| Summary: | gentoo-sources 2.6.31-r6: reboot and poweroff not working | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | Evgeniy Dushistov <dushistov> |
| Component: | [OLD] Core system | Assignee: | Gentoo Kernel Bug Wranglers and Kernel Maintainers <kernel> |
| Status: | RESOLVED FIXED | ||
| Severity: | major | ||
| Priority: | High | ||
| Version: | unspecified | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | linux-2.6.31,linux-2.6.32 | ||
| Package list: | Runtime testing required: | --- | |
| Attachments: |
linux-2.6.28-gentoo-r5/.config
linux-2.6.31-gentoo-r6/.config |
||
|
Description
Evgeniy Dushistov
2010-01-14 12:51:45 UTC
The both kerneles were build from gentoo-sources package, using genkernel package. The kernel config for 2.6.31 is build from 2.6.28 config with "make oldconfig" Created attachment 216488 [details]
linux-2.6.28-gentoo-r5/.config
Created attachment 216490 [details]
linux-2.6.31-gentoo-r6/.config
Using SysRq key I can see that rmmod process have "D" state Thanks for report, Evgeniy. Please, try to reproduce problem with 2.6.31-r9 and then 2.6.32-r2. yes, please test with the latest gentoo sources 2.6.32 which is -r3 as of this writing I am having this problem too with gentoo-sources-2.6.31-r10. With r9, this problem goes away. I am running packages from the ~x86 testing tree. I forgot to add that this issue also occurs on suspend to RAM, but does not (always) seem to occur with reboots. I was at a loss to diagnose the cause of this until I decided to roll back to the previous kernel. My configuration file was identical in both. >yes, please test with the latest gentoo sources 2.6.32 which is -r3 as of this
>writing
I tried linux-2.6.32-gentoo-r3, the same bug.
reboot/poweroff hang on a half.
rmmod deadlocked.
Here is part of backtrace:
(gdb) li *0xc04779e1
0xc04779e1 is in wait_for_common (kernel/sched.c:5794)
(gdb) li *0xc012e840
0xc012e840 is in default_wake_function (kernel/sched.c:5626)
(gdb) li *0xc0477ad2
0xc0477ad2 is in wait_for_completion (kernel/sched.c:5825)
(gdb) li *0xc015c74a
0xc015c74a is in sys_delete_module (kernel/module.c:789)
I wonder what module this is hanging on. What modules are you using? Have you tried to rmmod all of the modules in use and then tried shutdown/reboot ? >I wonder what module this is hanging on.
>What modules are you using? Have you tried to rmmod all of the modules in use
>and then tried shutdown/reboot ?
Yes, I tried.
The problem in one of "snd" modules, if I remove all stuff related to snd,
all works fine, I able to reboot.
Here from BAD 2.6.32:
snd_pcm_oss 29184 0
snd_mixer_oss 12352 1 snd_pcm_oss
snd_seq_oss 20768 0
snd_seq_midi_event 3520 1 snd_seq_oss
snd_seq 34832 4 snd_seq_oss,snd_seq_midi_event
snd_seq_device 3532 2 snd_seq_oss,snd_seq
snd_hda_codec_realtek 168612 1
snd_hda_intel 16712 0
snd_hda_codec 42016 2 snd_hda_codec_realtek,snd_hda_intel
snd_pcm 47204 3 snd_pcm_oss,snd_hda_intel,snd_hda_codec
snd_timer 13864 2 snd_seq,snd_pcm
snd 35652 10 snd_pcm_oss,snd_mixer_oss,snd_seq_oss,snd_seq,snd_seq_device,snd_hda_codec_realtek,snd_hda_intel,snd_hda_codec,snd_pcm,snd_timer
snd_page_alloc 5000 2 snd_hda_intel,snd_pcm
Here from GOOD 2.6.28:
snd_pcm_oss 33184 0
snd_mixer_oss 15232 1 snd_pcm_oss
snd_seq_oss 26624 0
snd_seq_midi_event 7680 1 snd_seq_oss
snd_seq 38640 4 snd_seq_oss,snd_seq_midi_event
snd_seq_device 7820 2 snd_seq_oss,snd_seq
snd_hda_intel 339216 0
snd_pcm 49668 2 snd_pcm_oss,snd_hda_intel
snd_timer 17672 2 snd_seq,snd_pcm
snd 37540 8 snd_pcm_oss,snd_mixer_oss,snd_seq_oss,snd_seq,snd_seq_device,snd_hda_intel,snd_pcm,snd_timer
snd_page_alloc 9224 2 snd_hda_intel,snd_pcm
With 2.6.32 snd_hda_codec_realtek loaded, there is no such with 2.6.28.
I can recall a similar bug in the past with alsa-* which failed to unload snd modules properly http://bugs.gentoo.org/show_bug.cgi?id=232875 Yes, those alsa modules do not successfully unload on shutdown. There is a closed bug on that somewhere. They should not be unloaded during shutdown. |
