* [2.5.74] bad: scheduling while atomic!
@ 2003-07-04 15:34 Roger Luethi
2003-07-05 2:28 ` Andrew Morton
0 siblings, 1 reply; 2+ messages in thread
From: Roger Luethi @ 2003-07-04 15:34 UTC (permalink / raw)
To: linux-kernel
I haven't had the time to investigate this, so I don't have much
information to share beyond the trace below. I think I have seen this at
least with 2.5.73, too. The system looks okay, then, usually hours later
(if at all, it's a rare event), something triggers a flood of those call
traces (many of them per second).
The syslog seems to suggest it might be related to IDE DMA:
Jul 4 17:17:28 [kernel] hda: dma_timer_expiry: dma status == 0x61
Jul 4 17:17:44 [kernel] hda: timeout waiting for DMA
Jul 4 17:17:44 [kernel] [<c0107000>] default_idle+0x0/0x40
Jul 4 17:17:44 [kernel] bad: scheduling while atomic!
Compiler is gcc 3.2.3.
bad: scheduling while atomic!
Call Trace:
[<c0107000>] default_idle+0x0/0x40
[<c011f110>] schedule+0x500/0x510
[<c0107063>] poll_idle+0x23/0x40
[<c0118073>] apm_cpu_idle+0xa3/0x140
[<c0117fd0>] apm_cpu_idle+0x0/0x140
[<c0107000>] default_idle+0x0/0x40
[<c01070b8>] cpu_idle+0x38/0x40
[<c0105000>] rest_init+0x0/0x30
[<c037c738>] start_kernel+0x138/0x140
[<c037c4c0>] unknown_bootoption+0x0/0x100
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [2.5.74] bad: scheduling while atomic!
2003-07-04 15:34 [2.5.74] bad: scheduling while atomic! Roger Luethi
@ 2003-07-05 2:28 ` Andrew Morton
0 siblings, 0 replies; 2+ messages in thread
From: Andrew Morton @ 2003-07-05 2:28 UTC (permalink / raw)
To: Roger Luethi; +Cc: linux-kernel, Bartlomiej Zolnierkiewicz
Roger Luethi <rl@hellgate.ch> wrote:
>
> I haven't had the time to investigate this, so I don't have much
> information to share beyond the trace below. I think I have seen this at
> least with 2.5.73, too. The system looks okay, then, usually hours later
> (if at all, it's a rare event), something triggers a flood of those call
> traces (many of them per second).
>
> The syslog seems to suggest it might be related to IDE DMA:
>
> Jul 4 17:17:28 [kernel] hda: dma_timer_expiry: dma status == 0x61
> Jul 4 17:17:44 [kernel] hda: timeout waiting for DMA
> Jul 4 17:17:44 [kernel] [<c0107000>] default_idle+0x0/0x40
> Jul 4 17:17:44 [kernel] bad: scheduling while atomic!
>
> Compiler is gcc 3.2.3.
>
> bad: scheduling while atomic!
> Call Trace:
> [<c0107000>] default_idle+0x0/0x40
> [<c011f110>] schedule+0x500/0x510
> [<c0107063>] poll_idle+0x23/0x40
> [<c0118073>] apm_cpu_idle+0xa3/0x140
> [<c0117fd0>] apm_cpu_idle+0x0/0x140
> [<c0107000>] default_idle+0x0/0x40
> [<c01070b8>] cpu_idle+0x38/0x40
> [<c0105000>] rest_init+0x0/0x30
> [<c037c738>] start_kernel+0x138/0x140
> [<c037c4c0>] unknown_bootoption+0x0/0x100
Possibly the IDE error handler has a locking imbalance. It returned from
the interrupt handler without having unlocked a lock which it should have
unlocked, and that left the currently-running process (the idle task in
this case) with an incorrect preempt count.
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2003-07-05 2:12 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2003-07-04 15:34 [2.5.74] bad: scheduling while atomic! Roger Luethi
2003-07-05 2:28 ` Andrew Morton
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox