linux 2.4.1 oops with arla-0.35.2

[Emil Egerfjord]

Hello all,

I get an "oops" when I try to run the program
"/afs/e.kth.se/usr/local/bin/mail-forward", run from within a "cvs ci "
command. The most remarkable thing is that it does not happen until the
second "cvs ci", i.e the first (since the system booted) attempt to send
the mail always succeeds, but the following will cause the "oops".

I guess it is arla-related since it referes to "[xfs]xfs_d_revalidate" in
the ksymoops, even if it seems to "sh" that triggered it.

I am running a Debian GNU/Linux unstable running Linux 2.4.1 with
arla version 0.35.2. This computer is running at home - not at the
"e.kth.se"-cell".

I get the following ksymoops:

--cut--
Mar  5 10:37:29 mille kernel: d084aee6
Mar  5 10:37:29 mille kernel: Oops: 0000
Mar  5 10:37:29 mille kernel: CPU:    0
Mar  5 10:37:29 mille kernel: EIP:    0010:[<d084aee6>]
Using defaults from ksymoops -t elf32-i386 -a i386
Mar  5 10:37:29 mille kernel: EFLAGS: 00010202
Mar  5 10:37:29 mille kernel: eax: c782ddf0   ebx: cf29f620   ecx: d084edbb   edx: c4783284
Mar  5 10:37:29 mille kernel: esi: 000408ce   edi: c782dde0   ebp: cede3e34   esp: cede3dac
Mar  5 10:37:29 mille kernel: ds: 0018   es: 0018   ss: 0018
Mar  5 10:37:29 mille kernel: Process sh (pid: 1191, stackpage=cede3000)
Mar  5 10:37:29 mille kernel: Stack: cf29f620 00000000 cf29fc20 cff8e300 cf18901c b083a1ba 0000000c c0138792 
Mar  5 10:37:29 mille kernel:        cf29f620 00000000 cede3e0c c0138f02 cf29fc20 cede3e0c 00000000 cede3e34 
Mar  5 10:37:29 mille kernel:        cf189000 00000000 cede3e34 be250b48 be8a7cda cede3e38 00000009 00000000 
Mar  5 10:37:29 mille kernel: Call Trace: [cached_lookup+46/84] [path_walk+1358/1980] [open_exec+42/200] [do_execve+30/488] [__alloc_pages_limit+138/168] [__alloc_pages+296/720] [do_wp_page+527/572] 
Mar  5 10:37:29 mille kernel: Code: f6 46 3c 10 74 1b f6 05 a0 fa 84 d0 10 74 e6 83 c4 f4 68 d5 

>>EIP; d084aee6 <[xfs]xfs_d_revalidate+fa/158>   <=====
Code;  d084aee6 <[xfs]xfs_d_revalidate+fa/158>
00000000 <_EIP>:
Code;  d084aee6 <[xfs]xfs_d_revalidate+fa/158>   <=====
   0:   f6 46 3c 10               testb  $0x10,0x3c(%esi)   <=====
Code;  d084aeea <[xfs]xfs_d_revalidate+fe/158>
   4:   74 1b                     je     21 <_EIP+0x21> d084af07 <[xfs]xfs_d_revalidate+11b/158>
Code;  d084aeec <[xfs]xfs_d_revalidate+100/158>
   6:   f6 05 a0 fa 84 d0 10      testb  $0x10,0xd084faa0
Code;  d084aef3 <[xfs]xfs_d_revalidate+107/158>
   d:   74 e6                     je     fffffff5 <_EIP+0xfffffff5> d084aedb <[xfs]xfs_d_revalidate+ef/158>
Code;  d084aef5 <[xfs]xfs_d_revalidate+109/158>
   f:   83 c4 f4                  add    $0xfffffff4,%esp
Code;  d084aef8 <[xfs]xfs_d_revalidate+10c/158>
  12:   68 d5 00 00 00            push   $0xd5
--cut--

More information may be available upon request.

Regards
/Emil Egerfjord
-- 
+----------------------------------+---------------------------------+
| Emil Egerfjord                   | All opinions in this letter are |
| Url: http://www.e.kth.se/~mille/ | mine, if not stated otherwise!  |
| Email: mille(at)e.kth.se         | /Emil Egerfjord                 |
+----------------------------------+---------------------------------+