[PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Shuah Khan]

Change ALSA driver to use Media Controller API to
share media resources with DVB and V4L2 drivers
on a AU0828 media device. Media Controller specific
initialization is done after sound card is registered.
ALSA creates Media interface and entity function graph
nodes for Control, Mixer, PCM Playback, and PCM Capture
devices.

snd_usb_hw_params() will call Media Controller enable
source handler interface to request the media resource.
If resource request is granted, it will release it from
snd_usb_hw_free(). If resource is busy, -EBUSY is returned.

Media specific cleanup is done in usb_audio_disconnect().

Signed-off-by: Shuah Khan <shuahkh@osg.samsung.com>
Acked-by: Takashi Iwai <tiwai@suse.de>
---

Changes since v4:
1. Renamed media_* interfaces in sound/usb to media_snd*
   to avoid potential conflicts with the MC interfaces.
   This addresses Laurent Pinchart's comment.
2. Added Acked-by from Takashi Iwai

 sound/usb/Kconfig        |   4 +
 sound/usb/Makefile       |   2 +
 sound/usb/card.c         |  14 +++
 sound/usb/card.h         |   3 +
 sound/usb/media.c        | 318 +++++++++++++++++++++++++++++++++++++++++++++++
 sound/usb/media.h        |  72 +++++++++++
 sound/usb/mixer.h        |   3 +
 sound/usb/pcm.c          |  28 ++++-
 sound/usb/quirks-table.h |   1 +
 sound/usb/stream.c       |   2 +
 sound/usb/usbaudio.h     |   6 +
 11 files changed, 448 insertions(+), 5 deletions(-)
 create mode 100644 sound/usb/media.c
 create mode 100644 sound/usb/media.h

diff --git a/sound/usb/Kconfig b/sound/usb/Kconfig
index a452ad7..d14bf41 100644
--- a/sound/usb/Kconfig
+++ b/sound/usb/Kconfig
@@ -15,6 +15,7 @@ config SND_USB_AUDIO
 	select SND_RAWMIDI
 	select SND_PCM
 	select BITREVERSE
+	select SND_USB_AUDIO_USE_MEDIA_CONTROLLER if MEDIA_CONTROLLER && (MEDIA_SUPPORT=y || MEDIA_SUPPORT=SND_USB_AUDIO)
 	help
 	  Say Y here to include support for USB audio and USB MIDI
 	  devices.
@@ -22,6 +23,9 @@ config SND_USB_AUDIO
 	  To compile this driver as a module, choose M here: the module
 	  will be called snd-usb-audio.
 
+config SND_USB_AUDIO_USE_MEDIA_CONTROLLER
+	bool
+
 config SND_USB_UA101
 	tristate "Edirol UA-101/UA-1000 driver"
 	select SND_PCM
diff --git a/sound/usb/Makefile b/sound/usb/Makefile
index 2d2d122..8dca3c4 100644
--- a/sound/usb/Makefile
+++ b/sound/usb/Makefile
@@ -15,6 +15,8 @@ snd-usb-audio-objs := 	card.o \
 			quirks.o \
 			stream.o
 
+snd-usb-audio-$(CONFIG_SND_USB_AUDIO_USE_MEDIA_CONTROLLER) += media.o
+
 snd-usbmidi-lib-objs := midi.o
 
 # Toplevel Module Dependency
diff --git a/sound/usb/card.c b/sound/usb/card.c
index 1f09d95..258cf70 100644
--- a/sound/usb/card.c
+++ b/sound/usb/card.c
@@ -66,6 +66,7 @@
 #include "format.h"
 #include "power.h"
 #include "stream.h"
+#include "media.h"
 
 MODULE_AUTHOR("Takashi Iwai <tiwai@suse.de>");
 MODULE_DESCRIPTION("USB Audio");
@@ -561,6 +562,11 @@ static int usb_audio_probe(struct usb_interface *intf,
 	if (err < 0)
 		goto __error;
 
+	if (quirk->media_device) {
+		/* don't want to fail when media_snd_device_create() fails */
+		media_snd_device_create(chip, intf);
+	}
+
 	usb_chip[chip->index] = chip;
 	chip->num_interfaces++;
 	usb_set_intfdata(intf, chip);
@@ -617,6 +623,14 @@ static void usb_audio_disconnect(struct usb_interface *intf)
 		list_for_each(p, &chip->midi_list) {
 			snd_usbmidi_disconnect(p);
 		}
+		/*
+		 * Nice to check quirk && quirk->media_device
+		 * need some special handlings. Doesn't look like
+		 * we have access to quirk here
+		 * Acceses mixer_list
+		*/
+		media_snd_device_delete(chip);
+
 		/* release mixer resources */
 		list_for_each_entry(mixer, &chip->mixer_list, list) {
 			snd_usb_mixer_disconnect(mixer);
diff --git a/sound/usb/card.h b/sound/usb/card.h
index 71778ca..34a0898 100644
--- a/sound/usb/card.h
+++ b/sound/usb/card.h
@@ -105,6 +105,8 @@ struct snd_usb_endpoint {
 	struct list_head list;
 };
 
+struct media_ctl;
+
 struct snd_usb_substream {
 	struct snd_usb_stream *stream;
 	struct usb_device *dev;
@@ -156,6 +158,7 @@ struct snd_usb_substream {
 	} dsd_dop;
 
 	bool trigger_tstamp_pending_update; /* trigger timestamp being updated from initial estimate */
+	struct media_ctl *media_ctl;
 };
 
