iptable_filter ip_tables ppp_generic slhc snd_seq_midi
snd_seq_midi_event snd_seq snd_ens1370 snd_ak4531_codec snd_ymfpci
snd_ac97_codec snd_pcm snd_opl3_lib snd_timer snd_hwdep snd_page_alloc
gameport snd_mpu401_uart snd_rawmidi snd_seq_device snd soundcore nfsd
exportfs lockd ipv6 parport_pc lp parport ds yenta_socket pcmcia_core
sunrpc autofs4 ne2k_pci 8390 3c59x microcode ext3 jbd dm_mod uhci_hcd
CPU:    0
EIP:    0060:[<16d35966>]    Not tainted
EFLAGS: 00010283   (2.6.6-1.435.2.3)
EIP is at ni_ai_munge+0x26/0x50 [ni_pcimio]
eax: 070ca000   ebx: 000077e8   ecx: 00000000   edx: 15c10000
esi: 7ffffff0   edi: 16d41030   ebp: 032e8238   esp: 0fab9f0c
ds: 007b   es: 007b   ss: 0068
Process qtscope (pid: 6118, threadinfo=0fab9000 task=0984c7d0)
Stack: 070ca000 ffffffe0 053d7974 032e8238 ffffffe0 16d24181 ffffffe0
00000008
       fef9f510 16d35940 00000000 070ca000 032e8238 053d7974 070ca000
fef9f510
       16d219b8 00000000 00000000 00000000 00001010 00000020 00001010
00000020
Call Trace:
 [<16d24181>] comedi_buf_munge+0x91/0x100 [comedi]
 [<16d35940>] ni_ai_munge+0x0/0x50 [ni_pcimio]
 [<16d219b8>] do_bufinfo_ioctl+0xf8/0x1a0 [comedi]
 [<0214f7f6>] sys_ioctl+0x1f2/0x224
 [<021421f4>] sys_read+0x2c/0x42

Code: 0f b7 04 5f 0f b7 94 4a 36 01 00 00 41 01 d0 66 89 04 5f 31

And ksymoops says:

>>EIP; 16d35966 <END_OF_CODE+149ac54a/????>   <=====

Trace; 16d24181 <END_OF_CODE+1499ad65/????>
Trace; 16d35940 <END_OF_CODE+149ac524/????>
Trace; 16d219b8 <END_OF_CODE+1499859c/????>
Trace; 0214f7f6 <kill_fasync+32a/35c>
Trace; 021421f4 <vfs_write+110/22c>

Code;  16d35966 <END_OF_CODE+149ac54a/????>
00000000 <_EIP>:
Code;  16d35966 <END_OF_CODE+149ac54a/????>   <=====
   0:   0f b7 04 5f               movzwl (%edi,%ebx,2),%eax   <=====
Code;  16d3596a <END_OF_CODE+149ac54e/????>
   4:   0f b7 94 4a 36 01 00      movzwl 0x136(%edx,%ecx,2),%edx
Code;  16d35971 <END_OF_CODE+149ac555/????>
   b:   00
Code;  16d35972 <END_OF_CODE+149ac556/????>
   c:   41                        inc    %ecx
Code;  16d35973 <END_OF_CODE+149ac557/????>
   d:   01 d0                     add    %edx,%eax
Code;  16d35975 <END_OF_CODE+149ac559/????>
   f:   66 89 04 5f               mov    %ax,(%edi,%ebx,2)
Code;  16d35979 <END_OF_CODE+149ac55d/????>
  13:   31 00                     xor    %eax,(%eax)


That's running current CVS.

  Eran

Received on 2004-07-16Z17:46:51