From: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> commit 98479ada421a8fd2123b98efd398a6f1379307ab upstream. If we park/unpark faster than we can respond to RPS events, we never will process a downclock event after expiring a waitboost, and thus we will forever restart the GPU at max clocks even if the workload switches and doesn't justify full power. Closes: https://gitlab.freedesktop.org/drm/intel/issues/1500 Fixes: 3e7abf814193 ("drm/i915: Extract GT render power state management") Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> Cc: Andi Shyti <andi.shyti@xxxxxxxxx> Cc: Lyude Paul <lyude@xxxxxxxxxx> Reviewed-by: Andi Shyti <andi.shyti@xxxxxxxxx> Link: https://patchwork.freedesktop.org/patch/msgid/20200322163225.28791-1-chris@xxxxxxxxxxxxxxxxxx Cc: <stable@xxxxxxxxxxxxxxx> # v5.5+ (cherry picked from commit 21abf0bf168dffff1192e0f072af1dc74ae1ff0e) Signed-off-by: Rodrigo Vivi <rodrigo.vivi@xxxxxxxxx> Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> --- drivers/gpu/drm/i915/gt/intel_rps.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) --- a/drivers/gpu/drm/i915/gt/intel_rps.c +++ b/drivers/gpu/drm/i915/gt/intel_rps.c @@ -763,6 +763,19 @@ void intel_rps_park(struct intel_rps *rp intel_uncore_forcewake_get(rps_to_uncore(rps), FORCEWAKE_MEDIA); rps_set(rps, rps->idle_freq); intel_uncore_forcewake_put(rps_to_uncore(rps), FORCEWAKE_MEDIA); + + /* + * Since we will try and restart from the previously requested + * frequency on unparking, treat this idle point as a downclock + * interrupt and reduce the frequency for resume. If we park/unpark + * more frequently than the rps worker can run, we will not respond + * to any EI and never see a change in frequency. + * + * (Note we accommodate Cherryview's limitation of only using an + * even bin by applying it to all.) + */ + rps->cur_freq = + max_t(int, round_down(rps->cur_freq - 1, 2), rps->min_freq); } void intel_rps_boost(struct i915_request *rq)