 struct snd_usb_stream {
diff --git a/sound/usb/media.c b/sound/usb/media.c
new file mode 100644
index 0000000..6c0dfd8
--- /dev/null
+++ b/sound/usb/media.c
@@ -0,0 +1,318 @@
+/*
+ * media.c - Media Controller specific ALSA driver code
+ *
+ * Copyright (c) 2016 Shuah Khan <shuahkh@osg.samsung.com>
+ * Copyright (c) 2016 Samsung Electronics Co., Ltd.
+ *
+ * This file is released under the GPLv2.
+ */
+
+/*
+ * This file adds Media Controller support to ALSA driver
+ * to use the Media Controller API to share tuner with DVB
+ * and V4L2 drivers that control media device. Media device
+ * is created based on existing quirks framework. Using this
+ * approach, the media controller API usage can be added for
+ * a specific device.
+*/
+
+#include <linux/init.h>
+#include <linux/list.h>
+#include <linux/mutex.h>
+#include <linux/slab.h>
+#include <linux/usb.h>
+
+#include <sound/pcm.h>
+#include <sound/core.h>
+
+#include "usbaudio.h"
+#include "card.h"
+#include "mixer.h"
+#include "media.h"
+
+static int media_snd_enable_source(struct media_ctl *mctl)
+{
+	if (mctl && mctl->media_dev->enable_source)
+		return mctl->media_dev->enable_source(&mctl->media_entity,
+						      &mctl->media_pipe);
+	return 0;
+}
+
+static void media_snd_disable_source(struct media_ctl *mctl)
+{
+	if (mctl && mctl->media_dev->disable_source)
+		mctl->media_dev->disable_source(&mctl->media_entity);
+}
+
+int media_snd_stream_init(struct snd_usb_substream *subs, struct snd_pcm *pcm,
+			int stream)
+{
+	struct media_device *mdev;
+	struct media_ctl *mctl;
+	struct device *pcm_dev = &pcm->streams[stream].dev;
+	u32 intf_type;
+	int ret = 0;
+	u16 mixer_pad;
+	struct media_entity *entity;
+
+	mdev = subs->stream->chip->media_dev;
+	if (!mdev)
+		return -ENODEV;
+
+	if (subs->media_ctl)
+		return 0;
+
+	/* allocate media_ctl */
+	mctl = kzalloc(sizeof(*mctl), GFP_KERNEL);
+	if (!mctl)
+		return -ENOMEM;
+
+	mctl->media_dev = mdev;
+	if (stream == SNDRV_PCM_STREAM_PLAYBACK) {
+		intf_type = MEDIA_INTF_T_ALSA_PCM_PLAYBACK;
+		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_PLAYBACK;
+		mctl->media_pad.flags = MEDIA_PAD_FL_SOURCE;
+		mixer_pad = 1;
+	} else {
+		intf_type = MEDIA_INTF_T_ALSA_PCM_CAPTURE;
+		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_CAPTURE;
+		mctl->media_pad.flags = MEDIA_PAD_FL_SINK;
+		mixer_pad = 2;
+	}
+	mctl->media_entity.name = pcm->name;
+	media_entity_pads_init(&mctl->media_entity, 1, &mctl->media_pad);
+	ret =  media_device_register_entity(mctl->media_dev,
+					    &mctl->media_entity);
+	if (ret)
+		goto err1;
+
+	mctl->intf_devnode = media_devnode_create(mdev, intf_type, 0,
+						  MAJOR(pcm_dev->devt),
+						  MINOR(pcm_dev->devt));
+	if (!mctl->intf_devnode) {
+		ret = -ENOMEM;
+		goto err2;
+	}
+	mctl->intf_link = media_create_intf_link(&mctl->media_entity,
+						 &mctl->intf_devnode->intf,
+						 MEDIA_LNK_FL_ENABLED);
+	if (!mctl->intf_link) {
+		ret = -ENOMEM;
+		goto err3;
+	}
+
+	/* create link between mixer and audio */
+	media_device_for_each_entity(entity, mdev) {
+		switch (entity->function) {
+		case MEDIA_ENT_F_AUDIO_MIXER:
+			ret = media_create_pad_link(entity, mixer_pad,
+						    &mctl->media_entity, 0,
+						    MEDIA_LNK_FL_ENABLED);
+			if (ret)
+				goto err4;
+			break;
+		}
+	}
+
+	subs->media_ctl = mctl;
+	return 0;
+
+err4:
+	media_remove_intf_link(mctl->intf_link);
+err3:
+	media_devnode_remove(mctl->intf_devnode);
+err2:
+	media_device_unregister_entity(&mctl->media_entity);
+err1:
+	kfree(mctl);
+	return ret;
+}
+
+void media_snd_stream_delete(struct snd_usb_substream *subs)
+{
+	struct media_ctl *mctl = subs->media_ctl;
+
+	if (mctl && mctl->media_dev) {
+		struct media_device *mdev;
+
+		mdev = subs->stream->chip->media_dev;
+		if (mdev && media_devnode_is_registered(&mdev->devnode)) {
+			media_devnode_remove(mctl->intf_devnode);
+			media_device_unregister_entity(&mctl->media_entity);
+			media_entity_cleanup(&mctl->media_entity);
+		}
+		kfree(mctl);
+		subs->media_ctl = NULL;
+	}
+}
+
+int media_snd_start_pipeline(struct snd_usb_substream *subs)
+{
+	struct media_ctl *mctl = subs->media_ctl;
+
+	if (mctl)
+		return media_snd_enable_source(mctl);
+	return 0;
+}
+
+void media_snd_stop_pipeline(struct snd_usb_substream *subs)
+{
+	struct media_ctl *mctl = subs->media_ctl;
+
+	if (mctl)
+		media_snd_disable_source(mctl);
+}
+
+int media_snd_mixer_init(struct snd_usb_audio *chip)
+{
+	struct device *ctl_dev = &chip->card->ctl_dev;
+	struct media_intf_devnode *ctl_intf;
+	struct usb_mixer_interface *mixer;
+	struct media_device *mdev = chip->media_dev;
+	struct media_mixer_ctl *mctl;
+	u32 intf_type = MEDIA_INTF_T_ALSA_CONTROL;
+	int ret;
+
+	if (!mdev)
+		return -ENODEV;
+
+	ctl_intf = chip->ctl_intf_media_devnode;
+	if (!ctl_intf) {
+		ctl_intf = media_devnode_create(mdev, intf_type, 0,
+						MAJOR(ctl_dev->devt),
+						MINOR(ctl_dev->devt));
+		if (!ctl_intf)
+			return -ENOMEM;
+		chip->ctl_intf_media_devnode = ctl_intf;
+	}
+
+	list_for_each_entry(mixer, &chip->mixer_list, list) {
+
+		if (mixer->media_mixer_ctl)
+			continue;
+
+		/* allocate media_mixer_ctl */
+		mctl = kzalloc(sizeof(*mctl), GFP_KERNEL);
+		if (!mctl)
+			return -ENOMEM;
+
+		mctl->media_dev = mdev;
+		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_MIXER;
+		mctl->media_entity.name = chip->card->mixername;
+		mctl->media_pad[0].flags = MEDIA_PAD_FL_SINK;
+		mctl->media_pad[1].flags = MEDIA_PAD_FL_SOURCE;
+		mctl->media_pad[2].flags = MEDIA_PAD_FL_SOURCE;
+		media_entity_pads_init(&mctl->media_entity, MEDIA_MIXER_PAD_MAX,
+				  mctl->media_pad);
+		ret =  media_device_register_entity(mctl->media_dev,
+						    &mctl->media_entity);
+		if (ret) {
+			kfree(mctl);
+			return ret;
+		}
+
+		mctl->intf_link = media_create_intf_link(&mctl->media_entity,
+							 &ctl_intf->intf,
+							 MEDIA_LNK_FL_ENABLED);
+		if (!mctl->intf_link) {
+			media_device_unregister_entity(&mctl->media_entity);
+			media_entity_cleanup(&mctl->media_entity);
+			kfree(mctl);
+			return -ENOMEM;
+		}
+		mctl->intf_devnode = ctl_intf;
+		mixer->media_mixer_ctl = mctl;
+	}
+	return 0;
+}
+
+static void media_snd_mixer_delete(struct snd_usb_audio *chip)
+{
+	struct usb_mixer_interface *mixer;
+	struct media_device *mdev = chip->media_dev;
+
+	if (!mdev)
+		return;
+
+	list_for_each_entry(mixer, &chip->mixer_list, list) {
+		struct media_mixer_ctl *mctl;
+
+		mctl = mixer->media_mixer_ctl;
+		if (!mixer->media_mixer_ctl)
+			continue;
+
+		if (media_devnode_is_registered(&mdev->devnode)) {
+			media_device_unregister_entity(&mctl->media_entity);
+			media_entity_cleanup(&mctl->media_entity);
+		}
+		kfree(mctl);
+		mixer->media_mixer_ctl = NULL;
+	}
+	if (media_devnode_is_registered(&mdev->devnode))
+		media_devnode_remove(chip->ctl_intf_media_devnode);
+	chip->ctl_intf_media_devnode = NULL;
+}
+
+int media_snd_device_create(struct snd_usb_audio *chip,
+			struct usb_interface *iface)
+{
+	struct media_device *mdev;
+	struct usb_device *usbdev = interface_to_usbdev(iface);
+	int ret;
+
+	mdev = media_device_get_devres(&usbdev->dev);
+	if (!mdev)
+		return -ENOMEM;
+	if (!mdev->dev) {
+		/* register media device */
+		mdev->dev = &usbdev->dev;
+		if (usbdev->product)
+			strlcpy(mdev->model, usbdev->product,
+				sizeof(mdev->model));
+		if (usbdev->serial)
+			strlcpy(mdev->serial, usbdev->serial,
+				sizeof(mdev->serial));
+		strcpy(mdev->bus_info, usbdev->devpath);
+		mdev->hw_revision = le16_to_cpu(usbdev->descriptor.bcdDevice);
+		media_device_init(mdev);
+	}
+	if (!media_devnode_is_registered(&mdev->devnode)) {
+		ret = media_device_register(mdev);
+		if (ret) {
+			dev_err(&usbdev->dev,
+				"Couldn't register media device. Error: %d\n",
+				ret);
+			return ret;
+		}
+	}
+
+	/* save media device - avoid lookups */
+	chip->media_dev = mdev;
+
+	/* Create media entities for mixer and control dev */
+	ret = media_snd_mixer_init(chip);
+	if (ret) {
+		dev_err(&usbdev->dev,
+			"Couldn't create media mixer entities. Error: %d\n",
+			ret);
+
+		/* clear saved media_dev */
+		chip->media_dev = NULL;
+
+		return ret;
+	}
+	return 0;
+}
+
+void media_snd_device_delete(struct snd_usb_audio *chip)
+{
+	struct media_device *mdev = chip->media_dev;
+
+	media_snd_mixer_delete(chip);
+
+	if (mdev) {
+		if (media_devnode_is_registered(&mdev->devnode))
+			media_device_unregister(mdev);
+		chip->media_dev = NULL;
+	}
+}
diff --git a/sound/usb/media.h b/sound/usb/media.h
new file mode 100644
index 0000000..1dcdcdc
--- /dev/null
+++ b/sound/usb/media.h
@@ -0,0 +1,72 @@
+/*
+ * media.h - Media Controller specific ALSA driver code
+ *
+ * Copyright (c) 2016 Shuah Khan <shuahkh@osg.samsung.com>
+ * Copyright (c) 2016 Samsung Electronics Co., Ltd.
+ *
+ * This file is released under the GPLv2.
+ */
+
+/*
+ * This file adds Media Controller support to ALSA driver
+ * to use the Media Controller API to share tuner with DVB
+ * and V4L2 drivers that control media device. Media device
+ * is created based on existing quirks framework. Using this
+ * approach, the media controller API usage can be added for
+ * a specific device.
+*/
+#ifndef __MEDIA_H
+
+#ifdef CONFIG_SND_USB_AUDIO_USE_MEDIA_CONTROLLER
+
+#include <media/media-device.h>
+#include <media/media-entity.h>
+#include <sound/asound.h>
+
+struct media_ctl {
+	struct media_device *media_dev;
+	struct media_entity media_entity;
+	struct media_intf_devnode *intf_devnode;
+	struct media_link *intf_link;
+	struct media_pad media_pad;
+	struct media_pipeline media_pipe;
+};
+
+/*
+ * One source pad each for SNDRV_PCM_STREAM_CAPTURE and
+ * SNDRV_PCM_STREAM_PLAYBACK. One for sink pad to link
+ * to AUDIO Source
+*/
+#define MEDIA_MIXER_PAD_MAX    (SNDRV_PCM_STREAM_LAST + 2)
+
+struct media_mixer_ctl {
+	struct media_device *media_dev;
+	struct media_entity media_entity;
+	struct media_intf_devnode *intf_devnode;
+	struct media_link *intf_link;
+	struct media_pad media_pad[MEDIA_MIXER_PAD_MAX];
+	struct media_pipeline media_pipe;
+};
+
+int media_snd_device_create(struct snd_usb_audio *chip,
+			    struct usb_interface *iface);
+void media_snd_device_delete(struct snd_usb_audio *chip);
+int media_snd_stream_init(struct snd_usb_substream *subs, struct snd_pcm *pcm,
+			  int stream);
+void media_snd_stream_delete(struct snd_usb_substream *subs);
+int media_snd_start_pipeline(struct snd_usb_substream *subs);
+void media_snd_stop_pipeline(struct snd_usb_substream *subs);
+#else
+static inline int media_snd_device_create(struct snd_usb_audio *chip,
+					  struct usb_interface *iface)
+						{ return 0; }
+static inline void media_snd_device_delete(struct snd_usb_audio *chip) { }
+static inline int media_snd_stream_init(struct snd_usb_substream *subs,
+					struct snd_pcm *pcm, int stream)
+						{ return 0; }
+static inline void media_snd_stream_delete(struct snd_usb_substream *subs) { }
+static inline int media_snd_start_pipeline(struct snd_usb_substream *subs)
+					{ return 0; }
+static inline void media_snd_stop_pipeline(struct snd_usb_substream *subs) { }
+#endif
+#endif /* __MEDIA_H */
diff --git a/sound/usb/mixer.h b/sound/usb/mixer.h
index 3417ef3..f378944 100644
--- a/sound/usb/mixer.h
+++ b/sound/usb/mixer.h
@@ -3,6 +3,8 @@
 
 #include <sound/info.h>
 
+struct media_mixer_ctl;
+
 struct usb_mixer_interface {
 	struct snd_usb_audio *chip;
 	struct usb_host_interface *hostif;
@@ -22,6 +24,7 @@ struct usb_mixer_interface {
 	struct urb *rc_urb;
 	struct usb_ctrlrequest *rc_setup_packet;
 	u8 rc_buffer[6];
+	struct media_mixer_ctl *media_mixer_ctl;
 };
 
 #define MAX_CHANNELS	16	/* max logical channels */
diff --git a/sound/usb/pcm.c b/sound/usb/pcm.c
index 9245f52..b0370d5 100644
--- a/sound/usb/pcm.c
+++ b/sound/usb/pcm.c
@@ -35,6 +35,7 @@
 #include "pcm.h"
 #include "clock.h"
 #include "power.h"
+#include "media.h"
 
 #define SUBSTREAM_FLAG_DATA_EP_STARTED	0
 #define SUBSTREAM_FLAG_SYNC_EP_STARTED	1
@@ -715,10 +716,14 @@ static int snd_usb_hw_params(struct snd_pcm_substream *substream,
 	struct audioformat *fmt;
 	int ret;
 
+	ret = media_snd_start_pipeline(subs);
+	if (ret)
+		return ret;
+
 	ret = snd_pcm_lib_alloc_vmalloc_buffer(substream,
 					       params_buffer_bytes(hw_params));
 	if (ret < 0)
-		return ret;
+		goto err_ret;
 
 	subs->pcm_format = params_format(hw_params);
 	subs->period_bytes = params_period_bytes(hw_params);
@@ -732,22 +737,27 @@ static int snd_usb_hw_params(struct snd_pcm_substream *substream,
 		dev_dbg(&subs->dev->dev,
 			"cannot set format: format = %#x, rate = %d, channels = %d\n",
 			   subs->pcm_format, subs->cur_rate, subs->channels);
-		return -EINVAL;
+		ret = -EINVAL;
+		goto err_ret;
 	}
 
 	ret = snd_usb_lock_shutdown(subs->stream->chip);
 	if (ret < 0)
-		return ret;
+		goto err_ret;
 	ret = set_format(subs, fmt);
 	snd_usb_unlock_shutdown(subs->stream->chip);
 	if (ret < 0)
-		return ret;
+		goto err_ret;
 
 	subs->interface = fmt->iface;
 	subs->altset_idx = fmt->altset_idx;
 	subs->need_setup_ep = true;
 
 	return 0;
+
+err_ret:
+	media_snd_stop_pipeline(subs);
+	return ret;
 }
 
 /*
@@ -759,6 +769,7 @@ static int snd_usb_hw_free(struct snd_pcm_substream *substream)
 {
 	struct snd_usb_substream *subs = substream->runtime->private_data;
 
+	media_snd_stop_pipeline(subs);
 	subs->cur_audiofmt = NULL;
 	subs->cur_rate = 0;
 	subs->period_bytes = 0;
@@ -1219,6 +1230,7 @@ static int snd_usb_pcm_open(struct snd_pcm_substream *substream, int direction)
 	struct snd_usb_stream *as = snd_pcm_substream_chip(substream);
 	struct snd_pcm_runtime *runtime = substream->runtime;
 	struct snd_usb_substream *subs = &as->substream[direction];
+	int ret;
 
 	subs->interface = -1;
 	subs->altset_idx = 0;
@@ -1232,7 +1244,12 @@ static int snd_usb_pcm_open(struct snd_pcm_substream *substream, int direction)
 	subs->dsd_dop.channel = 0;
 	subs->dsd_dop.marker = 1;
 
-	return setup_hw_info(runtime, subs);
+	ret = setup_hw_info(runtime, subs);
+	if (ret == 0)
+		ret = media_snd_stream_init(subs, as->pcm, direction);
+	if (ret)
+		snd_usb_autosuspend(subs->stream->chip);
+	return ret;
 }
 
 static int snd_usb_pcm_close(struct snd_pcm_substream *substream, int direction)
@@ -1241,6 +1258,7 @@ static int snd_usb_pcm_close(struct snd_pcm_substream *substream, int direction)
 	struct snd_usb_substream *subs = &as->substream[direction];
 
 	stop_endpoints(subs, true);
+	media_snd_stop_pipeline(subs);
 
 	if (subs->interface >= 0 &&
 	    !snd_usb_lock_shutdown(subs->stream->chip)) {
diff --git a/sound/usb/quirks-table.h b/sound/usb/quirks-table.h
index c60a776..9d087b1 100644
--- a/sound/usb/quirks-table.h
+++ b/sound/usb/quirks-table.h
@@ -2886,6 +2886,7 @@ YAMAHA_DEVICE(0x7010, "UB99"),
 		.product_name = pname, \
 		.ifnum = QUIRK_ANY_INTERFACE, \
 		.type = QUIRK_AUDIO_ALIGN_TRANSFER, \
+		.media_device = 1, \
 	} \
 }
 
diff --git a/sound/usb/stream.c b/sound/usb/stream.c
index c4dc577..51258a1 100644
--- a/sound/usb/stream.c
+++ b/sound/usb/stream.c
@@ -36,6 +36,7 @@
 #include "format.h"
 #include "clock.h"
 #include "stream.h"
+#include "media.h"
 
 /*
  * free a substream
@@ -52,6 +53,7 @@ static void free_substream(struct snd_usb_substream *subs)
 		kfree(fp);
 	}
 	kfree(subs->rate_list.list);
+	media_snd_stream_delete(subs);
 }
 
 
diff --git a/sound/usb/usbaudio.h b/sound/usb/usbaudio.h
index b665d85..a161c7c 100644
--- a/sound/usb/usbaudio.h
+++ b/sound/usb/usbaudio.h
@@ -30,6 +30,9 @@
  *
  */
 
+struct media_device;
+struct media_intf_devnode;
+
 struct snd_usb_audio {
 	int index;
 	struct usb_device *dev;
@@ -60,6 +63,8 @@ struct snd_usb_audio {
 	bool autoclock;			/* from the 'autoclock' module param */
 
 	struct usb_host_interface *ctrl_intf;	/* the audio control interface */
+	struct media_device *media_dev;
+	struct media_intf_devnode *ctl_intf_media_devnode;
 };
 
 #define usb_audio_err(chip, fmt, args...) \
@@ -110,6 +115,7 @@ struct snd_usb_audio_quirk {
 	const char *product_name;
 	int16_t ifnum;
 	uint16_t type;
+	bool media_device;
 	const void *data;
 };
 
-- 
2.5.0

[PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Shuah Khan]

Change ALSA driver to use Media Controller API to
share media resources with DVB and V4L2 drivers
on a AU0828 media device. Media Controller specific
initialization is done after sound card is registered.
ALSA creates Media interface and entity function graph
nodes for Control, Mixer, PCM Playback, and PCM Capture
devices.

snd_usb_hw_params() will call Media Controller enable
source handler interface to request the media resource.
If resource request is granted, it will release it from
snd_usb_hw_free(). If resource is busy, -EBUSY is returned.

Media specific cleanup is done in usb_audio_disconnect().

Signed-off-by: Shuah Khan <shuahkh@osg.samsung.com>
Acked-by: Takashi Iwai <tiwai@suse.de>
---

Changes since v4:
1. Renamed media_* interfaces in sound/usb to media_snd*
   to avoid potential conflicts with the MC interfaces.
   This addresses Laurent Pinchart's comment.
2. Added Acked-by from Takashi Iwai

 sound/usb/Kconfig        |   4 +
 sound/usb/Makefile       |   2 +
 sound/usb/card.c         |  14 +++
 sound/usb/card.h         |   3 +
 sound/usb/media.c        | 318 +++++++++++++++++++++++++++++++++++++++++++++++
 sound/usb/media.h        |  72 +++++++++++
 sound/usb/mixer.h        |   3 +
 sound/usb/pcm.c          |  28 ++++-
 sound/usb/quirks-table.h |   1 +
 sound/usb/stream.c       |   2 +
 sound/usb/usbaudio.h     |   6 +
 11 files changed, 448 insertions(+), 5 deletions(-)
 create mode 100644 sound/usb/media.c
 create mode 100644 sound/usb/media.h

diff --git a/sound/usb/Kconfig b/sound/usb/Kconfig
index a452ad7..d14bf41 100644
--- a/sound/usb/Kconfig
+++ b/sound/usb/Kconfig
@@ -15,6 +15,7 @@ config SND_USB_AUDIO
 	select SND_RAWMIDI
 	select SND_PCM
 	select BITREVERSE
+	select SND_USB_AUDIO_USE_MEDIA_CONTROLLER if MEDIA_CONTROLLER && (MEDIA_SUPPORT=y || MEDIA_SUPPORT=SND_USB_AUDIO)
 	help
 	  Say Y here to include support for USB audio and USB MIDI
 	  devices.
@@ -22,6 +23,9 @@ config SND_USB_AUDIO
 	  To compile this driver as a module, choose M here: the module
 	  will be called snd-usb-audio.
 
+config SND_USB_AUDIO_USE_MEDIA_CONTROLLER
+	bool
+
 config SND_USB_UA101
 	tristate "Edirol UA-101/UA-1000 driver"
 	select SND_PCM
diff --git a/sound/usb/Makefile b/sound/usb/Makefile
index 2d2d122..8dca3c4 100644
--- a/sound/usb/Makefile
+++ b/sound/usb/Makefile
@@ -15,6 +15,8 @@ snd-usb-audio-objs := 	card.o \
 			quirks.o \
 			stream.o
 
+snd-usb-audio-$(CONFIG_SND_USB_AUDIO_USE_MEDIA_CONTROLLER) += media.o
+
 snd-usbmidi-lib-objs := midi.o
 
 # Toplevel Module Dependency
diff --git a/sound/usb/card.c b/sound/usb/card.c
index 1f09d95..258cf70 100644
--- a/sound/usb/card.c
+++ b/sound/usb/card.c
@@ -66,6 +66,7 @@
 #include "format.h"
 #include "power.h"
 #include "stream.h"
+#include "media.h"
 
 MODULE_AUTHOR("Takashi Iwai <tiwai@suse.de>");
 MODULE_DESCRIPTION("USB Audio");
@@ -561,6 +562,11 @@ static int usb_audio_probe(struct usb_interface *intf,
 	if (err < 0)
 		goto __error;
 
+	if (quirk->media_device) {
+		/* don't want to fail when media_snd_device_create() fails */
+		media_snd_device_create(chip, intf);
+	}
+
 	usb_chip[chip->index] = chip;
 	chip->num_interfaces++;
 	usb_set_intfdata(intf, chip);
@@ -617,6 +623,14 @@ static void usb_audio_disconnect(struct usb_interface *intf)
 		list_for_each(p, &chip->midi_list) {
 			snd_usbmidi_disconnect(p);
 		}
+		/*
+		 * Nice to check quirk && quirk->media_device
+		 * need some special handlings. Doesn't look like
+		 * we have access to quirk here
+		 * Acceses mixer_list
+		*/
+		media_snd_device_delete(chip);
+
 		/* release mixer resources */
 		list_for_each_entry(mixer, &chip->mixer_list, list) {
 			snd_usb_mixer_disconnect(mixer);
diff --git a/sound/usb/card.h b/sound/usb/card.h
index 71778ca..34a0898 100644
--- a/sound/usb/card.h
+++ b/sound/usb/card.h
@@ -105,6 +105,8 @@ struct snd_usb_endpoint {
 	struct list_head list;
 };
 
+struct media_ctl;
+
 struct snd_usb_substream {
 	struct snd_usb_stream *stream;
 	struct usb_device *dev;
@@ -156,6 +158,7 @@ struct snd_usb_substream {
 	} dsd_dop;
 
 	bool trigger_tstamp_pending_update; /* trigger timestamp being updated from initial estimate */
+	struct media_ctl *media_ctl;
 };
 
