Re: [RFC PATCH 0/2] ALSA: hda - acomp probe fix for i915

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Tue, 06 Oct 2020 14:13:40 +0200,
Takashi Iwai wrote:
> 
> On Tue, 06 Oct 2020 13:30:40 +0200,
> Kai Vehmanen wrote:
> > 
> > Hi,
> > 
> > this simple bugfix started to feel a bit like getting stuck in quicksand,
> > so I'm looking for some early input via this RFC series.
> > 
> > Basicly hdac_i915.c should not use global state to track communication
> > with i915 driver. But how to get handle of "hdac_bus*? I considered
> > a few options:
> > 
> >   1) add hdac_bus as a member of drm_audio_component.h
> > 	-> seems wrong as this is really an audio side implementation)
> > 
> >   2) embed copy of drm_audio_component to 'struct hdac_bus', so
> >      I could use container_of() on the device handle to get
> >      to the bus 
> > 	-> wasted space to keep a copy at hdac_bus level
> > 	   (note: snd-hda-codec-hdmi do this by embedding a copy
> > 	    of ops to "struct hdmi_spec")
> > 
> >   3) add another devres entry to store the hdac_bus directly
> >      in acomp_init and a new helper function to query it
> > 
> > I now implemented option 3 in this RFC series as it seemed cleanest
> > and most local to hdac_component.c, where the problem stems from. It's still
> > somewhat messy, and I'm wondering if I'm overlooking some obvious alternative.
> > We could dig this deeper into i915 specific code, but OTOH, hdac_bus is
> > an argument snd_hdac_acomp_init(), so it's common for all.
> > 
> > Kai Vehmanen (2):
> >   ALSA: hda - keep track of HDA core bus instance in acomp
> >   ALSA: hda/i915 - fix list corruption with concurrent probes
> 
> Another option would be to move the completion into the common acomp
> helper from i915-specific one.  That is,
> 
> - Add bind_complete field into struct drm_audio_component,
>   initialize it at snd_hdac_acomp_init()
> 
> - Call complete_all(&acomp->bind_complete) at the end of
>   hdac_component_master_bind()
> 
> - Remove / replace i915's own completion with the hdac's one.
>   The i915_init_ops can be dropped.

So something like below (totally untested).


Takashi

---
--- a/include/drm/drm_audio_component.h
+++ b/include/drm/drm_audio_component.h
@@ -117,6 +117,10 @@ struct drm_audio_component {
 	 * @audio_ops: Ops implemented by hda driver, called by DRM driver
 	 */
 	const struct drm_audio_component_audio_ops *audio_ops;
+	/**
+	 * @master_bind_complete: completion held during component master binding
+	 */
+	struct completion master_bind_complete;
 };
 
 #endif /* _DRM_AUDIO_COMPONENT_H_ */
--- a/sound/hda/hdac_component.c
+++ b/sound/hda/hdac_component.c
@@ -210,12 +210,14 @@ static int hdac_component_master_bind(struct device *dev)
 			goto module_put;
 	}
 
+	complete_all(&acomp->master_bind_complete);
 	return 0;
 
  module_put:
 	module_put(acomp->ops->owner);
 out_unbind:
 	component_unbind_all(dev, acomp);
+	complete_all(&acomp->master_bind_complete);
 
 	return ret;
 }
@@ -296,6 +298,7 @@ int snd_hdac_acomp_init(struct hdac_bus *bus,
 	if (!acomp)
 		return -ENOMEM;
 	acomp->audio_ops = aops;
+	init_completion(&acomp->master_bind_complete);
 	bus->audio_component = acomp;
 	devres_add(dev, acomp);
 
--- a/sound/hda/hdac_i915.c
+++ b/sound/hda/hdac_i915.c
@@ -11,8 +11,6 @@
 #include <sound/hda_i915.h>
 #include <sound/hda_register.h>
 
-static struct completion bind_complete;
-
 #define IS_HSW_CONTROLLER(pci) (((pci)->device == 0x0a0c) || \
 				((pci)->device == 0x0c0c) || \
 				((pci)->device == 0x0d0c) || \
@@ -130,19 +128,6 @@ static bool i915_gfx_present(void)
 	return pci_dev_present(ids);
 }
 
-static int i915_master_bind(struct device *dev,
-			    struct drm_audio_component *acomp)
-{
-	complete_all(&bind_complete);
-	/* clear audio_ops here as it was needed only for completion call */
-	acomp->audio_ops = NULL;
-	return 0;
-}
-
-static const struct drm_audio_component_audio_ops i915_init_ops = {
-	.master_bind = i915_master_bind
-};
-
 /**
  * snd_hdac_i915_init - Initialize i915 audio component
  * @bus: HDA core bus
@@ -163,9 +148,7 @@ int snd_hdac_i915_init(struct hdac_bus *bus)
 	if (!i915_gfx_present())
 		return -ENODEV;
 
-	init_completion(&bind_complete);
-
-	err = snd_hdac_acomp_init(bus, &i915_init_ops,
+	err = snd_hdac_acomp_init(bus, NULL,
 				  i915_component_master_match,
 				  sizeof(struct i915_audio_component) - sizeof(*acomp));
 	if (err < 0)
@@ -177,7 +160,7 @@ int snd_hdac_i915_init(struct hdac_bus *bus)
 		if (!IS_ENABLED(CONFIG_MODULES) ||
 		    !request_module("i915")) {
 			/* 60s timeout */
-			wait_for_completion_timeout(&bind_complete,
+			wait_for_completion_timeout(&acomp->master_bind_complete,
 						   msecs_to_jiffies(60 * 1000));
 		}
 	}



[Index of Archives]     [ALSA User]     [Linux Audio Users]     [Pulse Audio]     [Kernel Archive]     [Asterisk PBX]     [Photo Sharing]     [Linux Sound]     [Video 4 Linux]     [Gimp]     [Yosemite News]

  Powered by Linux