During our selftests, we try reseting the GPU tens of thousands of
times, flooding the dmesg with out reset spam drowning out any potential
warnings. Add an option to i915_reset()/i915_reset_engine() to specify a
quiet reset for selftesting.
Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
---
drivers/gpu/drm/i915/i915_drv.c | 48 +++++++++++++-----------
drivers/gpu/drm/i915/i915_drv.h | 8 +++-
drivers/gpu/drm/i915/i915_gem_request.c | 2 +-
drivers/gpu/drm/i915/i915_irq.c | 4 +-
drivers/gpu/drm/i915/selftests/intel_hangcheck.c | 8 ++--
5 files changed, 39 insertions(+), 31 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 3ac8215c0e36..214555e813f1 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -1831,7 +1831,8 @@ static int i915_resume_switcheroo(struct drm_device *dev)
/**
* i915_reset - reset chip after a hang
- * @dev_priv: device private to reset
+ * @i915: #drm_i915_private to reset
+ * @flags: Instructions
*
* Reset the chip. Useful if a hang is detected. Marks the device as wedged
* on failure.
@@ -1846,34 +1847,34 @@ static int i915_resume_switcheroo(struct drm_device *dev)
* - re-init interrupt state
* - re-init display
*/
-void i915_reset(struct drm_i915_private *dev_priv)
+void i915_reset(struct drm_i915_private *i915, unsigned int flags)
{
- struct i915_gpu_error *error = &dev_priv->gpu_error;
+ struct i915_gpu_error *error = &i915->gpu_error;
int ret;
- lockdep_assert_held(&dev_priv->drm.struct_mutex);
+ lockdep_assert_held(&i915->drm.struct_mutex);
GEM_BUG_ON(!test_bit(I915_RESET_BACKOFF, &error->flags));
if (!test_bit(I915_RESET_HANDOFF, &error->flags))
return;
/* Clear any previous failed attempts at recovery. Time to try again. */
- if (!i915_gem_unset_wedged(dev_priv))
+ if (!i915_gem_unset_wedged(i915))
goto wakeup;
- dev_notice(dev_priv->drm.dev,
- "Resetting chip after gpu hang\n");
+ if (!(flags & I915_RESET_QUIET))
+ dev_notice(i915->drm.dev, "Resetting chip after gpu hang\n");
error->reset_count++;
- disable_irq(dev_priv->drm.irq);
- ret = i915_gem_reset_prepare(dev_priv);
+ disable_irq(i915->drm.irq);
+ ret = i915_gem_reset_prepare(i915);
if (ret) {
DRM_ERROR("GPU recovery failed\n");
- intel_gpu_reset(dev_priv, ALL_ENGINES);
+ intel_gpu_reset(i915, ALL_ENGINES);
goto error;
}
- ret = intel_gpu_reset(dev_priv, ALL_ENGINES);
+ ret = intel_gpu_reset(i915, ALL_ENGINES);
if (ret) {
if (ret != -ENODEV)
DRM_ERROR("Failed to reset chip: %i\n", ret);
@@ -1882,8 +1883,8 @@ void i915_reset(struct drm_i915_private *dev_priv)
goto error;
}
- i915_gem_reset(dev_priv);
- intel_overlay_reset(dev_priv);
+ i915_gem_reset(i915);
+ intel_overlay_reset(i915);
/* Ok, now get things going again... */
@@ -1899,17 +1900,17 @@ void i915_reset(struct drm_i915_private *dev_priv)
* was running at the time of the reset (i.e. we weren't VT
* switched away).
*/
- ret = i915_gem_init_hw(dev_priv);
+ ret = i915_gem_init_hw(i915);
if (ret) {
DRM_ERROR("Failed hw init on reset %d\n", ret);
goto error;
}
- i915_queue_hangcheck(dev_priv);
+ i915_queue_hangcheck(i915);
finish:
- i915_gem_reset_finish(dev_priv);
- enable_irq(dev_priv->drm.irq);
+ i915_gem_reset_finish(i915);
+ enable_irq(i915->drm.irq);
wakeup:
clear_bit(I915_RESET_HANDOFF, &error->flags);
@@ -1917,14 +1918,15 @@ void i915_reset(struct drm_i915_private *dev_priv)
return;
error:
- i915_gem_set_wedged(dev_priv);
- i915_gem_retire_requests(dev_priv);
+ i915_gem_set_wedged(i915);
+ i915_gem_retire_requests(i915);
goto finish;
}
/**
* i915_reset_engine - reset GPU engine to recover from a hang
* @engine: engine to reset
+ * @flags: options
*
* Reset a specific GPU engine. Useful if a hang is detected.
* Returns zero on successful reset or otherwise an error code.
