Hardware: Sun v20z SMP box. Cached on serial console: Slab corruption: (Not tainted) start=ffff8100d452d548, len=168 Redzone: 0x5a2cf071/0x5a2cf071. Last user: [<ffffffff80300547>](cfq_free_io_context+0x2f/0x74) Call Trace: <ffffffff8027bdb5>{check_poison_obj+118} <ffffffff8025f0c4>{mempool_alloc+66} <ffffffff8030278b>{cfq_set_request+926} <ffffffff8027bf19>{cache_alloc_debugcheck_after+48} <ffffffff8030278b>{cfq_set_request+926} <ffffffff8027c13c>{kmem_cache_alloc+185} <ffffffff8030278b>{cfq_set_request+926} <ffffffff802fa532>{get_request+412} <ffffffff802fa724>{get_request_wait+31} <ffffffff80301395>{cfq_add_crq_rb+190} <ffffffff80301561>{cfq_insert_request+198} <ffffffff8025f0c4>{mempool_alloc+66} <ffffffff802fabe2>{__make_request+882} <ffffffff8025f0c4>{mempool_alloc+66} <ffffffff802f828a>{generic_make_request+545} <ffffffff80286c50>{__bio_clone+113} <ffffffff881330c3>{:raid1:make_request+396} <ffffffff8027ba76>{poison_obj+38} <ffffffff8025f0c4>{mempool_alloc+66} <ffffffff8027c01c>{cache_alloc_debugcheck_after+307} <ffffffff8025f0c4>{mempool_alloc+66} <ffffffff802f828a>{generic_make_request+545} <ffffffff802f72ce>{blk_recount_segments+126} <ffffffff88141c57>{:dm_mod:__map_bio+199} <ffffffff88141160>{:dm_mod:clone_bio+37} <ffffffff88141e0a>{:dm_mod:__split_bio+379} <ffffffff881425f2>{:dm_mod:dm_request+341} <ffffffff802f828a>{generic_make_request+545} <ffffffff8810ad32>{:ext3:ext3_get_block+0} <ffffffff802f9921>{submit_bio+186} <ffffffff8810ad32>{:ext3:ext3_get_block+0} <ffffffff80263a7c>{__pagevec_lru_add+202} <ffffffff802a300e>{mpage_bio_submit+34} <ffffffff802a3fd7>{mpage_readpages+329} <ffffffff8810ad32>{:ext3:ext3_get_block+0} <ffffffff80243a54>{bit_waitqueue+56} <ffffffff8025fe0f>{bad_range+16} <ffffffff802616df>{get_page_from_freelist+702} <ffffffff80243a54>{bit_waitqueue+56} <ffffffff802618ed>{__alloc_pages+118} <ffffffff80262fd5>{__do_page_cache_readahead+288} <ffffffff8029bd5a>{mntput_no_expire+25} <ffffffff80291112>{link_path_walk+211} <ffffffff80263143>{blockable_page_cache_readahead+86} <ffffffff80263316>{page_cache_readahead+214} <ffffffff8025ce7a>{do_generic_mapping_read+295} <ffffffff8025eaa4>{file_read_actor+0} <ffffffff8025d919>{__generic_file_aio_read+351} <ffffffff802dec36>{avc_has_perm+70} <ffffffff8025d99b>{generic_file_aio_read+52} <ffffffff80281ba6>{do_sync_read+199} <ffffffff8028fbbf>{may_open+88} <ffffffff80243b06>{autoremove_wake_function+0} <ffffffff802594b1>{audit_syscall_entry+301} <ffffffff80282503>{vfs_read+203} <ffffffff802828df>{sys_read+69} <ffffffff8020986c>{tracesys+209} 090: 90 89 d5 9a 00 81 ff ff 6b 6b 6b 6b 6b 6b 6b 6b Prev obj: start=ffff8100d452d488, len=168 Redzone: 0x170fc2a5/0x170fc2a5. Last user: [<ffffffff8030278b>](cfq_set_request+0x39e/0x402) 000: 00 00 00 00 00 00 00 00 01 00 00 00 5a 5a 5a 5a 010: 48 d5 0e 74 00 81 ff ff 00 00 00 00 00 00 00 00 Next obj: start=ffff8100d452d608, len=168 Redzone: 0x170fc2a5/0x170fc2a5. Last user: [<ffffffff8030278b>](cfq_set_request+0x39e/0x402) 000: 00 00 00 00 00 00 00 00 01 00 00 00 5a 5a 5a 5a 010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
There were a number of cfq related corruption problems fixed in the run up to 2.6.17. Reopen this if you can reproduce this on any of the 2.6.17-1.xxxx kernels, but I'm pretty certain this is fixed.