View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0000285 | ardour | bugs | public | 2004-02-29 21:41 | 2004-05-28 09:42 |
| Reporter | v2 | Assigned To | |||
| Priority | normal | Severity | block | Reproducibility | always |
| Status | closed | Resolution | open | ||
| Summary | 0000285: Export hangs jackd, with jackd 0.94.0 | ||||
| Description | I get a segmentation fault when exporting a session. The session is recorded at 48khz, and I'm exporting it to 44.1, best resample & shaped dither. The session has a master bus, and I'm exporting the session through that. Ardour crashes with just 'Segmentation fault'. No other messages. I'm running Debian testing with the agnula 1.1.0 kernel. GCC 3.3.2. jackd 0.94.0: ./configure --prefix=/usr --enable-capabilities --enable-optimize --with-default-tmpdir=/dev/shm/ --enable-timestamps Ardour/GTK 0.452.0 running with libardour 0.727.0: ./configure --prefix=/usr --enable-optimize | ||||
| Tags | No tags attached. | ||||
|
|
does it always crash? can you run it under gdb and provide us with a backtrace? |
|
|
It always crashes.I ran gdb like this: gdb ardour Here's what I got: Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 49156 (LWP 725)] 0x084442d2 in SigC::Signal1<int, unsigned, SigC::Marshal<int> >::emit(unsigned const&) () at basic_signal.cc:25 25 Current language: auto; currently c++ (gdb) backtrace #0 0x084442d2 in SigC::Signal1<int, unsigned, SigC::Marshal<int> >::emit(unsigned const&) () at basic_signal.cc:25 Error accessing memory address 0xbf3ffa30: No such process. (gdb) Should I compile ardour with debugging, or give gdb other options? |
|
|
I couldn't get a backtrace after the window had died. I moved fast this time, and I wrote bt to gdb at the moment it dropped into the console. This is what I got: #0 0x084442d2 in SigC::Signal1<int, unsigned, SigC::Marshal<int> >::emit(unsigned const&) () at basic_signal.cc:25 0000001 0x08442998 in ARDOUR::AudioEngine::process_callback(unsigned) () at basic_signal.cc:25 #2 0x40287413 in jack_client_thread () from /usr/lib/libjack.so.0 #3 0x40294e51 in pthread_start_thread () from /lib/libpthread.so.0 0000004 0x40294ecf in pthread_start_thread_event () from /lib/libpthread.so.0 0000005 0x406d664a in clone () from /lib/libc.so.6 I don't know does that help you much more though... libpthread is libpthread-0.10.so, it's from the debian package libc6, version 2.3.2.ds1-11 (testing). |
|
|
This fixed itself. I have now a better low latency kernel, and a newer version of ardour. I can't say which one did the trick, but I suspect the kernel. But, if it was the kernel, I would still think of this as a bug. Latency problems shouldn't affect exporting sessions at all. Realtime issues should not affect it. edited on: 03-15-04 00:09 |
|
|
I'll take that back. 16 bit worked previously, but I forgot to set the samplerate to 44.1. I just tried exporting to 16 bit 44.1 (shaped noise + best quality) and ardour crashed about 15 seconds into the export (0000009:0000010% finished). |
|
|
Too many things have happened to Ardour since this bug was reported. If I run into exporting problems, I'll create a new bug. |
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2004-02-29 21:41 | v2 | New Issue | |
| 2004-03-01 02:57 | paul | Note Added: 0000481 | |
| 2004-03-01 02:57 | paul | Status | new => feedback |
| 2004-03-01 14:21 | v2 | Note Added: 0000491 | |
| 2004-03-01 14:29 | v2 | Note Added: 0000492 | |
| 2004-03-14 10:34 | v2 | Note Added: 0000541 | |
| 2004-03-15 08:09 | v2 | Note Edited: 0000541 | |
| 2004-03-15 15:59 | v2 | Note Added: 0000545 | |
| 2004-05-28 09:42 | v2 | Status | feedback => closed |
| 2004-05-28 09:42 | v2 | Note Added: 0000916 |
