On Mon, 24 Feb 1997, Ion Badulescu wrote:
> On Mon, 24 Feb 1997, Ion Badulescu wrote:
>
> > Really, folks, can you try out Bernd's code and see if it oopses on your
> > particular configuration? Please? :) Drop me a message if you got rid of
> > the message and I'll forward it to you.
>
> Follow-up on my own message:
>
> 1. the problem only occurs if core is dumped (my tests had been done with
> ulimit -c0 initially)
Ditto - using 2.0.29 with coredumpsize of zero, the program just
segfaulted. Unlimiting this gets the following oops:
Feb 25 12:02:41 bison kernel: Unable to handle kernel paging request at
virtual address 40008000
Feb 25 12:02:41 bison kernel: current->tss.cr3 = 0045f000, Pr3 = 0045f000
Feb 25 12:02:41 bison kernel: *pde = 00551067
Feb 25 12:02:41 bison kernel: *pte = 00000000
Feb 25 12:02:41 bison kernel: Oops: 0000
Feb 25 12:02:41 bison kernel: CPU: 0
Feb 25 12:02:41 bison kernel: EIP: 0010:[ext2_file_write+585/1116]
Feb 25 12:02:41 bison kernel: EFLAGS: 00010216
Feb 25 12:02:41 bison kernel: eax: 00522088 ebx: 00000400
ecx: 00000100 edx: 00747400
Feb 25 12:02:41 bison kernel: esi: 40008000 edi: 00747400
ebp: 00000400 esp: 004c6c08
Feb 25 12:02:41 bison kernel: ds: 0018 es: 0018 fs: 002b
gs: 002b ss: 0018
Feb 25 12:02:41 bison kernel: Process crashtest (pid: 2833, process nr:41,
stackpage=004c6000)
Feb 25 12:02:41 bison kernel: Stack: 00001000 40008000 40008000 00040000
00000000 00000000 001a8950 00000000
Feb 25 12:02:41 bison kernel: 00000000 00000014 00005000 00000000
00522088 ffffffe4 00dfc9b000563088
Feb 25 12:02:41 bison kernel: 004c6c94 008b19b0 00008180 004c6c94
008b19b0 00008180 0012458200563088
Feb 25 12:02:41 bison kernel: Call Trace:
[__brelse+34/68] [ext2_create+338/360]
[dump_write+28/44] [writenote+167/200]
[dump_write+28/44] [elf_core_dump+2488/2640]
[do_no_page+259/808]
Feb 25 12:02:41 bison kernel: [do_no_page+391/808] [do_no_page+0/808]
[do_signal+495/632] [signal_return+18/64]
Feb 25 12:02:41 bison kernel: Code: 64 f3 a5 83 e3 03 89 d9 64 f3 a4 55
8b 54 24 34 8b 52 24 03
Disassembly of code:
0x804946c <str>: repz movsl %ds:(%esi),%es:(%edi)
0x804946f <str+3>: andl $0x3,%ebx
0x8049472 <str+6>: movl %ebx,%ecx
0x8049474 <str+8>: repz movsb %ds:(%esi),%es:(%edi)
0x8049477 <str+11>: pushl %ebp
0x8049478 <str+12>: movl 0x34(%esp,1),%edx
0x804947c <str+16>: movl 0x24(%edx),%edx
0x804947f <str+19>: addl (%eax),%eax
0x8049481 <str+21>: addb %al,(%eax)
0x8049483 <str+23>: addb %bh,%bh
This is with kernel 2.0.29 with the ldt patch posted a couple of weeks
ago, together with the pc-speaker patch. Kernel was compiled with gcc
2.7.2.1, libc-5.4.23, binutils 2.7.0.3, ld.so 1.8.9, for a pentium. On
this machine, I've*never* had a signal 11 while compiling or other problem
related to bad memory, so I doubt if that's the cause.
Chris.
-----BEGIN PGP SIGNATURE-----
Version: 2.6.2i
Comment: PGP Signed with PineSign 2.0
iQCVAwUBMxLYx2ZVEN0KDxVBAQEStgQApFVGi7YG9Ze1J0ZXNB2yNWcInIKtXc/m
lz47/sOHvnXSB7Z6VTUF8GjrM3Ci1agrBTjOPTpJuqsvmNmXIhwkqz+MRXk/oE9d
rbqJgU6nNQiryR410wYT8t5f+mPtj4j/RKV1fExnmCLW+txPPaj8sSZsNolfcsEW
2Jys1rO0Hjs=
=4fSf
-----END PGP SIGNATURE-----