Summary: | [radeonsi] gtkperf hangs the system | ||
---|---|---|---|
Product: | DRI | Reporter: | darkbasic <darkbasic> |
Component: | DRM/Radeon | Assignee: | Default DRI bug account <dri-devel> |
Status: | RESOLVED MOVED | QA Contact: | |
Severity: | normal | ||
Priority: | medium | ||
Version: | XOrg git | ||
Hardware: | Other | ||
OS: | All | ||
Whiteboard: | |||
i915 platform: | i915 features: |
Description
darkbasic
2013-11-17 13:39:51 UTC
It is a regression because it used to work, not in the kernel becuase 3.12 still hangs. Can you narrow down which component (kernel, mesa, llvm, etc.) update caused the problem? It will not be that easy.. I don't use gtkperf since a few time and every single component of my graphic stack is pulled from git master. I recompiled kernel 3.12 which doesn't work but maybe the working one was 3.11, I tried downgrading glamor but X didn't start at all, maybe I needed to recompile xf86-video-ati. It's a time consuming process, being able to enable some kind of debug to have some hints would be preferable. It is probably the "GtkDrawingArea - Lines" component. The same problem is described in bug 68524. It is caused by the glamor software fallback being extremely slow. And obviously somehow mostly excluding other drawing operations. On my system, it doesn't crash the system, it only makes it extremely slow to respond. So slow that it doesn't even switch to a virtual terminal with CTRL+ALT+F1? I checked some of the tests I did in the past with 3.12-pre-rc1 and I didn't do the total time test (the one which currently hangs): http://openbenchmarking.org/result/1309135-SO-2D744911973 I was pretty sure I did it in the past, but considering I don't find a benchmark of the total time I assume to be wrong. > So slow that it doesn't even switch to a virtual terminal with CTRL+ALT+F1?
I could switch to a virtual terminal on my system. But X11 itself was frozen for a good while.
-- 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/drm/amd/issues/404. |
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.