 struct snd_usb_stream {
diff --git a/sound/usb/media.c b/sound/usb/media.c
new file mode 100644
index 0000000..6c0dfd8
--- /dev/null
+++ b/sound/usb/media.c
@@ -0,0 +1,318 @@
+/*
+ * media.c - Media Controller specific ALSA driver code
+ *
+ * Copyright (c) 2016 Shuah Khan <shuahkh@osg.samsung.com>
+ * Copyright (c) 2016 Samsung Electronics Co., Ltd.
+ *
+ * This file is released under the GPLv2.
+ */
+
+/*
+ * This file adds Media Controller support to ALSA driver
+ * to use the Media Controller API to share tuner with DVB
+ * and V4L2 drivers that control media device. Media device
+ * is created based on existing quirks framework. Using this
+ * approach, the media controller API usage can be added for
+ * a specific device.
+*/
+
+#include <linux/init.h>
+#include <linux/list.h>
+#include <linux/mutex.h>
+#include <linux/slab.h>
+#include <linux/usb.h>
+
+#include <sound/pcm.h>
+#include <sound/core.h>
+
+#include "usbaudio.h"
+#include "card.h"
+#include "mixer.h"
+#include "media.h"
+
+static int media_snd_enable_source(struct media_ctl *mctl)
+{
+	if (mctl && mctl->media_dev->enable_source)
+		return mctl->media_dev->enable_source(&mctl->media_entity,
+						      &mctl->media_pipe);
+	return 0;
+}
+
+static void media_snd_disable_source(struct media_ctl *mctl)
+{
+	if (mctl && mctl->media_dev->disable_source)
+		mctl->media_dev->disable_source(&mctl->media_entity);
+}
+
+int media_snd_stream_init(struct snd_usb_substream *subs, struct snd_pcm *pcm,
+			int stream)
+{
+	struct media_device *mdev;
+	struct media_ctl *mctl;
+	struct device *pcm_dev = &pcm->streams[stream].dev;
+	u32 intf_type;
+	int ret = 0;
+	u16 mixer_pad;
+	struct media_entity *entity;
+
+	mdev = subs->stream->chip->media_dev;
+	if (!mdev)
+		return -ENODEV;
+
+	if (subs->media_ctl)
+		return 0;
+
+	/* allocate media_ctl */
+	mctl = kzalloc(sizeof(*mctl), GFP_KERNEL);
+	if (!mctl)
+		return -ENOMEM;
+
+	mctl->media_dev = mdev;
+	if (stream == SNDRV_PCM_STREAM_PLAYBACK) {
+		intf_type = MEDIA_INTF_T_ALSA_PCM_PLAYBACK;
+		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_PLAYBACK;
+		mctl->media_pad.flags = MEDIA_PAD_FL_SOURCE;
+		mixer_pad = 1;
+	} else {
+		intf_type = MEDIA_INTF_T_ALSA_PCM_CAPTURE;
+		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_CAPTURE;
+		mctl->media_pad.flags = MEDIA_PAD_FL_SINK;
+		mixer_pad = 2;
+	}
+	mctl->media_entity.name = pcm->name;
+	media_entity_pads_init(&mctl->media_entity, 1, &mctl->media_pad);
+	ret =  media_device_register_entity(mctl->media_dev,
+					    &mctl->media_entity);
+	if (ret)
+		goto err1;
+
+	mctl->intf_devnode = media_devnode_create(mdev, intf_type, 0,
+						  MAJOR(pcm_dev->devt),
+						  MINOR(pcm_dev->devt));
+	if (!mctl->intf_devnode) {
+		ret = -ENOMEM;
+		goto err2;
+	}
+	mctl->intf_link = media_create_intf_link(&mctl->media_entity,
+						 &mctl->intf_devnode->intf,
+						 MEDIA_LNK_FL_ENABLED);
+	if (!mctl->intf_link) {
+		ret = -ENOMEM;
+		goto err3;
+	}
+
+	/* create link between mixer and audio */
+	media_device_for_each_entity(entity, mdev) {
+		switch (entity->function) {
+		case MEDIA_ENT_F_AUDIO_MIXER:
+			ret = media_create_pad_link(entity, mixer_pad,
+						    &mctl->media_entity, 0,
+						    MEDIA_LNK_FL_ENABLED);
+			if (ret)
+				goto err4;
+			break;
+		}
+	}
+
+	subs->media_ctl = mctl;
+	return 0;
+
+err4:
+	media_remove_intf_link(mctl->intf_link);
+err3:
+	media_devnode_remove(mctl->intf_devnode);
+err2:
+	media_device_unregister_entity(&mctl->media_entity);
+err1:
+	kfree(mctl);
+	return ret;
+}
+
+void media_snd_stream_delete(struct snd_usb_substream *subs)
+{
+	struct media_ctl *mctl = subs->media_ctl;
+
+	if (mctl && mctl->media_dev) {
+		struct media_device *mdev;
+
+		mdev = subs->stream->chip->media_dev;
+		if (mdev && media_devnode_is_registered(&mdev->devnode)) {
+			media_devnode_remove(mctl->intf_devnode);
+			media_device_unregister_entity(&mctl->media_entity);
+			media_entity_cleanup(&mctl->media_entity);
+		}
+		kfree(mctl);
+		subs->media_ctl = NULL;
+	}
+}
+
+int media_snd_start_pipeline(struct snd_usb_substream *subs)
+{
+	struct media_ctl *mctl = subs->media_ctl;
+
+	if (mctl)
+		return media_snd_enable_source(mctl);
+	return 0;
+}
+
+void media_snd_stop_pipeline(struct snd_usb_substream *subs)
+{
+	struct media_ctl *mctl = subs->media_ctl;
+
+	if (mctl)
+		media_snd_disable_source(mctl);
+}
+
+int media_snd_mixer_init(struct snd_usb_audio *chip)
+{
+	struct device *ctl_dev = &chip->card->ctl_dev;
+	struct media_intf_devnode *ctl_intf;
+	struct usb_mixer_interface *mixer;
+	struct media_device *mdev = chip->media_dev;
+	struct media_mixer_ctl *mctl;
+	u32 intf_type = MEDIA_INTF_T_ALSA_CONTROL;
+	int ret;
+
+	if (!mdev)
+		return -ENODEV;
+
+	ctl_intf = chip->ctl_intf_media_devnode;
+	if (!ctl_intf) {
+		ctl_intf = media_devnode_create(mdev, intf_type, 0,
+						MAJOR(ctl_dev->devt),
+						MINOR(ctl_dev->devt));
+		if (!ctl_intf)
+			return -ENOMEM;
+		chip->ctl_intf_media_devnode = ctl_intf;
+	}
+
+	list_for_each_entry(mixer, &chip->mixer_list, list) {
+
+		if (mixer->media_mixer_ctl)
+			continue;
+
+		/* allocate media_mixer_ctl */
+		mctl = kzalloc(sizeof(*mctl), GFP_KERNEL);
+		if (!mctl)
+			return -ENOMEM;
+
+		mctl->media_dev = mdev;
+		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_MIXER;
+		mctl->media_entity.name = chip->card->mixername;
+		mctl->media_pad[0].flags = MEDIA_PAD_FL_SINK;
+		mctl->media_pad[1].flags = MEDIA_PAD_FL_SOURCE;
+		mctl->media_pad[2].flags = MEDIA_PAD_FL_SOURCE;
+		media_entity_pads_init(&mctl->media_entity, MEDIA_MIXER_PAD_MAX,
+				  mctl->media_pad);
+		ret =  media_device_register_entity(mctl->media_dev,
+						    &mctl->media_entity);
+		if (ret) {
+			kfree(mctl);
+			return ret;
+		}
+
+		mctl->intf_link = media_create_intf_link(&mctl->media_entity,
+							 &ctl_intf->intf,
+							 MEDIA_LNK_FL_ENABLED);
+		if (!mctl->intf_link) {
+			media_device_unregister_entity(&mctl->media_entity);
+			media_entity_cleanup(&mctl->media_entity);
+			kfree(mctl);
+			return -ENOMEM;
+		}
+		mctl->intf_devnode = ctl_intf;
+		mixer->media_mixer_ctl = mctl;
+	}
+	return 0;
+}
+
+static void media_snd_mixer_delete(struct snd_usb_audio *chip)
+{
+	struct usb_mixer_interface *mixer;
+	struct media_device *mdev = chip->media_dev;
+
+	if (!mdev)
+		return;
+
+	list_for_each_entry(mixer, &chip->mixer_list, list) {
+		struct media_mixer_ctl *mctl;
+
+		mctl = mixer->media_mixer_ctl;
+		if (!mixer->media_mixer_ctl)
+			continue;
+
+		if (media_devnode_is_registered(&mdev->devnode)) {
+			media_device_unregister_entity(&mctl->media_entity);
+			media_entity_cleanup(&mctl->media_entity);
+		}
+		kfree(mctl);
+		mixer->media_mixer_ctl = NULL;
+	}
+	if (media_devnode_is_registered(&mdev->devnode))
+		media_devnode_remove(chip->ctl_intf_media_devnode);
+	chip->ctl_intf_media_devnode = NULL;
+}
+
+int media_snd_device_create(struct snd_usb_audio *chip,
+			struct usb_interface *iface)
+{
+	struct media_device *mdev;
+	struct usb_device *usbdev = interface_to_usbdev(iface);
+	int ret;
+
+	mdev = media_device_get_devres(&usbdev->dev);
+	if (!mdev)
+		return -ENOMEM;
+	if (!mdev->dev) {
+		/* register media device */
+		mdev->dev = &usbdev->dev;
+		if (usbdev->product)
+			strlcpy(mdev->model, usbdev->product,
+				sizeof(mdev->model));
+		if (usbdev->serial)
+			strlcpy(mdev->serial, usbdev->serial,
+				sizeof(mdev->serial));
+		strcpy(mdev->bus_info, usbdev->devpath);
+		mdev->hw_revision = le16_to_cpu(usbdev->descriptor.bcdDevice);
+		media_device_init(mdev);
+	}
+	if (!media_devnode_is_registered(&mdev->devnode)) {
+		ret = media_device_register(mdev);
+		if (ret) {
+			dev_err(&usbdev->dev,
+				"Couldn't register media device. Error: %d\n",
+				ret);
+			return ret;
+		}
+	}
+
+	/* save media device - avoid lookups */
+	chip->media_dev = mdev;
+
+	/* Create media entities for mixer and control dev */
+	ret = media_snd_mixer_init(chip);
+	if (ret) {
+		dev_err(&usbdev->dev,
+			"Couldn't create media mixer entities. Error: %d\n",
+			ret);
+
+		/* clear saved media_dev */
+		chip->media_dev = NULL;
+
+		return ret;
+	}
+	return 0;
+}
+
+void media_snd_device_delete(struct snd_usb_audio *chip)
+{
+	struct media_device *mdev = chip->media_dev;
+
+	media_snd_mixer_delete(chip);
+
+	if (mdev) {
+		if (media_devnode_is_registered(&mdev->devnode))
+			media_device_unregister(mdev);
+		chip->media_dev = NULL;
+	}
+}
diff --git a/sound/usb/media.h b/sound/usb/media.h
new file mode 100644
index 0000000..1dcdcdc
--- /dev/null
+++ b/sound/usb/media.h
@@ -0,0 +1,72 @@
+/*
+ * media.h - Media Controller specific ALSA driver code
+ *
+ * Copyright (c) 2016 Shuah Khan <shuahkh@osg.samsung.com>
+ * Copyright (c) 2016 Samsung Electronics Co., Ltd.
+ *
+ * This file is released under the GPLv2.
+ */
+
+/*
+ * This file adds Media Controller support to ALSA driver
+ * to use the Media Controller API to share tuner with DVB
+ * and V4L2 drivers that control media device. Media device
+ * is created based on existing quirks framework. Using this
+ * approach, the media controller API usage can be added for
+ * a specific device.
+*/
+#ifndef __MEDIA_H
+
+#ifdef CONFIG_SND_USB_AUDIO_USE_MEDIA_CONTROLLER
+
+#include <media/media-device.h>
+#include <media/media-entity.h>
+#include <sound/asound.h>
+
+struct media_ctl {
+	struct media_device *media_dev;
+	struct media_entity media_entity;
+	struct media_intf_devnode *intf_devnode;
+	struct media_link *intf_link;
+	struct media_pad media_pad;
+	struct media_pipeline media_pipe;
+};
+
+/*
+ * One source pad each for SNDRV_PCM_STREAM_CAPTURE and
+ * SNDRV_PCM_STREAM_PLAYBACK. One for sink pad to link
+ * to AUDIO Source
+*/
+#define MEDIA_MIXER_PAD_MAX    (SNDRV_PCM_STREAM_LAST + 2)
+
+struct media_mixer_ctl {
+	struct media_device *media_dev;
+	struct media_entity media_entity;
+	struct media_intf_devnode *intf_devnode;
+	struct media_link *intf_link;
+	struct media_pad media_pad[MEDIA_MIXER_PAD_MAX];
+	struct media_pipeline media_pipe;
+};
+
+int media_snd_device_create(struct snd_usb_audio *chip,
+			    struct usb_interface *iface);
+void media_snd_device_delete(struct snd_usb_audio *chip);
+int media_snd_stream_init(struct snd_usb_substream *subs, struct snd_pcm *pcm,
+			  int stream);
+void media_snd_stream_delete(struct snd_usb_substream *subs);
+int media_snd_start_pipeline(struct snd_usb_substream *subs);
+void media_snd_stop_pipeline(struct snd_usb_substream *subs);
+#else
+static inline int media_snd_device_create(struct snd_usb_audio *chip,
+					  struct usb_interface *iface)
+						{ return 0; }
+static inline void media_snd_device_delete(struct snd_usb_audio *chip) { }
+static inline int media_snd_stream_init(struct snd_usb_substream *subs,
+					struct snd_pcm *pcm, int stream)
+						{ return 0; }
+static inline void media_snd_stream_delete(struct snd_usb_substream *subs) { }
+static inline int media_snd_start_pipeline(struct snd_usb_substream *subs)
+					{ return 0; }
+static inline void media_snd_stop_pipeline(struct snd_usb_substream *subs) { }
+#endif
+#endif /* __MEDIA_H */
diff --git a/sound/usb/mixer.h b/sound/usb/mixer.h
index 3417ef3..f378944 100644
--- a/sound/usb/mixer.h
+++ b/sound/usb/mixer.h
@@ -3,6 +3,8 @@
 
