http://bugme.osdl.org/show_bug.cgi?id=1056
Summary: oops during hwscan in 2.6.0-test2-bk7
Kernel Version: 2.6.0-test2-bk7
Status: NEW
Severity: blocking
Owner: [email protected]
Submitter: [email protected]
Distribution:SLES 8
Hardware Environment:8way PIII, 8GB mem, aic7xxx
Software Environment:
Problem Description: Get the following oops during hwscan init script at boot time.
Starting service at daemondone
Starting hardware scan on boot200xx?200200200200Oops: 0000 [#1]
CPU: 1
EIP: 0060:[<c02fed1e>] Not tainted
EFLAGS: 00010007
EIP is at as_find_arq_hash+0x1e/0xa0
eax: 00000027 ebx: e3d973a0 ecx: c6728a80 edx: 20323178
esi: c6724a00 edi: 00000010 ebp: 203232b0 esp: f72d5c00
ds: 007b es: 007b ss: 0068
Process hwscan (pid: 815, threadinfo=f72d4000 task=dedbad40)
Stack: e3d973a0 c6724a00 00000010 00000000 c0300334 c6728a80 00000008 00000101
ded470a0 c6728a80 00000000 c6724a00 00000008 00000000 c02f7809 c6724a00
f72d5c6c e3d973a0 c02fa9a8 c6724a00 f72d5c6c e3d973a0 00000008 00000000
Call Trace:
[<c0300334>] as_merge+0xe4/0x170
[<c02f7809>] elv_merge+0x29/0x30
[<c02fa9a8>] __make_request+0x2d8/0x4f0
[<c02faced>] generic_make_request+0x12d/0x1b0
[<c015bff7>] bio_alloc+0xd7/0x1c0
[<c02fadc4>] submit_bio+0x54/0xa0
[<c015a5b7>] block_read_full_page+0x227/0x2f0
[<c0139161>] add_to_page_cache+0x71/0x110
[<c013f254>] read_pages+0xf4/0x150
[<c015e2e0>] blkdev_get_block+0x0/0x60
[<c013d311>] __alloc_pages+0xa1/0x330
[<c013f6a0>] __do_page_cache_readahead+0xf0/0x143
[<c013f3e5>] page_cache_readahead+0x75/0x180
[<c01399d3>] do_generic_mapping_read+0xf3/0x420
[<c0139d00>] file_read_actor+0x0/0xf0
[<c0139fe4>] __generic_file_aio_read+0x1f4/0x240
[<c0139d00>] file_read_actor+0x0/0xf0
[<c01478ee>] do_wp_page+0x37e/0x3a0
[<c013a13e>] generic_file_read+0x8e/0xb0
[<c01694e0>] __pollwait+0x0/0xd0
[<c0169bfe>] sys_select+0x24e/0x520
[<c01569ee>] vfs_read+0xbe/0x130
[<c0156c92>] sys_read+0x42/0x70
[<c010afef>] syscall_call+0x7/0xb
Code: 8b 5d 00 39 eb 89 d9 74 3f 89 f6 8d bc 27 00 00 00 00 8d 53
Steps to reproduce:
Boot machine.
I've had problems with hwscan ever since I started testing 2.5 on a
Suse system. I ended up deactivating it. Doesn't make any sense if the hardware
doesn't change and you boot faster :-)
The kernel shouldn't bork though.
Unfortunately it would appear that the source for hwscan is not available.
I reported this to (German) Suse back in June.
Margit
On Fri, Aug 08 2003, Margit Schubert-While wrote:
> Unfortunately it would appear that the source for hwscan is not available.
Nonsense, it's on the discs.
--
Jens Axboe