As of kernel 2.6.35.2, having problems with "BUG: scheduling while atomic: ..." on different processes. Kernel 2.6.35.1 and earlier don't have this issue. Representative trace info follows: ~ Andy ========================= BUG: scheduling while atomic: sh/13943/0x10000001 Modules linked in: fan snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device snd_pcm_oss snd_mixer_oss pl2303 fuse pcmcia snd_atiixp snd_ac97_codec ac97_bus yenta_socket ohci_hcd ehci_hcd thermal battery processor firewire_ohci ac shpchp snd_pcm wmi button pcmcia_rsrc video ide_cd_mod pcmcia_core snd_timer snd ssb snd_page_alloc Pid: 13943, comm: sh Not tainted 2.6.35.2 #1 Call Trace: [<c1028573>] __schedule_bug+0x58/0x5a [<c150f98f>] schedule+0x343/0x43e [<c150fb9b>] _cond_resched+0x2d/0x42 [<c10aa8f8>] anon_vma_prepare+0xf/0xbc [<c10a7239>] expand_downwards+0x12/0x124 [<c10a7353>] expand_stack+0x8/0xa [<c10a5460>] handle_mm_fault+0x7fb/0x82d [<c10a26f9>] ? follow_page+0x55/0x261 [<c10a5590>] __get_user_pages+0xfe/0x3ba [<c10a58dc>] get_user_pages+0x49/0x51 [<c10bfb57>] get_arg_page+0x46/0x8f [<c11ac948>] ? strnlen_user+0x1b/0x39 [<c10bfc6b>] copy_strings+0xcb/0x173 [<c10bfd2f>] copy_strings_kernel+0x1c/0x2b [<c10c1143>] do_execve+0x17c/0x299 [<c11aca85>] ? copy_to_user+0x2e/0x119 [<c11ac9ff>] ? strncpy_from_user+0x2a/0x4b [<c1007c20>] sys_execve+0x2d/0x54 [<c1002d4e>] ptregs_execve+0x12/0x18 [<c1510f1c>] ? syscall_call+0x7/0xb
Unfortunately I did not find bug #16588 until after submitting this one. I am changing the status to RESOLVED/CODE_FIX because applying commit hash 11ac5524... (http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=11ac552477e32835cb6970bf0a70c210807f5673), as suggested in that bug's comment log, seems to have fixed the problem on this end. ~ Andy