 #include <sound/info.h>
 
+struct media_mixer_ctl;
+
 struct usb_mixer_interface {
 	struct snd_usb_audio *chip;
 	struct usb_host_interface *hostif;
@@ -22,6 +24,7 @@ struct usb_mixer_interface {
 	struct urb *rc_urb;
 	struct usb_ctrlrequest *rc_setup_packet;
 	u8 rc_buffer[6];
+	struct media_mixer_ctl *media_mixer_ctl;
 };
 
 #define MAX_CHANNELS	16	/* max logical channels */
diff --git a/sound/usb/pcm.c b/sound/usb/pcm.c
index 9245f52..b0370d5 100644
--- a/sound/usb/pcm.c
+++ b/sound/usb/pcm.c
@@ -35,6 +35,7 @@
 #include "pcm.h"
 #include "clock.h"
 #include "power.h"
+#include "media.h"
 
 #define SUBSTREAM_FLAG_DATA_EP_STARTED	0
 #define SUBSTREAM_FLAG_SYNC_EP_STARTED	1
@@ -715,10 +716,14 @@ static int snd_usb_hw_params(struct snd_pcm_substream *substream,
 	struct audioformat *fmt;
 	int ret;
 
+	ret = media_snd_start_pipeline(subs);
+	if (ret)
+		return ret;
+
 	ret = snd_pcm_lib_alloc_vmalloc_buffer(substream,
 					       params_buffer_bytes(hw_params));
 	if (ret < 0)
-		return ret;
+		goto err_ret;
 
 	subs->pcm_format = params_format(hw_params);
 	subs->period_bytes = params_period_bytes(hw_params);
@@ -732,22 +737,27 @@ static int snd_usb_hw_params(struct snd_pcm_substream *substream,
 		dev_dbg(&subs->dev->dev,
 			"cannot set format: format = %#x, rate = %d, channels = %d\n",
 			   subs->pcm_format, subs->cur_rate, subs->channels);
-		return -EINVAL;
+		ret = -EINVAL;
+		goto err_ret;
 	}
 
 	ret = snd_usb_lock_shutdown(subs->stream->chip);
 	if (ret < 0)
-		return ret;
+		goto err_ret;
 	ret = set_format(subs, fmt);
 	snd_usb_unlock_shutdown(subs->stream->chip);
 	if (ret < 0)
-		return ret;
+		goto err_ret;
 
 	subs->interface = fmt->iface;
 	subs->altset_idx = fmt->altset_idx;
 	subs->need_setup_ep = true;
 
 	return 0;
+
+err_ret:
+	media_snd_stop_pipeline(subs);
+	return ret;
 }
 
 /*
@@ -759,6 +769,7 @@ static int snd_usb_hw_free(struct snd_pcm_substream *substream)
 {
 	struct snd_usb_substream *subs = substream->runtime->private_data;
 
+	media_snd_stop_pipeline(subs);
 	subs->cur_audiofmt = NULL;
 	subs->cur_rate = 0;
 	subs->period_bytes = 0;
@@ -1219,6 +1230,7 @@ static int snd_usb_pcm_open(struct snd_pcm_substream *substream, int direction)
 	struct snd_usb_stream *as = snd_pcm_substream_chip(substream);
 	struct snd_pcm_runtime *runtime = substream->runtime;
 	struct snd_usb_substream *subs = &as->substream[direction];
+	int ret;
 
 	subs->interface = -1;
 	subs->altset_idx = 0;
@@ -1232,7 +1244,12 @@ static int snd_usb_pcm_open(struct snd_pcm_substream *substream, int direction)
 	subs->dsd_dop.channel = 0;
 	subs->dsd_dop.marker = 1;
 
-	return setup_hw_info(runtime, subs);
+	ret = setup_hw_info(runtime, subs);
+	if (ret == 0)
+		ret = media_snd_stream_init(subs, as->pcm, direction);
+	if (ret)
+		snd_usb_autosuspend(subs->stream->chip);
+	return ret;
 }
 
 static int snd_usb_pcm_close(struct snd_pcm_substream *substream, int direction)
@@ -1241,6 +1258,7 @@ static int snd_usb_pcm_close(struct snd_pcm_substream *substream, int direction)
 	struct snd_usb_substream *subs = &as->substream[direction];
 
 	stop_endpoints(subs, true);
+	media_snd_stop_pipeline(subs);
 
 	if (subs->interface >= 0 &&
 	    !snd_usb_lock_shutdown(subs->stream->chip)) {
diff --git a/sound/usb/quirks-table.h b/sound/usb/quirks-table.h
index c60a776..9d087b1 100644
--- a/sound/usb/quirks-table.h
+++ b/sound/usb/quirks-table.h
@@ -2886,6 +2886,7 @@ YAMAHA_DEVICE(0x7010, "UB99"),
 		.product_name = pname, \
 		.ifnum = QUIRK_ANY_INTERFACE, \
 		.type = QUIRK_AUDIO_ALIGN_TRANSFER, \
+		.media_device = 1, \
 	} \
 }
 
diff --git a/sound/usb/stream.c b/sound/usb/stream.c
index c4dc577..51258a1 100644
--- a/sound/usb/stream.c
+++ b/sound/usb/stream.c
@@ -36,6 +36,7 @@
 #include "format.h"
 #include "clock.h"
 #include "stream.h"
+#include "media.h"
 
 /*
  * free a substream
@@ -52,6 +53,7 @@ static void free_substream(struct snd_usb_substream *subs)
 		kfree(fp);
 	}
 	kfree(subs->rate_list.list);
+	media_snd_stream_delete(subs);
 }
 
 
diff --git a/sound/usb/usbaudio.h b/sound/usb/usbaudio.h
index b665d85..a161c7c 100644
--- a/sound/usb/usbaudio.h
+++ b/sound/usb/usbaudio.h
@@ -30,6 +30,9 @@
  *
  */
 
+struct media_device;
+struct media_intf_devnode;
+
 struct snd_usb_audio {
 	int index;
 	struct usb_device *dev;
@@ -60,6 +63,8 @@ struct snd_usb_audio {
 	bool autoclock;			/* from the 'autoclock' module param */
 
 	struct usb_host_interface *ctrl_intf;	/* the audio control interface */
+	struct media_device *media_dev;
+	struct media_intf_devnode *ctl_intf_media_devnode;
 };
 
 #define usb_audio_err(chip, fmt, args...) \
@@ -110,6 +115,7 @@ struct snd_usb_audio_quirk {
 	const char *product_name;
 	int16_t ifnum;
 	uint16_t type;
+	bool media_device;
 	const void *data;
 };
 
-- 
2.5.0

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Dan Carpenter]

On Wed, Mar 02, 2016 at 09:50:31AM -0700, Shuah Khan wrote:
> +	mctl = kzalloc(sizeof(*mctl), GFP_KERNEL);
> +	if (!mctl)
> +		return -ENOMEM;
> +
> +	mctl->media_dev = mdev;
> +	if (stream == SNDRV_PCM_STREAM_PLAYBACK) {
> +		intf_type = MEDIA_INTF_T_ALSA_PCM_PLAYBACK;
> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_PLAYBACK;
> +		mctl->media_pad.flags = MEDIA_PAD_FL_SOURCE;
> +		mixer_pad = 1;
> +	} else {
> +		intf_type = MEDIA_INTF_T_ALSA_PCM_CAPTURE;
> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_CAPTURE;
> +		mctl->media_pad.flags = MEDIA_PAD_FL_SINK;
> +		mixer_pad = 2;
> +	}
> +	mctl->media_entity.name = pcm->name;
> +	media_entity_pads_init(&mctl->media_entity, 1, &mctl->media_pad);
> +	ret =  media_device_register_entity(mctl->media_dev,
> +					    &mctl->media_entity);
> +	if (ret)
> +		goto err1;

Could we give this label a meaningful name instead of a number?
goto free_mctl;

> +
> +	mctl->intf_devnode = media_devnode_create(mdev, intf_type, 0,
> +						  MAJOR(pcm_dev->devt),
> +						  MINOR(pcm_dev->devt));
> +	if (!mctl->intf_devnode) {
> +		ret = -ENOMEM;
> +		goto err2;

goto unregister_device;

> +	}
> +	mctl->intf_link = media_create_intf_link(&mctl->media_entity,
> +						 &mctl->intf_devnode->intf,
> +						 MEDIA_LNK_FL_ENABLED);
> +	if (!mctl->intf_link) {
> +		ret = -ENOMEM;
> +		goto err3;

goto delete_devnode;

> +	}
> +
> +	/* create link between mixer and audio */
> +	media_device_for_each_entity(entity, mdev) {
> +		switch (entity->function) {
> +		case MEDIA_ENT_F_AUDIO_MIXER:
> +			ret = media_create_pad_link(entity, mixer_pad,
> +						    &mctl->media_entity, 0,
> +						    MEDIA_LNK_FL_ENABLED);
> +			if (ret)
> +				goto err4;

This is a bit weird because we're inside a loop.  Shouldn't we call
media_entity_remove_links() or something if this is the second time
through the loop?

