Overview
Request 1113092 revoked
This is the latest iteration of the upstream MR https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3293 that reliably works, that is I do not see the crashes on my wayland session any more. Note that gnome-shell needs to rebuilt against the patched mutter to work.
Probably should replace sr#1113073.
Thanks in advance.
- Created by badshah400
- In state revoked
Request History
badshah400 created request
This is the latest iteration of the upstream MR https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3293 that reliably works, that is I do not see the crashes on my wayland session any more. Note that gnome-shell needs to rebuilt against the patched mutter to work.
Probably should replace sr#1113073.
Thanks in advance.
gnome-review-bot accepted review
Check script succeeded
gnome-review-bot approved review
Check script succeeded
badshah400 revoked request
Revoking as this does not work either and isn't finalised upstream either.
Upstream MR as of this comment: https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3293#note_1851613
Not as reliably, unfortunately. I still get crashes after rebooting my system. Out of ideas at the moment.
Any idea if I need to build any other packages against the patched mutter? I build gnome-shell against it already, but that does not seem to be enough.
Is the previous sub stable on your end? (the one we have in GN right now). If yes, then we just go with the big hammer :-)
Not really, I can still reproduce the crash after a reboot or two... don't know if these are unrelated or not. Will be posting on the upstream thread to figure this out.
The only thing that does work for me all the time is doing:
sudo chrt -a -o -p 0 PID_OF_CURRENT_GNOME_SHELL_PROCESSafter each reboot.
I am sorry for the multiple subs and the associated confusion I seem to have created; as of now, I don't know of a working solution.
The only thing I can do at the moment is run this after every login to prevent the crash:
sudo chrt -a -o -p 0 PID_OF_CURRENT_GNOME_SHELL_PROCESSHi
Could you pull the latest checkin I did (the +45 one) - It brings in the work Ray Strode have been doing, and hopefully this fixes your issue.
Hi, many thanks! I am just testing your package and it is working great so far. Give me a day or two of testing and I shall let you know for sure. In the past, the issue would have reappeared after a reboot or two, so I want to put it through the paces a bit more.
Great - thanks for your feedback!
If you find this "working" I'll be pushing it to Factory :-)
Btw - do we have a "downstream" bug for this?
There is a related boo#1215695 but someone seems to have wrongly marked it (IMO) as duplicate of a kernel bug.
@iznogood I think it is green light from me! Many thanks for the update.