On 5/11/2012, at 5:59 AM, Sjoerd Simons wrote:
On Sun, Feb 12, 2012 at 01:46:27PM +1300, Michael Cree wrote:
Other hints that this is a kernel problem introduced somewhere
between
2.6.38 and 2.6.39 is that the eglibc test suite passes on 2.6.38 and
older kernels, but fails on 2.6.39 and newer
close 649641
thanks,
On Mon, 2012-11-05 at 09:02 +1300, Michael Cree wrote:
> On 5/11/2012, at 5:59 AM, Sjoerd Simons wrote:
> It is now fixed in the kernel; commit 62aca403657fe upstream. It was
> a problem with futexes.
Great thanks, closing this bug :)
--
Sjoerd Simons
--
To UNSUBSCRIB
On Sun, Feb 12, 2012 at 01:46:27PM +1300, Michael Cree wrote:
> Other hints that this is a kernel problem introduced somewhere between
> 2.6.38 and 2.6.39 is that the eglibc test suite passes on 2.6.38 and
> older kernels, but fails on 2.6.39 and newer kernels with failures in
> mutex related tests
The pulseaudio problem on Alpha is worse than just causing a crash in
iceweasel---pulseaudio cannot auto-detect audio cards via the udev
mechanism.
I have a number of audio cards (HDA-intel on a newer Radeon graphics
card, and M-Audio Revolution 5.1) in my Alpha XP1000 but these are not
detected d
Verified that the previously-mentioned workaround is applicable to
Alpha, and that it works: no more mutex failures. The specific
implementation of the workaround for the 1.1-1 source package is to
patch "m4/ax_pthread.m4" line 266ff to read as follows:
AC_CACHE_CHECK([for PTHREAD_PRIO_IN
Package: pulseaudio
Version: 1.1-1
Severity: important
Iceweasel and firefox browsers crash with the following error:
Assertion 'pthread_mutex_unlock(&m->mutex) == 0' failed at
pulsecore/mutex-posix.c:106, function pa_mutex_unlock(). Aborting.
Always reproducible by starting either firefox or i
6 matches
Mail list logo
