On Thu, 2007-05-24 at 16:00 +0300, Nedko Arnaudov wrote:
William Weston <sysex.net(a)sysex.net> writes:
Ahhh... I misunderstood and thought you were
having troubles only
when JACK was requesting a shutdown (which is only partly solved by
the above patch). Thanks for being clear on this. The segfault on
shutdown was caused by a race condition on thread cancellations and
threads exiting normally. Version 0.10.3 contains the fix for this
(which also improves shutdown time) as well as fixes for slowness
when accessing the file menu for the first time and slow display of
file dialogs.
I'm still getting crash on exit:
(gdb) r
Starting program: /usr/bin/phasex
[Thread debugging using libthread_db enabled]
[New Thread -1220261392 (LWP 9044)]
[New Thread -1220265040 (LWP 9047)]
JACK tmpdir identified as [/tmp]
[New Thread -1229153360 (LWP 9049)]
[New Thread -1237546064 (LWP 9050)]
[New Thread -1248474192 (LWP 9051)]
[New Thread -1256866896 (LWP 9052)]
Thank you for using PHASEX!
(C) 1999-2007 William Weston <weston(a)sysex.net>
Released under the GNU Public License, Ver. 2
[Thread -1237546064 (LWP 9050) exited]
Program received signal SIG32, Real-time event 32.
This is not a crash. If you're debugging JACK applications, it can be
helpful to put this into .gdbinit:
handle SIG33 nostop noprint pass
handle SIG32 nostop noprint pass