Bug 112921
Summary: | [radeon] Lock-up after screensaver | ||
---|---|---|---|
Product: | Drivers | Reporter: | Jean Delvare (jdelvare) |
Component: | Video(DRI - non Intel) | Assignee: | drivers_video-dri |
Status: | RESOLVED CODE_FIX | ||
Severity: | normal | CC: | alexdeucher |
Priority: | P1 | ||
Hardware: | x86-64 | ||
OS: | Linux | ||
Kernel Version: | 4.4 | Subsystem: | |
Regression: | Yes | Bisected commit-id: | |
Attachments: |
dmesg output
Xorg.0.log |
Description
Jean Delvare
2016-02-23 09:23:36 UTC
I compared the boot messages from kernels v4.3.5 and v4.4.2. The only difference is the following error in v4.3.5 which is no longer present in v4.4.2: [drm:btc_dpm_set_power_state [radeon]] *ERROR* rv770_set_sw_state failed I have no idea if this is related. Created attachment 204371 [details]
dmesg output
Created attachment 204381 [details]
Xorg.0.log
This looks like the vblank regression bug. See this thread for more info: https://lists.freedesktop.org/archives/dri-devel/2016-January/098823.html and the fdo bug: https://bugs.freedesktop.org/show_bug.cgi?id=93746 Thanks for the pointer Alex, I'll test Mario's patch right now. Unfortunately, Mario's patch does not fix my problem. Which patch did you try? Mario actually came up with a number of separate fixes to address this regression. I think all of them should be in Linus' tree by now (for 4.5). I used the patch attached in bugzilla comment https://bugs.freedesktop.org/show_bug.cgi?id=93746#c18 on top of v4.4.3. This is the same patch that went upstream. I'll test v4.4.4 next, as I see many radeon fixes went into this version. I've been running v4.4.4 for some time now and the lock-up never happened. So I guess it's fixed there, thanks. |