drm/i915: POWER_DOMAIN_AUDIO ref-count debug logs
diff mbox series

Message ID 20200617095001.19220-1-anshuman.gupta@intel.com
State New
Headers show
Series
  • drm/i915: POWER_DOMAIN_AUDIO ref-count debug logs
Related show

Commit Message

Anshuman Gupta June 17, 2020, 9:50 a.m. UTC
Debug print for power domain audio get/put.
This will help to deubg the CI s2idle incomplete
failures.

Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com>
---
 drivers/gpu/drm/i915/display/intel_audio.c | 2 ++
 1 file changed, 2 insertions(+)

Comments

Chris Wilson June 17, 2020, 10:12 a.m. UTC | #1
Quoting Anshuman Gupta (2020-06-17 10:50:01)
> Debug print for power domain audio get/put.
> This will help to deubg the CI s2idle incomplete
> failures.

Do we not already print the mismatching pm, and who we are unable to
find a corresponding release for?
-Chris
Anshuman Gupta June 17, 2020, 10:17 a.m. UTC | #2
On 2020-06-17 at 11:12:12 +0100, Chris Wilson wrote:
> Quoting Anshuman Gupta (2020-06-17 10:50:01)
> > Debug print for power domain audio get/put.
> > This will help to deubg the CI s2idle incomplete
> > failures.
> 
> Do we not already print the mismatching pm, and who we are unable to
> find a corresponding release for?
Thanks Chris for review comment,
Yes it is there, but POWER_DOMAIN_AUDIO get/put request initiated by
snd_audio along with i915. So in order to confirm if it is a bug from
snd_audio module we would require to track the get/put resuest from
snd_audio.
Thanks,
Anshuman Gupta.
> -Chris
Chris Wilson June 17, 2020, 10:46 a.m. UTC | #3
Quoting Anshuman Gupta (2020-06-17 11:17:15)
> On 2020-06-17 at 11:12:12 +0100, Chris Wilson wrote:
> > Quoting Anshuman Gupta (2020-06-17 10:50:01)
> > > Debug print for power domain audio get/put.
> > > This will help to deubg the CI s2idle incomplete
> > > failures.
> > 
> > Do we not already print the mismatching pm, and who we are unable to
> > find a corresponding release for?
> Thanks Chris for review comment,
> Yes it is there, but POWER_DOMAIN_AUDIO get/put request initiated by
> snd_audio along with i915. So in order to confirm if it is a bug from
> snd_audio module we would require to track the get/put resuest from
> snd_audio.

But we are, see the cookie.
-Chris

Patch
diff mbox series

diff --git a/drivers/gpu/drm/i915/display/intel_audio.c b/drivers/gpu/drm/i915/display/intel_audio.c
index ad4aa66fd676..145afd2f1d4a 100644
--- a/drivers/gpu/drm/i915/display/intel_audio.c
+++ b/drivers/gpu/drm/i915/display/intel_audio.c
@@ -1012,6 +1012,7 @@  static unsigned long i915_audio_component_get_power(struct device *kdev)
 	/* Catch potential impedance mismatches before they occur! */
 	BUILD_BUG_ON(sizeof(intel_wakeref_t) > sizeof(unsigned long));
 
+	drm_dbg_kms(&dev_priv->drm, "get audio power domian power\n");
 	ret = intel_display_power_get(dev_priv, POWER_DOMAIN_AUDIO);
 
 	if (dev_priv->audio_power_refcount++ == 0) {
@@ -1045,6 +1046,7 @@  static void i915_audio_component_put_power(struct device *kdev,
 		if (IS_GEMINILAKE(dev_priv))
 			glk_force_audio_cdclk(dev_priv, false);
 
+	drm_dbg_kms(&dev_priv->drm, "put audio power domian power\n");
 	intel_display_power_put(dev_priv, POWER_DOMAIN_AUDIO, cookie);
 }