Bug 570345
| Summary: | [abrt] crash in muse-1:1.0.1-1.fc12: Process /usr/bin/muse was killed by signal 11 (SIGSEGV) | ||||||
|---|---|---|---|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Adam Huffman <bloch> | ||||
| Component: | muse | Assignee: | Orcan Ogetbil <oget.fedora> | ||||
| Status: | CLOSED DUPLICATE | QA Contact: | Fedora Extras Quality Assurance <extras-qa> | ||||
| Severity: | medium | Docs Contact: | |||||
| Priority: | low | ||||||
| Version: | 12 | CC: | nando, oget.fedora | ||||
| Target Milestone: | --- | ||||||
| Target Release: | --- | ||||||
| Hardware: | x86_64 | ||||||
| OS: | Linux | ||||||
| Whiteboard: | abrt_hash:bd8d1ee73587881a1e3dca904ccc4c1aed898fcd | ||||||
| Fixed In Version: | Doc Type: | Bug Fix | |||||
| Doc Text: | Story Points: | --- | |||||
| Clone Of: | Environment: | ||||||
| Last Closed: | 2010-03-09 06:48:53 UTC | Type: | --- | ||||
| Regression: | --- | Mount Type: | --- | ||||
| Documentation: | --- | CRM: | |||||
| Verified Versions: | Category: | --- | |||||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||
| Cloudforms Team: | --- | Target Upstream Version: | |||||
| Embargoed: | |||||||
| Attachments: |
|
||||||
|
Description
Adam Huffman
2010-03-03 23:21:13 UTC
Created attachment 397680 [details]
File: backtrace
The key line is this: MusE: WARNING: Recommend setting audio realtime priority to at least 5! How do you set up jack? If you are using qjackctl, go to Setup, and set the priority to at least 5. I usually set it to 20. If you are using pure jackd, pass a priority flag, such as "-P20". Let us know if it works. I'm not currently running an RT kernel, so I can't make that setting. Uh, then how can we make this compatible with both kernels? We have to support the Fedora kernel, since this is Fedora right? I can patch up muse to pull the priority requirement down but that will screw up users who run the Fedora kernel. I will think about this. Let me know if you have a solution. (In reply to comment #3) > I'm not currently running an RT kernel, so I can't make that setting. What error do you get if you try: jackd -R -d alsa -d hw:0 (assuming your first soundcard is the one you want to use). You do not need at rt kernel to run jackd with "-R". If you are using the default jack package from Fedora you need to belong to a certain group to be able to use realtime priority (I forget what the name is for the group). Maybe that is what is missing. Oops. I missed the "not" in his sentence. I thought he was running RT kernel. My bad. Fedora's jack configuration (the groups to add yourself to) is explained at /usr/share/doc/jack-audio-connection-kit-0.118.0/README.Fedora I assume every jack user already does this configuration. Maybe this is not a good assumption. Ah, I'm glad to see there is such documentation as I've looked in vain for something like that. Perhaps it should be mentioned more widely - on the Audio Creation SIG wiki page, for instance? I was already in the pulse-rt group, but not the jackuser one. Will rectify that and try again. Sorry - my account is already in the 'jackuser' group as well and I already had '-R' and '-P' in my jackd invocation: jackd -R -P4 -dalsa -r44100 -p512 -n4 -D -Chw:0 -Phw:0 > Ah, I'm glad to see there is such documentation as I've looked in vain for > something like that. Perhaps it should be mentioned more widely - on the Audio > Creation SIG wiki page, for instance? Good idea. > Sorry - my account is already in the 'jackuser' group as well and I already had > '-R' and '-P' in my jackd invocation: > Make sure you logout/login after you add yourself to those groups. > jackd -R -P4 -dalsa -r44100 -p512 -n4 -D -Chw:0 -Phw:0 Okay, now you want to have a priority of at least 5. So change that -P4 flag to -Px where x>4 Okay, I've invoked jackd with -P5 and started muse. It still crashed... muse Denormal protection enabled. Muse: __i386__ or __x86_64__ defined. Using optimized float buffer copying (asm movsl). Using SSE optimized routines no locale <muse_en_GB.utf8>/</usr/share/muse/locale> Trying RTC timer... fatal error: open /dev/rtc failed: Permission denied hint: check if 'rtc' kernel module is loaded, or used by something else Trying ALSA timer... AlsaTimer::initTimer(): best available ALSA timer: system timer got timer = 50 starting with default template JACK ERROR: cannot use real-time scheduling (FIFO at priority 0) [for thread -2069096688, from thread -2069096688] (22: Invalid argument) MusE: JackAudioDevice::realtimePriority: JACK is not running realtime MusE: WARNING: Recommend setting audio realtime priority to at least 5! Cannot set scheduling priority -5 for RT thread (Resource temporarily unavailable) creating thread <Prefetch> failed: Invalid argument AlsaTimer::setTimerTicks(): requested freq 1024 Hz too high for timer (max is 1000) freq stays at 1000 Hz Cannot set scheduling priority -2 for RT thread (Resource temporarily unavailable) creating thread <Midi> failed: Invalid argument looping waiting for sequencer thread to start looping waiting for sequencer thread to start looping waiting for sequencer thread to start looping waiting for sequencer thread to start looping waiting for sequencer thread to start looping waiting for sequencer thread to start looping waiting for sequencer thread to start looping waiting for sequencer thread to start looping waiting for sequencer thread to start looping waiting for sequencer thread to start midi sequencer thread does not start!? Exiting... Thanks. But we can't proceed much further if you don't give us enough information. Please tell us all the flags jack now starts with. Can you kill your jack if it is running, then start it over? Then start muse. For instance my jack-0.118.0 starts with /usr/bin/jackd -R -P20 -dalsa -dhw:0 -r44100 -p1024 -n2 -Xseq and muse-1.0.1 works fine. I mentioned the only parameter I changed from the invocation listed earlier... Anyway, here it is: jackd -R -P5 -dalsa -r44100 -p512 -n4 -D -Chw:0 -Phw:0 Did you try anything higher than -P5? Like -P6 or -P10 or -P20 etc? Yes, it did startup successfully with -P20. Still think it shouldn't crash with lower values. You are right. It shouldn't. I'll talk to the developers. I realized that this is the same crash on start that Fanis reported a while ago. There were a couple more reports of this crash too. Let's collect them all at one place. *** This bug has been marked as a duplicate of bug 561704 *** |