[Intel-gfx] [PATCH] drm/i915: Extend rpm wakelock during i915_handle_error()

[Chris Wilson]

On Tue, Mar 14, 2017 at 10:24:42AM -0700, Michel Thierry wrote:
> 
> 
> On 3/14/2017 10:18 AM, Chris Wilson wrote:
> >We take the runtime pm wakelock during i915_handle_error() to ensure
> >that all paths that reach the error capture keep the device awake during
> >the hw reads. However, we need to extend that from the reset handler to
> >include the earlier capture routines.
> >
> >Reported-by: Antonio Argenziano <antonio.argenziano at intel.com>
> >Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
> >Cc: Michel Thierry <michel.thierry at intel.com>
> >---
> > drivers/gpu/drm/i915/i915_irq.c | 25 ++++++++++++++-----------
> > 1 file changed, 14 insertions(+), 11 deletions(-)
> >
> >diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
> >index 31f0d7c8992f..52e1fe8a98fb 100644
> >--- a/drivers/gpu/drm/i915/i915_irq.c
> >+++ b/drivers/gpu/drm/i915/i915_irq.c
> >@@ -2664,14 +2664,6 @@ static void i915_reset_and_wakeup(struct drm_i915_private *dev_priv)
> > 	DRM_DEBUG_DRIVER("resetting chip\n");
> > 	kobject_uevent_env(kobj, KOBJ_CHANGE, reset_event);
> >
> >-	/*
> >-	 * In most cases it's guaranteed that we get here with an RPM
> >-	 * reference held, for example because there is a pending GPU
> >-	 * request that won't finish until the reset is done. This
> >-	 * isn't the case at least when we get here by doing a
> >-	 * simulated reset via debugs, so get an RPM reference.
> >-	 */
> >-	intel_runtime_pm_get(dev_priv);
> > 	intel_prepare_reset(dev_priv);
> >
> > 	do {
> >@@ -2693,7 +2685,6 @@ static void i915_reset_and_wakeup(struct drm_i915_private *dev_priv)
> > 				     HZ));
> >
> > 	intel_finish_reset(dev_priv);
> >-	intel_runtime_pm_put(dev_priv);
> >
> > 	if (!test_bit(I915_WEDGED, &dev_priv->gpu_error.flags))
> > 		kobject_uevent_env(kobj,
> >@@ -2780,15 +2771,24 @@ void i915_handle_error(struct drm_i915_private *dev_priv,
> > 	vscnprintf(error_msg, sizeof(error_msg), fmt, args);
> > 	va_end(args);
> >
> >+	/*
> >+	 * In most cases it's guaranteed that we get here with an RPM
> >+	 * reference held, for example because there is a pending GPU
> >+	 * request that won't finish until the reset is done. This
> >+	 * isn't the case at least when we get here by doing a
> >+	 * simulated reset via debugfs, so get an RPM reference.
> >+	 */
> >+	intel_runtime_pm_get(dev_priv);
> >+
> > 	i915_capture_error_state(dev_priv, engine_mask, error_msg);
> > 	i915_clear_error_registers(dev_priv);
> >
> > 	if (!engine_mask)
> >-		return;
> >+		goto out;
> >
> > 	if (test_and_set_bit(I915_RESET_IN_PROGRESS,
> > 			     &dev_priv->gpu_error.flags))
> >-		return;
> >+		goto out;
> >
> > 	/*
> > 	 * Wakeup waiting processes so that the reset function
> >@@ -2805,6 +2805,9 @@ void i915_handle_error(struct drm_i915_private *dev_priv,
> > 	i915_error_wake_up(dev_priv);
> >
> > 	i915_reset_and_wakeup(dev_priv);
> >+
> >+out:
> >+	intel_runtime_pm_put(dev_priv);
> > }
> >
> > /* Called from drm generic code, passed 'crtc' which
> >
> 
> Reviewed-by: Michel Thierry <michel.thierry at intel.com>

Thanks for the original patch, now CI has an idle moment, pushed.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre