BUG: block_read_full_page with NULL page->mapping

From: Yuri Pudgorodsky (yur@asplinux.ru)
Date: Thu Oct 26 2000 - 08:06:12 EST


Hi,

I got the following OOPS while doing some heavy compilations (make -j4) on dual SMP P-III,
2.4.0-test10-pre5 and reiserfs 3.6.18 for 2.4.0-test9.

It looks like page without page->mapping has been passed to
block_read_full_page().

Unable to handle kernel NULL pointer dereference at virtual address 00000010
*pde = 00000000
Oops: 0000
CPU: 1
EIP: 0010:[<c0133a96>]
Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00010282
eax: 00000000 ebx: 00000000 ecx: c0a49f24 edx: c1164f88
esi: c1164f88 edi: c28e913c ebp: 00000000 esp: c1cbbee4
ds: 0018 es: 0018 ss: 0018
Process bash (pid: 25783, stackpage=c1cbb000)
Stack: 00000000 c1164f88 c28e913c 00000000 00000001 c0710000 00000246 c1cbbf1c
       c0124e37 00000000 c1164f88 c28e913c 00000000 c1cbbf1c 01234567 c1cba000
       c09dbf24 c0a49f24 c01624a7 c1164f88 c015fe68 c0125776 c3d078c0 c1164f88
Call Trace: [<c0124e37>] [<c01624a7>] [<c015fe68>] [<c0125776>] [<c0125a6f>] [<c01259c0>] [<c0130ff7>]
       [<c010ac67>]
Code: 8b 40 10 89 44 24 24 c7 44 24 18 00 00 00 00 8b 42 18 a8 01

>>EIP; c0133a96 <block_read_full_page+e/1f4> <=====
Trace; c0124e37 <___wait_on_page+e3/ec>
Trace; c01624a7 <reiserfs_readpage+f/14>
Trace; c015fe68 <reiserfs_get_block+0/ddc>
Trace; c0125776 <do_generic_file_read+2ba/504>
Trace; c0125a6f <generic_file_read+5b/78>
Trace; c01259c0 <file_read_actor+0/54>
Trace; c0130ff7 <sys_read+8f/c4>
Trace; c010ac67 <system_call+33/38>
Code; c0133a96 <block_read_full_page+e/1f4>
00000000 <_EIP>:
Code; c0133a96 <block_read_full_page+e/1f4> <=====
   0: 8b 40 10 mov 0x10(%eax),%eax <=====
Code; c0133a99 <block_read_full_page+11/1f4>
   3: 89 44 24 24 mov %eax,0x24(%esp,1)
Code; c0133a9d <block_read_full_page+15/1f4>
   7: c7 44 24 18 00 00 00 movl $0x0,0x18(%esp,1)
Code; c0133aa4 <block_read_full_page+1c/1f4>
   e: 00
Code; c0133aa5 <block_read_full_page+1d/1f4>
   f: 8b 42 18 mov 0x18(%edx),%eax
Code; c0133aa8 <block_read_full_page+20/1f4>
  12: a8 01 test $0x1,%al

Hope it helps to fix some bugs..

Regards,
Yuri Pudgorodsky

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Tue Oct 31 2000 - 21:00:18 EST