I don't understand this.  The kernel has the media_entity_cleanup() stub
function which is supposed to do this but it hasn't been implemented
yet?

regards,
dan carpenter

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Dan Carpenter]

On Wed, Mar 02, 2016 at 09:50:31AM -0700, Shuah Khan wrote:
> +	mctl = kzalloc(sizeof(*mctl), GFP_KERNEL);
> +	if (!mctl)
> +		return -ENOMEM;
> +
> +	mctl->media_dev = mdev;
> +	if (stream == SNDRV_PCM_STREAM_PLAYBACK) {
> +		intf_type = MEDIA_INTF_T_ALSA_PCM_PLAYBACK;
> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_PLAYBACK;
> +		mctl->media_pad.flags = MEDIA_PAD_FL_SOURCE;
> +		mixer_pad = 1;
> +	} else {
> +		intf_type = MEDIA_INTF_T_ALSA_PCM_CAPTURE;
> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_CAPTURE;
> +		mctl->media_pad.flags = MEDIA_PAD_FL_SINK;
> +		mixer_pad = 2;
> +	}
> +	mctl->media_entity.name = pcm->name;
> +	media_entity_pads_init(&mctl->media_entity, 1, &mctl->media_pad);
> +	ret =  media_device_register_entity(mctl->media_dev,
> +					    &mctl->media_entity);
> +	if (ret)
> +		goto err1;

Could we give this label a meaningful name instead of a number?
goto free_mctl;

> +
> +	mctl->intf_devnode = media_devnode_create(mdev, intf_type, 0,
> +						  MAJOR(pcm_dev->devt),
> +						  MINOR(pcm_dev->devt));
> +	if (!mctl->intf_devnode) {
> +		ret = -ENOMEM;
> +		goto err2;

goto unregister_device;

> +	}
> +	mctl->intf_link = media_create_intf_link(&mctl->media_entity,
> +						 &mctl->intf_devnode->intf,
> +						 MEDIA_LNK_FL_ENABLED);
> +	if (!mctl->intf_link) {
> +		ret = -ENOMEM;
> +		goto err3;

goto delete_devnode;

> +	}
> +
> +	/* create link between mixer and audio */
> +	media_device_for_each_entity(entity, mdev) {
> +		switch (entity->function) {
> +		case MEDIA_ENT_F_AUDIO_MIXER:
> +			ret = media_create_pad_link(entity, mixer_pad,
> +						    &mctl->media_entity, 0,
> +						    MEDIA_LNK_FL_ENABLED);
> +			if (ret)
> +				goto err4;

This is a bit weird because we're inside a loop.  Shouldn't we call
media_entity_remove_links() or something if this is the second time
through the loop?

I don't understand this.  The kernel has the media_entity_cleanup() stub
function which is supposed to do this but it hasn't been implemented
yet?

regards,
dan carpenter

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Shuah Khan]

On 03/02/2016 01:41 PM, Dan Carpenter wrote:
> On Wed, Mar 02, 2016 at 09:50:31AM -0700, Shuah Khan wrote:
>> +	mctl = kzalloc(sizeof(*mctl), GFP_KERNEL);
>> +	if (!mctl)
>> +		return -ENOMEM;
>> +
>> +	mctl->media_dev = mdev;
>> +	if (stream == SNDRV_PCM_STREAM_PLAYBACK) {
>> +		intf_type = MEDIA_INTF_T_ALSA_PCM_PLAYBACK;
>> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_PLAYBACK;
>> +		mctl->media_pad.flags = MEDIA_PAD_FL_SOURCE;
>> +		mixer_pad = 1;
>> +	} else {
>> +		intf_type = MEDIA_INTF_T_ALSA_PCM_CAPTURE;
>> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_CAPTURE;
>> +		mctl->media_pad.flags = MEDIA_PAD_FL_SINK;
>> +		mixer_pad = 2;
>> +	}
>> +	mctl->media_entity.name = pcm->name;
>> +	media_entity_pads_init(&mctl->media_entity, 1, &mctl->media_pad);
>> +	ret =  media_device_register_entity(mctl->media_dev,
>> +					    &mctl->media_entity);
>> +	if (ret)
>> +		goto err1;
> 
> Could we give this label a meaningful name instead of a number?
> goto free_mctl;

I do see other places where numbered labels are used.
Names might help with code readability.

register_entity_fail probably makes more sense as a
label than free_mctl. In any case, I can address the
labels in a follow-on patch.

> 
>> +
>> +	mctl->intf_devnode = media_devnode_create(mdev, intf_type, 0,
>> +						  MAJOR(pcm_dev->devt),
>> +						  MINOR(pcm_dev->devt));
>> +	if (!mctl->intf_devnode) {
>> +		ret = -ENOMEM;
>> +		goto err2;
> 
> goto unregister_device;
> 
>> +	}
>> +	mctl->intf_link = media_create_intf_link(&mctl->media_entity,
>> +						 &mctl->intf_devnode->intf,
>> +						 MEDIA_LNK_FL_ENABLED);
>> +	if (!mctl->intf_link) {
>> +		ret = -ENOMEM;
>> +		goto err3;
> 
> goto delete_devnode;
> 
>> +	}
>> +
>> +	/* create link between mixer and audio */
>> +	media_device_for_each_entity(entity, mdev) {
>> +		switch (entity->function) {
>> +		case MEDIA_ENT_F_AUDIO_MIXER:
>> +			ret = media_create_pad_link(entity, mixer_pad,
>> +						    &mctl->media_entity, 0,
>> +						    MEDIA_LNK_FL_ENABLED);
>> +			if (ret)
>> +				goto err4;
> 
> This is a bit weird because we're inside a loop.  Shouldn't we call
> media_entity_remove_links() or something if this is the second time
> through the loop?

Links are removed from media_device_unregister_entity()
which is called in the error path.

> 
> I don't understand this.  The kernel has the media_entity_cleanup() stub
> function which is supposed to do this but it hasn't been implemented
> yet?
> 

Please see above. Links are removed when entity is
unregistered. media_entity_cleanup() is a stub. It
isn't intended for removing links.

thanks,
-- Shuah

-- 
Shuah Khan
Sr. Linux Kernel Developer
Open Source Innovation Group
Samsung Research America (Silicon Valley)
shuahkh@osg.samsung.com | (970) 217-8978

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Shuah Khan]

On 03/02/2016 01:41 PM, Dan Carpenter wrote:
> On Wed, Mar 02, 2016 at 09:50:31AM -0700, Shuah Khan wrote:
>> +	mctl = kzalloc(sizeof(*mctl), GFP_KERNEL);
>> +	if (!mctl)
>> +		return -ENOMEM;
>> +
>> +	mctl->media_dev = mdev;
>> +	if (stream == SNDRV_PCM_STREAM_PLAYBACK) {
>> +		intf_type = MEDIA_INTF_T_ALSA_PCM_PLAYBACK;
>> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_PLAYBACK;
>> +		mctl->media_pad.flags = MEDIA_PAD_FL_SOURCE;
>> +		mixer_pad = 1;
>> +	} else {
>> +		intf_type = MEDIA_INTF_T_ALSA_PCM_CAPTURE;
>> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_CAPTURE;
>> +		mctl->media_pad.flags = MEDIA_PAD_FL_SINK;
>> +		mixer_pad = 2;
>> +	}
>> +	mctl->media_entity.name = pcm->name;
>> +	media_entity_pads_init(&mctl->media_entity, 1, &mctl->media_pad);
>> +	ret =  media_device_register_entity(mctl->media_dev,
>> +					    &mctl->media_entity);
>> +	if (ret)
>> +		goto err1;
> 
> Could we give this label a meaningful name instead of a number?
> goto free_mctl;

I do see other places where numbered labels are used.
Names might help with code readability.

register_entity_fail probably makes more sense as a
label than free_mctl. In any case, I can address the
labels in a follow-on patch.

> 
>> +
>> +	mctl->intf_devnode = media_devnode_create(mdev, intf_type, 0,
>> +						  MAJOR(pcm_dev->devt),
>> +						  MINOR(pcm_dev->devt));
>> +	if (!mctl->intf_devnode) {
>> +		ret = -ENOMEM;
>> +		goto err2;
> 
> goto unregister_device;
> 
>> +	}
>> +	mctl->intf_link = media_create_intf_link(&mctl->media_entity,
>> +						 &mctl->intf_devnode->intf,
>> +						 MEDIA_LNK_FL_ENABLED);
>> +	if (!mctl->intf_link) {
>> +		ret = -ENOMEM;
>> +		goto err3;
> 
> goto delete_devnode;
> 
>> +	}
>> +
>> +	/* create link between mixer and audio */
>> +	media_device_for_each_entity(entity, mdev) {
>> +		switch (entity->function) {
>> +		case MEDIA_ENT_F_AUDIO_MIXER:
>> +			ret = media_create_pad_link(entity, mixer_pad,
>> +						    &mctl->media_entity, 0,
>> +						    MEDIA_LNK_FL_ENABLED);
>> +			if (ret)
>> +				goto err4;
> 
> This is a bit weird because we're inside a loop.  Shouldn't we call
> media_entity_remove_links() or something if this is the second time
> through the loop?

Links are removed from media_device_unregister_entity()
which is called in the error path.

> 
> I don't understand this.  The kernel has the media_entity_cleanup() stub
> function which is supposed to do this but it hasn't been implemented
> yet?
> 

Please see above. Links are removed when entity is
unregistered. media_entity_cleanup() is a stub. It
isn't intended for removing links.

thanks,
-- Shuah

-- 
Shuah Khan
Sr. Linux Kernel Developer
Open Source Innovation Group
Samsung Research America (Silicon Valley)
shuahkh@osg.samsung.com | (970) 217-8978

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Dan Carpenter]

On Wed, Mar 02, 2016 at 03:57:03PM -0700, Shuah Khan wrote:
> On 03/02/2016 01:41 PM, Dan Carpenter wrote:
> > On Wed, Mar 02, 2016 at 09:50:31AM -0700, Shuah Khan wrote:
> >> +	mctl = kzalloc(sizeof(*mctl), GFP_KERNEL);
> >> +	if (!mctl)
> >> +		return -ENOMEM;
> >> +
> >> +	mctl->media_dev = mdev;
> >> +	if (stream == SNDRV_PCM_STREAM_PLAYBACK) {
> >> +		intf_type = MEDIA_INTF_T_ALSA_PCM_PLAYBACK;
> >> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_PLAYBACK;
> >> +		mctl->media_pad.flags = MEDIA_PAD_FL_SOURCE;
> >> +		mixer_pad = 1;
> >> +	} else {
> >> +		intf_type = MEDIA_INTF_T_ALSA_PCM_CAPTURE;
> >> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_CAPTURE;
> >> +		mctl->media_pad.flags = MEDIA_PAD_FL_SINK;
> >> +		mixer_pad = 2;
> >> +	}
> >> +	mctl->media_entity.name = pcm->name;
> >> +	media_entity_pads_init(&mctl->media_entity, 1, &mctl->media_pad);
> >> +	ret =  media_device_register_entity(mctl->media_dev,
> >> +					    &mctl->media_entity);
> >> +	if (ret)
> >> +		goto err1;
> > 
> > Could we give this label a meaningful name instead of a number?
> > goto free_mctl;
> 
> I do see other places where numbered labels are used.

