Resetting the engine requires us to hold the forcewake wakeref to prevent RC6 trying to happen in the middle of the reset sequence. Normally, this is taken by i915_handle_error(), but as we are calling the lowlevel functions ourselves, we need to hold it. Wrap the entire live_hangcheck set of subtests in a single forcewake section for simplicity. This greatly improves the reliability of drv_selftest/live_hangcheck on Haswell, where it would exhibit an inability to restart a request because it lost its PD registers (PD_DIR_BASE reported as 0). Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> Cc: Mika Kuoppala <mika.kuoppala@xxxxxxxxxxxxxxx> --- drivers/gpu/drm/i915/selftests/intel_hangcheck.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/selftests/intel_hangcheck.c b/drivers/gpu/drm/i915/selftests/intel_hangcheck.c index 7e1bdd88eda3..a9e0decccc1f 100644 --- a/drivers/gpu/drm/i915/selftests/intel_hangcheck.c +++ b/drivers/gpu/drm/i915/selftests/intel_hangcheck.c @@ -878,9 +878,18 @@ int intel_hangcheck_live_selftests(struct drm_i915_private *i915) SUBTEST(igt_reset_queue), SUBTEST(igt_handle_error), }; + int err; if (!intel_has_gpu_reset(i915)) return 0; - return i915_subtests(tests, i915); + intel_runtime_pm_get(i915); + intel_uncore_forcewake_get(i915, FORCEWAKE_ALL); + + err = i915_subtests(tests, i915); + + intel_uncore_forcewake_put(i915, FORCEWAKE_ALL); + intel_runtime_pm_put(i915); + + return err; } -- 2.14.2 _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx
