Created attachment 116819 [details] GPU crash dump from /sys/class/drm/card0/error Dmesg shows the following when comming out of powershave (screensaver). Unsure what caused this, but the dmesg told me to file a bug report here, including attaching the crashdump. [ 8835.191819] intel_powerclamp: Start idle injection to reduce power [ 8867.237426] intel_powerclamp: Stop forced idle injection [ 9035.893149] intel_powerclamp: Start idle injection to reduce power [ 9069.214468] intel_powerclamp: Stop forced idle injection [ 9341.965675] [drm] stuck on render ring [ 9341.966371] [drm] GPU HANG: ecode 6:0:0x94eeffb8, in compiz [18008], reason: Ring hung, action: reset [ 9341.966372] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace. [ 9341.966373] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel [ 9341.966374] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue. [ 9341.966374] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it. [ 9341.966376] [drm] GPU crash dump saved to /sys/class/drm/card0/error [ 9341.981630] drm/i915: Resetting chip after gpu hang [ 9843.791418] [drm] stuck on render ring [ 9843.792079] [drm] GPU HANG: ecode 6:0:0x94eeffb8, in compiz [18008], reason: Ring hung, action: reset [ 9843.799419] drm/i915: Resetting chip after gpu hang
In the middle of a snb blorp hiz resolve?
@Chris Wilson To be honest, I'm not sure what you're asking me. I was using dmesg to find the usb-stick I had just connected when I saw this message. As the dmesg output requested to send the crash dump, I did this. The only thing I can tell you is that the timestamp corresponds with logging in after a coffee break. What I see in dmesg is that the GPU HANG: ecode 6:0:0x94eeffb8 does repeat (3 times in the last 15 minutes). Only one of these GPU hangs (the first one) actually asks me to send the crash dump.
We don't have enough information. - What kernel version are you using? (uname -a) - What Mesa version are you using? (glxinfo | grep Mesa) blorp (a component of the i965 driver) has undergone enormous changes since this bug was filed. Please test a new version of Mesa (12 or 13) and mark as REOPENED if you can reproduce and RESOLVED/* if you cannot reproduce.
I have not seen this issue again and cannot reproduce. I assume that this has dissapeared when using a newer kernel (and possibly newer Mesa version as well). I had a clean and up-to-date ubuntu 15.04 on the machine at the time on an Asus n53sv laptop, which is no longer in my posession.
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.