pulseaudio crashed with SIGABRT in raise() after manual reloading alsa

[msutkowski]

In order to listen any sound I have to do from console "sudo alsa reload".
After that i was trying to run virtualbox, which warned me, the resource (media) is occupied (or sth like that). Nevertheless it run. After a moment pulseaudio crashed.

ProblemType: Crash
DistroRelease: Ubuntu 11.10
Package: pulseaudio 1:0.99.3-0ubuntu5
ProcVersionSignature: Ubuntu 3.0.0-11.18-generic 3.0.4
Uname: Linux 3.0.0-11-generic x86_64
AlsaVersion: Advanced Linux Sound Architecture Driver Version 1.0.24.
AplayDevices:
 **** List of PLAYBACK Hardware Devices ****
 card 0: SB [HDA ATI SB], device 0: AD198x Analog [AD198x Analog]
   Subdevices: 0/1
   Subdevice #0: subdevice #0
ApportVersion: 1.23-0ubuntu1
Architecture: amd64
ArecordDevices:
 **** List of CAPTURE Hardware Devices ****
 card 0: SB [HDA ATI SB], device 0: AD198x Analog [AD198x Analog]
   Subdevices: 1/1
   Subdevice #0: subdevice #0
AudioDevicesInUse: Error: command ['fuser', '-v', '/dev/snd/seq', '/dev/snd/by-path', '/dev/snd/controlC0', '/dev/snd/hwC0D0', '/dev/snd/hwC0D1', '/dev/snd/pcmC0D0c', '/dev/snd/pcmC0D0p', '/dev/snd/timer'] failed with exit code 1:
Card0.Amixer.info:
 Card hw:0 'SB'/'HDA ATI SB at 0xc8100000 irq 16'
   Mixer name : 'Analog Devices AD1981'
   Components : 'HDA:11d41981,103c30c2,00100200 HDA:11c11040,103c1378,00100200'
   Controls : 11
   Simple ctrls : 9
Date: Mon Sep 19 02:30:56 2011
ExecutablePath: /usr/bin/pulseaudio
InstallationMedia: Ubuntu 11.10 "Oneiric Ocelot" - Beta amd64+mac (20110901)
ProcCmdline: /usr/bin/pulseaudio --start --log-target=syslog
ProcEnviron:
 PATH=(custom, no user)
 LANG=pl_PL.UTF-8
 SHELL=/bin/bash
Signal: 6
SourcePackage: pulseaudio
StacktraceTop:
 raise () from /lib/x86_64-linux-gnu/libc.so.6
 abort () from /lib/x86_64-linux-gnu/libc.so.6
 pa_fdsem_after_poll () from /usr/lib/libpulsecore-0.99.so
 pa_asyncq_write_after_poll () from /usr/lib/libpulsecore-0.99.so
 ?? () from /usr/lib/libpulsecore-0.99.so
Title: pulseaudio crashed with SIGABRT in raise()
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups: adm admin audio avahi backup cdrom dialout fax floppy libvirtd liquidsoap lpadmin netdev plugdev pulse pulse-access sambashare scanner syslog vboxusers video voice
dmi.bios.date: 11/25/2008
dmi.bios.vendor: Hewlett-Packard
dmi.bios.version: 68YTT Ver. F.0E
dmi.board.name: 30C2
dmi.board.vendor: Hewlett-Packard
dmi.board.version: KBC Version 71.2D
dmi.chassis.asset.tag: CNU7311X5T
dmi.chassis.type: 10
dmi.chassis.vendor: Hewlett-Packard
dmi.modalias: dmi:bvnHewlett-Packard:bvr68YTTVer.F.0E:bd11/25/2008:svnHewlett-Packard:pnHPCompaq6715s(GC078ES#AKD):pvrF.0E:rvnHewlett-Packard:rn30C2:rvrKBCVersion71.2D:cvnHewlett-Packard:ct10:cvr:
dmi.product.name: HP Compaq 6715s (GC078ES#AKD)
dmi.product.version: F.0E
dmi.sys.vendor: Hewlett-Packard

