Yes. I am editing the Actions. I place one marker at the beginning of the file, then create one or more actions such as PC to an external device, setting a value or a caption on a named widget. I am using only one rackspace. I use the Streaming Audio File Player to reference one folder containing my backing tracks that I play over.
no, I was not deleting any marker while this happened (at least I don’t recall), but I do have a few markers at position 0:00:00 in my songs (sometimes 1, sometimes 3, sometimes 0). Maybe switching to songs which have markers at this (the same) position (playhead was at 0 as well) looks to the program like some of them are disappearing - just guessing? (but I wasn’t even playing anything while it happened).
By any chance are you able to reproduce the crash? Even if you can’t crash on demand, is there something you do that even sometimes causes the crash? What actions are you using? Does it happen only when you are editing or does it happen by itself when you are playing a song?
Can you post a crash report here? That would be very helpful
I’m experienced similar crashes yesterday - also seems to be related to Streaming Audio Player. Today I was able to add ~60 markers with actions to next song in the same project - no crashes so far - some nasty heisenbug.
Sep1-crash.txt (66.4 KB)
Here is the crash report from today. There is one from the 30th and two on the 29th. Thanks for looking into this.
This is the same MacBook/OS I have. I have tried to replicate a crash when editing Actions in the Streaming Audio File Player, but haven’t so far. Any more details about your setup would be helpful.
- How many songs?
- What file types do you use?
- How many markers per song on average?
- How many actions per marker on average?
- Are you only using markers/actions at the beginning of the file?
- When you send PC messages, is this to the Mac’s IAC Bus, or to a MIDI interface?
- Do you have any widgets mapped to any of the SAFP parameters?
- Is the global playhead (or SAFP) playing at the time of the crash?
I cannot reproduce either. There is clearly an issue but I don’t know how to make it happen
Both reported crashed (one on Intel, one on Apple Silicon) are happening when the system is trying to update the current number of actions for a marker in the list of markers. That’s a clue but I still can’t figure out how to reproduce the problem.
If you experiment further to track down the issue, please remember what exactly you did and especially what step triggered the crash. That will be super useful for devs to fix the issue.
Sorry for the delay but i was on tour. No scripts on those rackspaces
I have merged the two topics that are reporting the identical bug into a single topic
20 songs
WAV files
All but one song have only one marker at the very beginning of the file
Currently there are 3 actions triggered by that first marker (one may be unnecessary since I discovered the 0.0 dB column in the FILELIST VIEW)
One song has a PC later on in the song
IAC Bus or MIDI interface? Hmm. I had to install the BOSS ME-80 driver for it to be recognized as a device in MIDI Studio. Then GP recognized it as well.
Several:
Label => Track Number
Button => Previous Track
Button => Next Track
Knob => File Player Volume
Switch (On) => Notify Playhead Position
Slider => Position
Button => Play
The playhead wasn’t active when the crash happened. I hope this helps.
What are the actions?
Fresh crash. I woke computer our from sleep, selected next song I want to convert to markers and tried to add new marker. Crash log is similar, but not the same.
Exception Type: EXC_BAD_ACCESS (SIGSEGV)
Exception Codes: KERN_INVALID_ADDRESS at 0x0000000000000010
Exception Codes: 0x0000000000000001, 0x0000000000000010
Exception Type: EXC_BAD_ACCESS (SIGSEGV)
Exception Codes: KERN_INVALID_ADDRESS at 0x0000a637a439cfd8 -> 0x00002637a439cfd8 (possible pointer authentication failure)
Exception Codes: 0x0000000000000001, 0x0000a637a439cfd8
This awfully looks like use-after-free, double release or something similar.
gp-crash-20230902.txt (48.9 KB)
I think “waking up computer from sleep” might be a clue here as well. On Thursday it also happened not far away from beginning my work that day.
and another hang… Again, happened after not using GP for ~2 hours (it was opened all the time). Here is spindump:
gp-Spindump-20230902144700.zip (453.7 KB)
Including sampling of the process at the time of hang as well.
gp-sampled-process-20230902145100.zip (38.0 KB)
I looked at that spin dump and it looks like something happened with your font manager.
There’s nothing going on in GP that would cause it to suddenly get stuck after a few hours by itself. It’s just going to be sitting there waiting for MIDI or other triggers to start generating audio again.
Very strange
It happened while I was renaming markers (I renamed ~40 of them and suddenly this hang happened) so font manager might have been used to render marker’s label - but hard to tell.
I’m confused. Above you said that GP just hung after two hours when you were not using it
After this two hours I started using it. What I was trying to say that I observe the crashes / hangs more often after some time of inactivity, which might be complete coincidence but I mention it nevertheless because I don’t see any other pattern.