Yeah.  But it's the wrong idea.  If you remove a label then you have to
renumber everything.  Plus it doesn't say what the goto does.

> Names might help with code readability.
> 
> register_entity_fail probably makes more sense as a
> label than free_mctl.

A lot of people do that but the you wind up with code like:

	foo = whatever_register();
	if (!foo)
		goto whatever_failed;

	if (something_else)
		goto whatever_failed;

So it doesn't make sense.  Also it doesn't tell you what the goto does.
(You can already see that whatever_failed from looking at the if
statement on the line before.)  This function is larger than a page so
you have to flip down to the other page to see what the goto does, then
you have to flip back and find your place again.  But if the label says
what the goto does then you can just say, "Have we allocated anything
since mctl?  No, then freeing mctl is the right thing."  You don't need
to flip to the othe page.

regards,
dan carpenter

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Dan Carpenter]

On Wed, Mar 02, 2016 at 03:57:03PM -0700, Shuah Khan wrote:
> On 03/02/2016 01:41 PM, Dan Carpenter wrote:
> > On Wed, Mar 02, 2016 at 09:50:31AM -0700, Shuah Khan wrote:
> >> +	mctl = kzalloc(sizeof(*mctl), GFP_KERNEL);
> >> +	if (!mctl)
> >> +		return -ENOMEM;
> >> +
> >> +	mctl->media_dev = mdev;
> >> +	if (stream == SNDRV_PCM_STREAM_PLAYBACK) {
> >> +		intf_type = MEDIA_INTF_T_ALSA_PCM_PLAYBACK;
> >> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_PLAYBACK;
> >> +		mctl->media_pad.flags = MEDIA_PAD_FL_SOURCE;
> >> +		mixer_pad = 1;
> >> +	} else {
> >> +		intf_type = MEDIA_INTF_T_ALSA_PCM_CAPTURE;
> >> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_CAPTURE;
> >> +		mctl->media_pad.flags = MEDIA_PAD_FL_SINK;
> >> +		mixer_pad = 2;
> >> +	}
> >> +	mctl->media_entity.name = pcm->name;
> >> +	media_entity_pads_init(&mctl->media_entity, 1, &mctl->media_pad);
> >> +	ret =  media_device_register_entity(mctl->media_dev,
> >> +					    &mctl->media_entity);
> >> +	if (ret)
> >> +		goto err1;
> > 
> > Could we give this label a meaningful name instead of a number?
> > goto free_mctl;
> 
> I do see other places where numbered labels are used.

Yeah.  But it's the wrong idea.  If you remove a label then you have to
renumber everything.  Plus it doesn't say what the goto does.

> Names might help with code readability.
> 
> register_entity_fail probably makes more sense as a
> label than free_mctl.

A lot of people do that but the you wind up with code like:

	foo = whatever_register();
	if (!foo)
		goto whatever_failed;

	if (something_else)
		goto whatever_failed;

So it doesn't make sense.  Also it doesn't tell you what the goto does.
(You can already see that whatever_failed from looking at the if
statement on the line before.)  This function is larger than a page so
you have to flip down to the other page to see what the goto does, then
you have to flip back and find your place again.  But if the label says
what the goto does then you can just say, "Have we allocated anything
since mctl?  No, then freeing mctl is the right thing."  You don't need
to flip to the othe page.

regards,
dan carpenter

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Shuah Khan]

On 03/03/2016 03:36 AM, Dan Carpenter wrote:
> On Wed, Mar 02, 2016 at 03:57:03PM -0700, Shuah Khan wrote:
>> On 03/02/2016 01:41 PM, Dan Carpenter wrote:
>>> On Wed, Mar 02, 2016 at 09:50:31AM -0700, Shuah Khan wrote:
>>>> +	mctl = kzalloc(sizeof(*mctl), GFP_KERNEL);
>>>> +	if (!mctl)
>>>> +		return -ENOMEM;
>>>> +
>>>> +	mctl->media_dev = mdev;
>>>> +	if (stream == SNDRV_PCM_STREAM_PLAYBACK) {
>>>> +		intf_type = MEDIA_INTF_T_ALSA_PCM_PLAYBACK;
>>>> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_PLAYBACK;
>>>> +		mctl->media_pad.flags = MEDIA_PAD_FL_SOURCE;
>>>> +		mixer_pad = 1;
>>>> +	} else {
>>>> +		intf_type = MEDIA_INTF_T_ALSA_PCM_CAPTURE;
>>>> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_CAPTURE;
>>>> +		mctl->media_pad.flags = MEDIA_PAD_FL_SINK;
>>>> +		mixer_pad = 2;
>>>> +	}
>>>> +	mctl->media_entity.name = pcm->name;
>>>> +	media_entity_pads_init(&mctl->media_entity, 1, &mctl->media_pad);
>>>> +	ret =  media_device_register_entity(mctl->media_dev,
>>>> +					    &mctl->media_entity);
>>>> +	if (ret)
>>>> +		goto err1;
>>>
>>> Could we give this label a meaningful name instead of a number?
>>> goto free_mctl;
>>
>> I do see other places where numbered labels are used.
> 
> Yeah.  But it's the wrong idea.  If you remove a label then you have to
> renumber everything.  Plus it doesn't say what the goto does.
> 
>> Names might help with code readability.
>>
>> register_entity_fail probably makes more sense as a
>> label than free_mctl.
> 
> A lot of people do that but the you wind up with code like:
> 
> 	foo = whatever_register();
> 	if (!foo)
> 		goto whatever_failed;
> 
> 	if (something_else)
> 		goto whatever_failed;
> 
> So it doesn't make sense.  Also it doesn't tell you what the goto does.
> (You can already see that whatever_failed from looking at the if
> statement on the line before.)  This function is larger than a page so
> you have to flip down to the other page to see what the goto does, then
> you have to flip back and find your place again.  But if the label says
> what the goto does then you can just say, "Have we allocated anything
> since mctl?  No, then freeing mctl is the right thing."  You don't need
> to flip to the othe page.
> 

Yeah. You are right about large routines and needing
to check labels. I will fix it in a followup patch.

thanks,
-- Shuah

-- 
Shuah Khan
Sr. Linux Kernel Developer
Open Source Innovation Group
Samsung Research America (Silicon Valley)
shuahkh@osg.samsung.com | (970) 217-8978

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Shuah Khan]

On 03/03/2016 03:36 AM, Dan Carpenter wrote:
> On Wed, Mar 02, 2016 at 03:57:03PM -0700, Shuah Khan wrote:
>> On 03/02/2016 01:41 PM, Dan Carpenter wrote:
>>> On Wed, Mar 02, 2016 at 09:50:31AM -0700, Shuah Khan wrote:
>>>> +	mctl = kzalloc(sizeof(*mctl), GFP_KERNEL);
>>>> +	if (!mctl)
>>>> +		return -ENOMEM;
>>>> +
>>>> +	mctl->media_dev = mdev;
>>>> +	if (stream == SNDRV_PCM_STREAM_PLAYBACK) {
>>>> +		intf_type = MEDIA_INTF_T_ALSA_PCM_PLAYBACK;
>>>> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_PLAYBACK;
>>>> +		mctl->media_pad.flags = MEDIA_PAD_FL_SOURCE;
>>>> +		mixer_pad = 1;
>>>> +	} else {
>>>> +		intf_type = MEDIA_INTF_T_ALSA_PCM_CAPTURE;
>>>> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_CAPTURE;
>>>> +		mctl->media_pad.flags = MEDIA_PAD_FL_SINK;
>>>> +		mixer_pad = 2;
>>>> +	}
>>>> +	mctl->media_entity.name = pcm->name;
>>>> +	media_entity_pads_init(&mctl->media_entity, 1, &mctl->media_pad);
>>>> +	ret =  media_device_register_entity(mctl->media_dev,
>>>> +					    &mctl->media_entity);
>>>> +	if (ret)
>>>> +		goto err1;
>>>
>>> Could we give this label a meaningful name instead of a number?
>>> goto free_mctl;
>>
>> I do see other places where numbered labels are used.
> 
> Yeah.  But it's the wrong idea.  If you remove a label then you have to
> renumber everything.  Plus it doesn't say what the goto does.
> 
>> Names might help with code readability.
>>
>> register_entity_fail probably makes more sense as a
>> label than free_mctl.
> 
> A lot of people do that but the you wind up with code like:
> 
> 	foo = whatever_register();
> 	if (!foo)
> 		goto whatever_failed;
> 
> 	if (something_else)
> 		goto whatever_failed;
> 
> So it doesn't make sense.  Also it doesn't tell you what the goto does.
> (You can already see that whatever_failed from looking at the if
> statement on the line before.)  This function is larger than a page so
> you have to flip down to the other page to see what the goto does, then
> you have to flip back and find your place again.  But if the label says
> what the goto does then you can just say, "Have we allocated anything
> since mctl?  No, then freeing mctl is the right thing."  You don't need
> to flip to the othe page.
> 

Yeah. You are right about large routines and needing
to check labels. I will fix it in a followup patch.

thanks,
-- Shuah

-- 
Shuah Khan
Sr. Linux Kernel Developer
Open Source Innovation Group
Samsung Research America (Silicon Valley)
shuahkh@osg.samsung.com | (970) 217-8978

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Mauro Carvalho Chehab]

Em Wed, 2 Mar 2016 15:57:03 -0700
Shuah Khan <shuahkh@osg.samsung.com> escreveu:

