Bug#639211: [pulseaudio] PA gets killed if module-combine and realtime scheduling enabled.
Antonio Ospite
ospite at studenti.unina.it
Thu Oct 6 10:34:42 UTC 2011
Package: pulseaudio
Version: 1.0-4
Followup-For: Bug #639211
Severity: important
Hi,
I too am experiencing this problem, this is the output of strace, I am on
x86_64:
-[ BEGIN ]-------------------------------------------------------------
....
open("/usr/lib/pulse-1.0/modules/module-combine-sink.so", O_RDONLY) = 31
read(31, "\177ELF\2\1\1\0\0\0\0\0\0\0\0\0\3\0>\0\1\0\0\0\3403\0\0\0\0\0\0"..., 832) = 832
fstat(31, {st_mode=S_IFREG|0644, st_size=40648, ...}) = 0
mmap(NULL, 2135760, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 31, 0) = 0x7fb3e8ca7000
mprotect(0x7fb3e8cb0000, 2097152, PROT_NONE) = 0
mmap(0x7fb3e8eb0000, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 31, 0x9000) = 0x7fb3e8eb0000
close(31) = 0
eventfd2(0, O_CLOEXEC) = 31
eventfd2(0, O_CLOEXEC) = 32
eventfd2(0, O_CLOEXEC) = 33
eventfd2(0, O_CLOEXEC) = 34
eventfd2(0, O_CLOEXEC) = 35
eventfd2(0, O_CLOEXEC) = 36
eventfd2(0, O_CLOEXEC) = 37
eventfd2(0, O_CLOEXEC) = 38
mmap(NULL, 8392704, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS|MAP_STACK, -1, 0) = 0x7fb3e84a6000
mprotect(0x7fb3e84a6000, 4096, PROT_NONE) = 0
clone(child_stack=0x7fb3e8ca5fb0, flags=CLONE_VM|CLONE_FS|CLONE_FILES|CLONE_SIGHAND|CLONE_THREAD|CLONE_SYSVSEM|CLONE_SETTLS|CLONE_PARENT_SETTID|CLONE_CHILD_CLEARTID, parent_tidptr=0x7fb3e8ca69d0, tls=0x7fb3e8ca6700, child_tidptr=0x7fb3e8ca69d0) = 23378
futex(0xe4d350, FUTEX_WAIT_PRIVATE, 0, NULL) = 0
sendto(32, "\1\0\0\0\0\0\0\0", 8, MSG_NOSIGNAL, NULL, 0) = -1 ENOTSOCK (Socket operation on non-socket)
write(32, "\1\0\0\0\0\0\0\0", 8) = 8
futex(0xe4d350, FUTEX_WAIT_PRIVATE, 0, NULL <unfinished ...>
+++ killed by SIGKILL +++
Ucciso
-[ END ]---------------------------------------------------------------
As a workaround I am running "pulseaudio --realtime=false" in a console after
logging in, but maybe I could even disable realtime in some config file.
I also gave a quick test to the latest upstream version and it runs fine, either with
--realtime=false or --realtime=true, but I didn't use all the compile options
as the Debian package, I am attaching the ldd difference between the two
binaries, just to contextualize the quick test.
Please let me know if you need more info.
Thanks,
Antonio Ospite
http://ao2.it
-- System Information:
Debian Release: wheezy/sid
APT prefers unstable
APT policy: (900, 'unstable'), (600, 'experimental')
Architecture: amd64 (x86_64)
Kernel: Linux 3.1.0-rc9-ao2 (SMP w/1 CPU core)
Locale: LANG=it_IT.utf8, LC_CTYPE=it_IT.utf8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash
Versions of packages pulseaudio depends on:
ii adduser 3.113
ii consolekit 0.4.5-1
ii libasound2 1.0.24.1-4
ii libasound2-plugins 1.0.24-2+b1
ii libc6 2.13-21
ii libcap2 1:2.22-1
ii libdbus-1-3 1.4.16-1
ii libfftw3-3 3.2.2-1
ii libice6 2:1.0.7-2
ii libltdl7 2.4-4
ii liborc-0.4-0 1:0.4.15-2
ii libpulse0 1.0-4
ii libsamplerate0 0.1.8-1
ii libsm6 2:1.2.0-2
ii libsndfile1 1.0.25-3
ii libspeexdsp1 1.2~rc1-1
ii libtdb1 1.2.9-4+b1
ii libudev0 172-1
ii libx11-6 2:1.4.4-2
ii libx11-xcb1 2:1.4.4-2
ii libxcb1 1.7-3
ii libxtst6 2:1.2.0-3
ii lsb-base 3.2-28
ii udev 172-1
Versions of packages pulseaudio recommends:
ii gstreamer0.10-pulseaudio 0.10.30-1
ii pulseaudio-esound-compat 1.0-4
ii pulseaudio-module-x11 1.0-4
ii rtkit 0.10-2
Versions of packages pulseaudio suggests:
ii paman <none>
ii paprefs 0.9.9-2
ii pavucontrol 0.99.2-1
ii pavumeter <none>
ii pulseaudio-utils 1.0-4
-- no debconf information
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ldd_diff_upstream_Debian.diff
Type: text/x-diff
Size: 2381 bytes
Desc: not available
URL: <http://lists.alioth.debian.org/pipermail/pkg-pulseaudio-devel/attachments/20111006/0003a09f/attachment.diff>
More information about the pkg-pulseaudio-devel
mailing list