From 3744d49c6b2d2d08f06715285196a946215887ba Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Fri, 21 Jul 2017 13:32:35 +0100 Subject: [PATCH] drm/i915/selftest: Refactor reset locking Git-commit: 3744d49c6b2d2d08f06715285196a946215887ba Patch-mainline: v4.14-rc1 References: FATE#322643 bsc#1055900 Extract the common barrier against rogue hangchecks from disrupting our direct testing of resets, and in the process expand the lock to include the per-engine reset shortcuts. Signed-off-by: Chris Wilson Cc: Mika Kuoppala Cc: Michel Thierry Link: https://patchwork.freedesktop.org/patch/msgid/20170721123238.16428-17-chris@chris-wilson.co.uk Reviewed-by: Michel Thierry Signed-off-by: Daniel Vetter Acked-by: Takashi Iwai --- drivers/gpu/drm/i915/selftests/intel_hangcheck.c | 58 +++++++++++++++++------ 1 file changed, 43 insertions(+), 15 deletions(-) --- a/drivers/gpu/drm/i915/selftests/intel_hangcheck.c +++ b/drivers/gpu/drm/i915/selftests/intel_hangcheck.c @@ -297,6 +297,37 @@ unlock: return err; } +static void global_reset_lock(struct drm_i915_private *i915) +{ + struct intel_engine_cs *engine; + enum intel_engine_id id; + + while (test_and_set_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags)) + wait_event(i915->gpu_error.reset_queue, + !test_bit(I915_RESET_BACKOFF, + &i915->gpu_error.flags)); + + for_each_engine(engine, i915, id) { + while (test_and_set_bit(I915_RESET_ENGINE + id, + &i915->gpu_error.flags)) + wait_on_bit(&i915->gpu_error.flags, + I915_RESET_ENGINE + id, + TASK_UNINTERRUPTIBLE); + } +} + +static void global_reset_unlock(struct drm_i915_private *i915) +{ + struct intel_engine_cs *engine; + enum intel_engine_id id; + + for_each_engine(engine, i915, id) + clear_bit(I915_RESET_ENGINE + id, &i915->gpu_error.flags); + + clear_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags); + wake_up_all(&i915->gpu_error.reset_queue); +} + static int igt_global_reset(void *arg) { struct drm_i915_private *i915 = arg; @@ -305,7 +336,7 @@ static int igt_global_reset(void *arg) /* Check that we can issue a global GPU reset */ - set_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags); + global_reset_lock(i915); set_bit(I915_RESET_HANDOFF, &i915->gpu_error.flags); mutex_lock(&i915->drm.struct_mutex); @@ -320,8 +351,7 @@ static int igt_global_reset(void *arg) mutex_unlock(&i915->drm.struct_mutex); GEM_BUG_ON(test_bit(I915_RESET_HANDOFF, &i915->gpu_error.flags)); - clear_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags); - wake_up_all(&i915->gpu_error.reset_queue); + global_reset_unlock(i915); if (i915_terminally_wedged(&i915->gpu_error)) err = -EIO; @@ -571,7 +601,7 @@ static int igt_wait_reset(void *arg) /* Check that we detect a stuck waiter and issue a reset */ - set_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags); + global_reset_lock(i915); mutex_lock(&i915->drm.struct_mutex); err = hang_init(&h, i915); @@ -616,8 +646,7 @@ fini: hang_fini(&h); unlock: mutex_unlock(&i915->drm.struct_mutex); - clear_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags); - wake_up_all(&i915->gpu_error.reset_queue); + global_reset_unlock(i915); if (i915_terminally_wedged(&i915->gpu_error)) return -EIO; @@ -638,7 +667,8 @@ static int igt_reset_queue(void *arg) if (!igt_can_mi_store_dword_imm(i915)) return 0; - set_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags); + global_reset_lock(i915); + mutex_lock(&i915->drm.struct_mutex); err = hang_init(&h, i915); if (err) @@ -732,8 +762,7 @@ fini: hang_fini(&h); unlock: mutex_unlock(&i915->drm.struct_mutex); - clear_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags); - wake_up_all(&i915->gpu_error.reset_queue); + global_reset_unlock(i915); if (i915_terminally_wedged(&i915->gpu_error)) return -EIO; @@ -755,7 +784,8 @@ static int igt_render_engine_reset_fallb if (!intel_has_reset_engine(i915)) return 0; - set_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags); + global_reset_lock(i915); + mutex_lock(&i915->drm.struct_mutex); err = hang_init(&h, i915); @@ -785,8 +815,7 @@ static int igt_render_engine_reset_fallb /* unlock since we'll call handle_error */ mutex_unlock(&i915->drm.struct_mutex); - clear_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags); - wake_up_all(&i915->gpu_error.reset_queue); + global_reset_unlock(i915); i915_handle_error(i915, intel_engine_flag(engine), "live test"); @@ -808,7 +837,7 @@ static int igt_render_engine_reset_fallb * more full reset to re-enable the hw. */ if (i915_terminally_wedged(&i915->gpu_error)) { - set_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags); + global_reset_lock(i915); rq->fence.error = 0; mutex_lock(&i915->drm.struct_mutex); @@ -829,8 +858,7 @@ out_rq: i915_gem_request_put(rq); hang_fini(&h); out_backoff: - clear_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags); - wake_up_all(&i915->gpu_error.reset_queue); + global_reset_unlock(i915); if (i915_terminally_wedged(&i915->gpu_error)) return -EIO;