Bug? scheduling while atomic: wine-preloader

[Joshua Kwan <joshk@triplehelix.org>]

Hello,

I got this on a stock 2.6.12 (haven't added .5 yet) setup while trying
to do crazy stuff with Wine.

scheduling while atomic: wine-preloader/0x1000000c/13047
 [<c032f517>] schedule+0x5d7/0x5e0
 [<c01923f7>] ext3_ordered_commit_write+0xc7/0xf0
 [<c032fcb9>] cond_resched+0x29/0x40
 [<c0138971>] generic_file_buffered_write+0x2e1/0x650
 [<c016ff02>] inode_update_time+0x52/0xe0
 [<c013900b>] __generic_file_aio_write_nolock+0x32b/0x570
 [<c0191b60>] ext3_get_block+0x0/0xa0
 [<c013b6a2>] __alloc_pages+0x2d2/0x420
 [<c0139502>] generic_file_aio_write+0x72/0xe0
 [<c018fab4>] ext3_file_write+0x44/0xd0
 [<c015488e>] do_sync_write+0xbe/0xf0
 [<c0145e7e>] do_no_page+0x1ae/0x310
 [<c01401c8>] mark_page_accessed+0x28/0x30
 [<c014498e>] __follow_page+0x9e/0xc0
 [<c012c770>] autoremove_wake_function+0x0/0x60
 [<c0144ad8>] get_user_pages+0xb8/0x390
 [<c017ef2b>] dump_write+0x2b/0x40
 [<c017ffcd>] elf_core_dump+0xa5d/0xc70
 [<c0160dab>] do_coredump+0x1cb/0x208
 [<c0121d35>] __dequeue_signal+0xf5/0x1d0
 [<c0121e45>] dequeue_signal+0x35/0xe0
 [<c0123b19>] get_signal_to_deliver+0x209/0x2f0
 [<c0102aa8>] do_signal+0x98/0x130
 [<c0109fa3>] restore_i387_fxsave+0x83/0x90
 [<c0123c6d>] sigprocmask+0x3d/0xb0
 [<c0123d62>] sys_rt_sigprocmask+0x82/0x100
 [<c0112730>] do_page_fault+0x0/0x599
 [<c0102b75>] do_notify_resume+0x35/0x38
 [<c0102d32>] work_notifysig+0x13/0x15
note: wine-preloader[13047] exited with preempt_count 12

I'm assuming that This Shouldn't Happen, so I thought that kernel
hackers in tune with current developments might like to know about this.

Please CC me on replies.

-- 
Joshua Kwan

Attachment: signature.asc
Description: OpenPGP digital signature