Stefan Bidigaray wrote: > Not sure if this is my fault or not! I've gotten this quite often in > testing the NSSound stuff... I've attached a copy of the program. I'm > not sure if it's something I'm doing, or what? I'm using SVN r28323 > (it's kind of old). > > Here's the gdb output: > (gdb) r > Starting program: /root/obj/nssoundtest ../sample.au > [Thread debugging using libthread_db enabled] > [New Thread 0x7f5c43b546f0 (LWP 16334)] > 2009-06-21 18:09:54.846 nssoundtest[16334] Unknown time zone name `CST'. > 2009-06-21 18:09:54.846 nssoundtest[16334] Using time zone with absolute > offset 0. > 2009-06-21 18:09:54.842 nssoundtest[16334] Duration: 6309616 > > Program received signal SIGSEGV, Segmentation fault. > [Switching to Thread 0x7f5c43b546f0 (LWP 16334)] > 0x00007f5c429a3817 in get_imp () from /usr/lib/libobjc.so.2 > (gdb) bt > #0 0x00007f5c429a3817 in get_imp () from /usr/lib/libobjc.so.2 > #1 0x00007f5c42f534db in -[NSAutoreleasePool emptyPool] (self=0xa302a0, > _cmd=0x7f5c434267f0) at NSAutoreleasePool.m:438 > #2 0x00007f5c42f532de in -[NSAutoreleasePool dealloc] (self=0xa302a0, > _cmd=0x7f5c434267e0) at NSAutoreleasePool.m:343 > #3 0x00007f5c42f5329a in -[NSAutoreleasePool release] (self=0xa302a0, > _cmd=0x604790) at NSAutoreleasePool.m:336 > #4 0x0000000000402178 in main (argc=2, argv=0x7fff4bb683c8) > at nssound_test.m:73 > > Let me know if you need all the other files that go along with this.
I did not run your test (You didn't provide a GNUmakefile and it seems to require external headers, which is bad for a test), but from looking at the code I notice that you are releasing "file", which is an already autoreleased NSString. Perhaps it would be sufficient to just drop that line. Cheers, Fred _______________________________________________ Gnustep-dev mailing list Gnustep-dev@gnu.org http://lists.gnu.org/mailman/listinfo/gnustep-dev