Created attachment 137587 [details] apitrace trace GPU: HD7750 Kernel: 4.14.17 (Debian package) Driver: both radeon and amdgpu Mesa: 17.3.4 LLVM: 5 When I start QT5 fork of celestia, my GPU gets locked. With radeon I have a few seconds to switch to a terminal, with amdgpu I don't, sometimes everything is locked so even REISUB doesn't help. Even apitrace dump-images locks. Feb 24 23:07:50 deneb kernel: [ 7701.664110] radeon 0000:01:00.0: ring 0 stalled for more than 10244msec Feb 24 23:07:50 deneb kernel: [ 7701.664118] radeon 0000:01:00.0: ring 3 stalled for more than 10244msec Feb 24 23:07:50 deneb kernel: [ 7701.664129] radeon 0000:01:00.0: GPU lockup (current fence id 0x0000000000168b3f last fence id 0x0000000000168cdb on ring 3) Feb 24 23:07:50 deneb kernel: [ 7701.664513] radeon 0000:01:00.0: GPU lockup (current fence id 0x00000000000ae7cc last fence id 0x00000000000ae85f on ring 0) Feb 24 23:07:50 deneb kernel: [ 7701.718269] do_general_protection: 15 callbacks suppressed Feb 24 23:07:50 deneb kernel: [ 7701.718272] traps: goldendict[4507] general protection ip:7f3bf7cc3185 sp:7ffedf2ba330 error:0 in libc-2.26.so[7f3bf7c8c000+1ad000] Feb 24 23:07:50 deneb kernel: [ 7701.936036] radeon_dp_aux_transfer_native: 74 callbacks suppressed Feb 24 23:07:50 deneb kernel: [ 7702.176091] radeon 0000:01:00.0: ring 3 stalled for more than 10756msec Feb 24 23:07:50 deneb kernel: [ 7702.176135] radeon 0000:01:00.0: ring 0 stalled for more than 10756msec Feb 24 23:07:50 deneb kernel: [ 7702.176142] radeon 0000:01:00.0: GPU lockup (current fence id 0x00000000000ae7cc last fence id 0x00000000000ae85f on ring 0) Feb 24 23:07:50 deneb kernel: [ 7702.176485] radeon 0000:01:00.0: GPU lockup (current fence id 0x0000000000168b3f last fence id 0x0000000000168ced on ring 3) Feb 24 23:07:51 deneb kernel: [ 7702.688091] radeon 0000:01:00.0: ring 0 stalled for more than 11268msec Feb 24 23:07:51 deneb kernel: [ 7702.688096] radeon 0000:01:00.0: ring 3 stalled for more than 11268msec
Works on Devuan 2 (Debian 9) with Linux 4.9 and 4.15 and Mesa 13.0.6, so I assume that the problem is in Mesa.
-- GitLab Migration Automatic Message -- This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity. You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/mesa/mesa/issues/1305.
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.