New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Alacritty is crashing rarely, but randomly #24521
Comments
New dmesg: |
Perhaps related, I have had this issue and an even more egregious variant where entire applications launched from alacritty would disappear. On Fedora 31 Gnome Wayland. Optimus laptop with zram constantly being utilized. Discrete GPU was turned off in bios at the time. Solution was to turn on the discrete GPU and use Nouveau. I suspect that for whatever reason the Intel GPU <-> system memory interaction was handled more poorly than with Nouveau somehow mediating it. The switch solved other graphical problems when memory was being full too (stuttering sound in videos + kernel cpu utilization through the roof), which ironically was what I wanted to avoid by turning off the discrete GPU to begin with. Alacritty uses gpu api (I recall being unable to even run it on some very old hardware because of opengl version or something) which probably makes it subject to GPU memory constraints in some way. |
I have had no problems like this with alacritty on bspwm, on a computer with only Intel internal graphics. |
Yeah, Musl. Do you think it could be blowing past the per thread stack, perhaps? |
The Fedora system I had this exact issue happen to wasn't musl. But I found a fix, zram with vm.watermark_scale_factor=800 |
There isn't really anything to act on, and I haven't experienced it for some time. @tarkov2213 interestingly I have also recently enabled zram |
System
Void 5.8.3_1 x86_64-musl GenuineIntel notuptodate rrnDDFFFFFFFFFFFFFFFFFFFFFFFFFFFF
alacritty-0.5.0_1
Expected behavior
It will stay open indefinitely.
Actual behavior
It crashes randomly, usually when I'm away from the device. I come back and the workspace occupied by alacritty is now empty. I managed to grab a core file and it doesn't have too much info, but notes that there was a
SIGILL
that killed the process.dmesg:
Interestingly enough, another alacritty instance just disappeared, without leaving complaints in dmesg or a core file. This is usually what happens.
Steps to reproduce the behavior
I'm opening this issue here first because we do patch stuff around the relevant toolchains, but I will also look into upstream.
The text was updated successfully, but these errors were encountered: