https://bugs.freedesktop.org/show_bug.cgi?id=79980
--- Comment #125 from Christian König deathsimple@vodafone.de --- (In reply to comment #124)
I can very quickly, almost deterministically, hang the GPU (radeonsi, Cape Verde) with the following command:
LIBGL_ALWAYS_SOFTWARE=1 mpv --fs --vo=opengl:sw /path/to/some_video
this works on both 3.16.0 and 3.17rc3. Try seeking, it often happens directly after a seek. In most cases, the hang is unrecoverable and crashes the kernel after some "atombios stuck in a loop" messages. Very strange indeed, software rendered glxgears doesn't cause this.
Can anyone verify? A somewhat reliable test case might be a good start to finally fixing this.
Well this is interesting, so you're saying that using software rendering on the client side can crash the GPU? That only leaves glamor and maybe the compositor as the only one using the hardware driver.