On Fri, May 1, 2020 at 11:26 AM Jordan Crouse <jcrouse@xxxxxxxxxxxxxx> wrote: > > Writing to the devfreq sysfs nodes while the GPU is powered down can > result in a system crash (on a5xx) or a nasty GMU error (on a6xx): > > $ /sys/class/devfreq/5000000.gpu# echo 500000000 > min_freq > [ 104.841625] platform 506a000.gmu: [drm:a6xx_gmu_set_oob] > *ERROR* Timeout waiting for GMU OOB set GPU_DCVS: 0x0 > > Despite the fact that we carefully try to suspend the devfreq device when > the hardware is powered down there are lots of holes in the governors that > don't check for the suspend state and blindly call into the devfreq > callbacks that end up triggering hardware reads in the GPU driver. > > Check the power state in the gpu_busy() and gpu_set_freq() callbacks for > a5xx and a6xx to make sure that the hardware is active before trying to > access it. Chatted on IRC -- while this avoids the instaboot on db820c when setting /sys/class/devfreq/devfreq1/min_freq, I think we should be using pm_runtime_get_if_in_use() to avoid the races while still avoiding bringing up the GPU.