[Apport retracing service (apport)]

StacktraceTop:
 pa_fdsem_after_poll (f=0x25d3a90) at pulsecore/fdsem.c:315
 pa_asyncq_write_after_poll (l=0x25d3250) at pulsecore/asyncq.c:319
 asyncmsgq_read_cb (api=<optimized out>, e=<optimized out>, fd=<optimized out>, events=PA_IO_EVENT_INPUT, userdata=0x25d1a50) at pulsecore/thread-mq.c:45
 dispatch_pollfds (m=0x2595d10) at pulse/mainloop.c:679
 pa_mainloop_dispatch (m=0x2595d10) at pulse/mainloop.c:929

[David Henningsson (diwic)]

Some notes:

#9 0x00007f5a4c6b096b in pa_fdsem_after_poll (f=0x25d3a90) at pulsecore/fdsem.c:315
        __func__ = "pa_fdsem_after_poll"
        __PRETTY_FUNCTION__ = "pa_fdsem_after_poll"
#10 0x00007f5a4c6a08fc in pa_asyncq_write_after_poll (l=0x25d3250) at pulsecore/asyncq.c:319
        __func__ = "pa_asyncq_write_after_poll"
        __PRETTY_FUNCTION__ = "pa_asyncq_write_after_poll"

#2 0x00007f5a4c6b0970 in pa_fdsem_after_poll (f=0x25d3a90) at pulsecore/fdsem.c:315
        __func__ = "pa_fdsem_after_poll"
        __PRETTY_FUNCTION__ = "pa_fdsem_after_poll"
#3 0x00007f5a4c6a08fc in pa_asyncq_write_after_poll (l=0x25d3250) at pulsecore/asyncq.c:319
        __func__ = "pa_asyncq_write_after_poll"
        __PRETTY_FUNCTION__ = "pa_asyncq_write_after_poll"

I assume, both threads should not call this simultaneously on the same object, which is module-null-sink's thread_mq->outq. The outq writes in the thread and reads from the main thread. Therefore it seems like the main thread should not call write_after_poll.

[David Henningsson (diwic)]

I'm trying to trace down a crash:

https://bugs.launchpad.net/ubuntu/+source/pulseaudio/+bug/853560
https://launchpadlibrarian.net/80418699/ThreadStacktrace.txt

I'm seeing that both threads call "pa_asyncq_write_after_poll"
simultaneously on the same object, which is module-null-sink's
thread_mq->outq. The outq writes in the thread and reads from the main
thread. Therefore it seems like the main thread should not call
write_after_poll. From what I can see, there almost seems to be a typo
in asyncmsgq_read_cb (see suggested patch below).

But since this is deep down, has been there for three years without
causing trouble in the past (for what we know), I'd like a second
opinion before actually applying this patch. I've just run a a quick
test here and it didn't seem to break anything.

diff --git a/src/pulsecore/thread-mq.c b/src/pulsecore/thread-mq.c
index 51fb765..b492793 100644
--- a/src/pulsecore/thread-mq.c
+++ b/src/pulsecore/thread-mq.c
@@ -42,7 +42,7 @@ static void asyncmsgq_read_cb(pa_mainloop_api*api,
pa_io_event* e, int fd, pa_io
      pa_assert(events == PA_IO_EVENT_INPUT);

      pa_asyncmsgq_ref(aq = q->outq);
- pa_asyncmsgq_write_after_poll(aq);
+ pa_asyncmsgq_read_after_poll(aq);

      for (;;) {
          pa_msgobject *object;

--
David Henningsson, Canonical Ltd.
http://launchpad.net/~diwic

[David Henningsson (diwic)]

I have now sent a patch for discussion upstream before I apply it to pulseaudio.

[David Henningsson (diwic)]

Applied upstream.

[David Henningsson (diwic)]

Fixed in PulseAudio 1.0