@@ -1934,7 +1936,7 @@ void i915_reset(struct drm_i915_private *dev_priv)
* - reset engine (which will force the engine to idle)
* - re-init/configure engine
*/
-int i915_reset_engine(struct intel_engine_cs *engine)
+int i915_reset_engine(struct intel_engine_cs *engine, unsigned int flags)
{
struct i915_gpu_error *error = &engine->i915->gpu_error;
struct drm_i915_gem_request *active_request;
@@ -1942,8 +1944,10 @@ int i915_reset_engine(struct intel_engine_cs *engine)
GEM_BUG_ON(!test_bit(I915_RESET_ENGINE + engine->id, &error->flags));
- dev_notice(engine->i915->drm.dev,
- "Resetting %s after gpu hang\n", engine->name);
+ if (!(flags & I915_RESET_QUIET)) {
+ dev_notice(engine->i915->drm.dev,
+ "Resetting %s after gpu hang\n", engine->name);
+ }
error->reset_engine_count[engine->id]++;
active_request = i915_gem_reset_prepare_engine(engine);
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index b051122c960b..619f756a83cd 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -3099,8 +3099,12 @@ extern int i915_driver_load(struct pci_dev *pdev,
extern void i915_driver_unload(struct drm_device *dev);
extern int intel_gpu_reset(struct drm_i915_private *dev_priv, u32 engine_mask);
extern bool intel_has_gpu_reset(struct drm_i915_private *dev_priv);
-extern void i915_reset(struct drm_i915_private *dev_priv);
-extern int i915_reset_engine(struct intel_engine_cs *engine);
+
+#define I915_RESET_QUIET BIT(0)
+extern void i915_reset(struct drm_i915_private *i915, unsigned int flags);
+extern int i915_reset_engine(struct intel_engine_cs *engine,
+ unsigned int flags);
+
extern bool intel_has_reset_engine(struct drm_i915_private *dev_priv);
extern int intel_guc_reset(struct drm_i915_private *dev_priv);
extern void intel_engine_init_hangcheck(struct intel_engine_cs *engine);
diff --git a/drivers/gpu/drm/i915/i915_gem_request.c b/drivers/gpu/drm/i915/i915_gem_request.c
index 68e406a53c04..9eedd33eb524 100644
--- a/drivers/gpu/drm/i915/i915_gem_request.c
+++ b/drivers/gpu/drm/i915/i915_gem_request.c
@@ -1071,7 +1071,7 @@ static bool __i915_wait_request_check_and_reset(struct drm_i915_gem_request *req
return false;
__set_current_state(TASK_RUNNING);
- i915_reset(request->i915);
+ i915_reset(request->i915, 0);
return true;
}
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 5244fb16fa55..469f854dd395 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -2624,7 +2624,7 @@ static void i915_reset_device(struct drm_i915_private *dev_priv)
*/
do {
if (mutex_trylock(&dev_priv->drm.struct_mutex)) {
- i915_reset(dev_priv);
+ i915_reset(dev_priv, 0);
mutex_unlock(&dev_priv->drm.struct_mutex);
}
} while (wait_on_bit_timeout(&dev_priv->gpu_error.flags,
@@ -2739,7 +2739,7 @@ void i915_handle_error(struct drm_i915_private *dev_priv,
&dev_priv->gpu_error.flags))
continue;
- if (i915_reset_engine(engine) == 0)
+ if (i915_reset_engine(engine, 0) == 0)
engine_mask &= ~intel_engine_flag(engine);
clear_bit(I915_RESET_ENGINE + engine->id,
diff --git a/drivers/gpu/drm/i915/selftests/intel_hangcheck.c b/drivers/gpu/drm/i915/selftests/intel_hangcheck.c
index e83503db3b66..4129d01b67a1 100644
--- a/drivers/gpu/drm/i915/selftests/intel_hangcheck.c
+++ b/drivers/gpu/drm/i915/selftests/intel_hangcheck.c
@@ -342,7 +342,7 @@ static int igt_global_reset(void *arg)
mutex_lock(&i915->drm.struct_mutex);
reset_count = i915_reset_count(&i915->gpu_error);
- i915_reset(i915);
+ i915_reset(i915, I915_RESET_QUIET);
if (i915_reset_count(&i915->gpu_error) == reset_count) {
pr_err("No GPU reset recorded!\n");
@@ -378,7 +378,7 @@ static int igt_reset_engine(void *arg)
reset_engine_count = i915_reset_engine_count(&i915->gpu_error,
engine);
- err = i915_reset_engine(engine);
+ err = i915_reset_engine(engine, I915_RESET_QUIET);
if (err) {
pr_err("i915_reset_engine failed\n");
break;
@@ -512,7 +512,7 @@ static int igt_reset_active_engines(void *arg)
set_bit(I915_RESET_ENGINE + engine->id, &i915->gpu_error.flags);
do {
- err = i915_reset_engine(engine);
+ err = i915_reset_engine(engine, I915_RESET_QUIET);
if (err) {
pr_err("i915_reset_engine(%s) failed, err=%d\n",
engine->name, err);
@@ -716,7 +716,7 @@ static int igt_reset_queue(void *arg)
reset_count = fake_hangcheck(prev);
- i915_reset(i915);
+ i915_reset(i915, I915_RESET_QUIET);
GEM_BUG_ON(test_bit(I915_RESET_HANDOFF,
&i915->gpu_error.flags));
--
2.13.3
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
