I compiled 2.1.25 yesterday without a problem. Booted it today and it
seemed to work ok for the first two hours. After which
Feb 3 21:34:53 channely kernel:
do_wp_page: bogus page-middle at address 080afd24 (42b06700)
zap_pte_range: bad pmd (42b06700)
zap_pte_range: bad pmd (ad806700)
zap_pte_range: bad pmd (e9e06700)
appeared in the logs, followed by
Feb 3 21:34:58 channely kernel:
Unable to handle kernel NULL pointer dereference at virtual address 00000080
current->tss.cr3 = 02efa000,
*pde = 00000000
Oops: 0000
CPU: 0
EIP: 0010:[<c0125b8e>]
EFLAGS: 00010246
eax: 00001800 ebx: 00000000 ecx: c2a89618 edx: 00000c00
esi: 00000674 edi: 00000000 ebp: bffffe3c esp: c2efef84
ds: 0018 es: 0018 ss: 0018
Process update (pid: 7, process nr: 7, stackpage=c2efe000)
Stack: c2f91c0c 00000100 00000000 c0128a7b 00000000 00000000 c2f91c0c 00000100
00000000 bffffe3c ffffffff c0128b85 c2f91c0c 00000100 bffffe48 c010a468
00000001 00000000 00000000 00000100 bffffe48 bffffe3c 00000086 0000002b
Call Trace: [<c0128a7b>] [<c0128b85>] [<c010a468>]
Code: 80 bb 80 00 00 00 00 74 09 53 e8 e7 05 00 00 83 c4 04 80 bb
After which loads of other followed (I can send the logs if anyone wants
them).
Using `/usr/src/linux-2.1/System.map' to map addresses to symbols.
>>EIP: c0125b8e <sync_inodes+2a/64>
Trace: c0128a7b <sync_old_buffers+13/ec>
Trace: c0128b85 <sys_bdflush+31/a4>
Trace: c010a468 <system_call+38/40>
Code: c0125b8e <sync_inodes+2a/64> cmpb $0x0,0x80(%ebx)
Code: c0125b95 <sync_inodes+31/64> je c0125ba0 <sync_inodes+3c/64>
Code: c0125b97 <sync_inodes+33/64> pushl %ebx
Code: c0125b98 <sync_inodes+34/64> call c0126184 <__wait_on_inode>
Code: c0125b9d <sync_inodes+39/64> addl $0x4,%esp
Code: c0125ba0 <sync_inodes+3c/64>
This is 2.1.25 without any further patches. It started while PPP-ing. I
hope this is somebody can use this.
--Rafael D'Halleweyn rdhall@dma.be
"Style is mighty in the shower." q+>