> On 03/02/2016 01:41 PM, Dan Carpenter wrote:
> > On Wed, Mar 02, 2016 at 09:50:31AM -0700, Shuah Khan wrote:
> >> +	mctl = kzalloc(sizeof(*mctl), GFP_KERNEL);
> >> +	if (!mctl)
> >> +		return -ENOMEM;
> >> +
> >> +	mctl->media_dev = mdev;
> >> +	if (stream == SNDRV_PCM_STREAM_PLAYBACK) {
> >> +		intf_type = MEDIA_INTF_T_ALSA_PCM_PLAYBACK;
> >> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_PLAYBACK;
> >> +		mctl->media_pad.flags = MEDIA_PAD_FL_SOURCE;
> >> +		mixer_pad = 1;
> >> +	} else {
> >> +		intf_type = MEDIA_INTF_T_ALSA_PCM_CAPTURE;
> >> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_CAPTURE;
> >> +		mctl->media_pad.flags = MEDIA_PAD_FL_SINK;
> >> +		mixer_pad = 2;
> >> +	}
> >> +	mctl->media_entity.name = pcm->name;
> >> +	media_entity_pads_init(&mctl->media_entity, 1, &mctl->media_pad);
> >> +	ret =  media_device_register_entity(mctl->media_dev,
> >> +					    &mctl->media_entity);
> >> +	if (ret)
> >> +		goto err1;
> > 
> > Could we give this label a meaningful name instead of a number?
> > goto free_mctl;
> 
> I do see other places where numbered labels are used.
> Names might help with code readability.
> 
> register_entity_fail probably makes more sense as a
> label than free_mctl. In any case, I can address the
> labels in a follow-on patch.
> 
> > 
> >> +
> >> +	mctl->intf_devnode = media_devnode_create(mdev, intf_type, 0,
> >> +						  MAJOR(pcm_dev->devt),
> >> +						  MINOR(pcm_dev->devt));
> >> +	if (!mctl->intf_devnode) {
> >> +		ret = -ENOMEM;
> >> +		goto err2;
> > 
> > goto unregister_device;
> > 
> >> +	}
> >> +	mctl->intf_link = media_create_intf_link(&mctl->media_entity,
> >> +						 &mctl->intf_devnode->intf,
> >> +						 MEDIA_LNK_FL_ENABLED);
> >> +	if (!mctl->intf_link) {
> >> +		ret = -ENOMEM;
> >> +		goto err3;
> > 
> > goto delete_devnode;
> > 
> >> +	}
> >> +
> >> +	/* create link between mixer and audio */
> >> +	media_device_for_each_entity(entity, mdev) {
> >> +		switch (entity->function) {
> >> +		case MEDIA_ENT_F_AUDIO_MIXER:
> >> +			ret = media_create_pad_link(entity, mixer_pad,
> >> +						    &mctl->media_entity, 0,
> >> +						    MEDIA_LNK_FL_ENABLED);
> >> +			if (ret)
> >> +				goto err4;
> > 
> > This is a bit weird because we're inside a loop.  Shouldn't we call
> > media_entity_remove_links() or something if this is the second time
> > through the loop?
> 
> Links are removed from media_device_unregister_entity()
> which is called in the error path.
> 
> > 
> > I don't understand this.  The kernel has the media_entity_cleanup() stub
> > function which is supposed to do this but it hasn't been implemented
> > yet?
> > 
> 
> Please see above. Links are removed when entity is
> unregistered. media_entity_cleanup() is a stub. It
> isn't intended for removing links.

Shuah,

Please address the issue that Dan is pointing on a separate
patch.

Thanks,
Mauro

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Mauro Carvalho Chehab]

Em Wed, 2 Mar 2016 15:57:03 -0700
Shuah Khan <shuahkh@osg.samsung.com> escreveu:

> On 03/02/2016 01:41 PM, Dan Carpenter wrote:
> > On Wed, Mar 02, 2016 at 09:50:31AM -0700, Shuah Khan wrote:
> >> +	mctl = kzalloc(sizeof(*mctl), GFP_KERNEL);
> >> +	if (!mctl)
> >> +		return -ENOMEM;
> >> +
> >> +	mctl->media_dev = mdev;
> >> +	if (stream == SNDRV_PCM_STREAM_PLAYBACK) {
> >> +		intf_type = MEDIA_INTF_T_ALSA_PCM_PLAYBACK;
> >> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_PLAYBACK;
> >> +		mctl->media_pad.flags = MEDIA_PAD_FL_SOURCE;
> >> +		mixer_pad = 1;
> >> +	} else {
> >> +		intf_type = MEDIA_INTF_T_ALSA_PCM_CAPTURE;
> >> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_CAPTURE;
> >> +		mctl->media_pad.flags = MEDIA_PAD_FL_SINK;
> >> +		mixer_pad = 2;
> >> +	}
> >> +	mctl->media_entity.name = pcm->name;
> >> +	media_entity_pads_init(&mctl->media_entity, 1, &mctl->media_pad);
> >> +	ret =  media_device_register_entity(mctl->media_dev,
> >> +					    &mctl->media_entity);
> >> +	if (ret)
> >> +		goto err1;
> > 
> > Could we give this label a meaningful name instead of a number?
> > goto free_mctl;
> 
> I do see other places where numbered labels are used.
> Names might help with code readability.
> 
> register_entity_fail probably makes more sense as a
> label than free_mctl. In any case, I can address the
> labels in a follow-on patch.
> 
> > 
> >> +
> >> +	mctl->intf_devnode = media_devnode_create(mdev, intf_type, 0,
> >> +						  MAJOR(pcm_dev->devt),
> >> +						  MINOR(pcm_dev->devt));
> >> +	if (!mctl->intf_devnode) {
> >> +		ret = -ENOMEM;
> >> +		goto err2;
> > 
> > goto unregister_device;
> > 
> >> +	}
> >> +	mctl->intf_link = media_create_intf_link(&mctl->media_entity,
> >> +						 &mctl->intf_devnode->intf,
> >> +						 MEDIA_LNK_FL_ENABLED);
> >> +	if (!mctl->intf_link) {
> >> +		ret = -ENOMEM;
> >> +		goto err3;
> > 
> > goto delete_devnode;
> > 
> >> +	}
> >> +
> >> +	/* create link between mixer and audio */
> >> +	media_device_for_each_entity(entity, mdev) {
> >> +		switch (entity->function) {
> >> +		case MEDIA_ENT_F_AUDIO_MIXER:
> >> +			ret = media_create_pad_link(entity, mixer_pad,
> >> +						    &mctl->media_entity, 0,
> >> +						    MEDIA_LNK_FL_ENABLED);
> >> +			if (ret)
> >> +				goto err4;
> > 
> > This is a bit weird because we're inside a loop.  Shouldn't we call
> > media_entity_remove_links() or something if this is the second time
> > through the loop?
> 
> Links are removed from media_device_unregister_entity()
> which is called in the error path.
> 
> > 
> > I don't understand this.  The kernel has the media_entity_cleanup() stub
> > function which is supposed to do this but it hasn't been implemented
> > yet?
> > 
> 
> Please see above. Links are removed when entity is
> unregistered. media_entity_cleanup() is a stub. It
> isn't intended for removing links.

Shuah,

Please address the issue that Dan is pointing on a separate
patch.

Thanks,
Mauro

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Shuah Khan]

On 03/03/2016 11:21 AM, Mauro Carvalho Chehab wrote:
> Em Wed, 2 Mar 2016 15:57:03 -0700
> Shuah Khan <shuahkh@osg.samsung.com> escreveu:
> 
>> On 03/02/2016 01:41 PM, Dan Carpenter wrote:
>>> On Wed, Mar 02, 2016 at 09:50:31AM -0700, Shuah Khan wrote:
>>>> +	mctl = kzalloc(sizeof(*mctl), GFP_KERNEL);
>>>> +	if (!mctl)
>>>> +		return -ENOMEM;
>>>> +
>>>> +	mctl->media_dev = mdev;
>>>> +	if (stream == SNDRV_PCM_STREAM_PLAYBACK) {
>>>> +		intf_type = MEDIA_INTF_T_ALSA_PCM_PLAYBACK;
>>>> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_PLAYBACK;
>>>> +		mctl->media_pad.flags = MEDIA_PAD_FL_SOURCE;
>>>> +		mixer_pad = 1;
>>>> +	} else {
>>>> +		intf_type = MEDIA_INTF_T_ALSA_PCM_CAPTURE;
>>>> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_CAPTURE;
>>>> +		mctl->media_pad.flags = MEDIA_PAD_FL_SINK;
>>>> +		mixer_pad = 2;
>>>> +	}
>>>> +	mctl->media_entity.name = pcm->name;
>>>> +	media_entity_pads_init(&mctl->media_entity, 1, &mctl->media_pad);
>>>> +	ret =  media_device_register_entity(mctl->media_dev,
>>>> +					    &mctl->media_entity);
>>>> +	if (ret)
>>>> +		goto err1;
>>>
>>> Could we give this label a meaningful name instead of a number?
>>> goto free_mctl;
>>
>> I do see other places where numbered labels are used.
>> Names might help with code readability.
>>
>> register_entity_fail probably makes more sense as a
>> label than free_mctl. In any case, I can address the
>> labels in a follow-on patch.
>>
>>>
>>>> +
>>>> +	mctl->intf_devnode = media_devnode_create(mdev, intf_type, 0,
>>>> +						  MAJOR(pcm_dev->devt),
>>>> +						  MINOR(pcm_dev->devt));
>>>> +	if (!mctl->intf_devnode) {
>>>> +		ret = -ENOMEM;
>>>> +		goto err2;
>>>
>>> goto unregister_device;
>>>
>>>> +	}
>>>> +	mctl->intf_link = media_create_intf_link(&mctl->media_entity,
>>>> +						 &mctl->intf_devnode->intf,
>>>> +						 MEDIA_LNK_FL_ENABLED);
>>>> +	if (!mctl->intf_link) {
>>>> +		ret = -ENOMEM;
>>>> +		goto err3;
>>>
>>> goto delete_devnode;
>>>
>>>> +	}
>>>> +
>>>> +	/* create link between mixer and audio */
>>>> +	media_device_for_each_entity(entity, mdev) {
>>>> +		switch (entity->function) {
>>>> +		case MEDIA_ENT_F_AUDIO_MIXER:
>>>> +			ret = media_create_pad_link(entity, mixer_pad,
>>>> +						    &mctl->media_entity, 0,
>>>> +						    MEDIA_LNK_FL_ENABLED);
>>>> +			if (ret)
>>>> +				goto err4;
>>>
>>> This is a bit weird because we're inside a loop.  Shouldn't we call
>>> media_entity_remove_links() or something if this is the second time
>>> through the loop?
>>
>> Links are removed from media_device_unregister_entity()
>> which is called in the error path.
>>
>>>
>>> I don't understand this.  The kernel has the media_entity_cleanup() stub
>>> function which is supposed to do this but it hasn't been implemented
>>> yet?
>>>
>>
>> Please see above. Links are removed when entity is
>> unregistered. media_entity_cleanup() is a stub. It
>> isn't intended for removing links.
> 
> Shuah,
> 
> Please address the issue that Dan is pointing on a separate
> patch.
> 

Mauro,

Sent it a couple of hours ago. 

https://lkml.org/lkml/2016/3/3/601
https://www.mail-archive.com/linux-media@vger.kernel.org/msg95141.html

thanks,
-- Shuah


-- 
Shuah Khan
Sr. Linux Kernel Developer
Open Source Innovation Group
Samsung Research America (Silicon Valley)
shuahkh@osg.samsung.com | (970) 217-8978

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Shuah Khan]

On 03/03/2016 11:21 AM, Mauro Carvalho Chehab wrote:
> Em Wed, 2 Mar 2016 15:57:03 -0700
> Shuah Khan <shuahkh@osg.samsung.com> escreveu:
> 
>> On 03/02/2016 01:41 PM, Dan Carpenter wrote:
>>> On Wed, Mar 02, 2016 at 09:50:31AM -0700, Shuah Khan wrote:
>>>> +	mctl = kzalloc(sizeof(*mctl), GFP_KERNEL);
>>>> +	if (!mctl)
>>>> +		return -ENOMEM;
>>>> +
>>>> +	mctl->media_dev = mdev;
>>>> +	if (stream == SNDRV_PCM_STREAM_PLAYBACK) {
>>>> +		intf_type = MEDIA_INTF_T_ALSA_PCM_PLAYBACK;
>>>> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_PLAYBACK;
>>>> +		mctl->media_pad.flags = MEDIA_PAD_FL_SOURCE;
>>>> +		mixer_pad = 1;
>>>> +	} else {
>>>> +		intf_type = MEDIA_INTF_T_ALSA_PCM_CAPTURE;
>>>> +		mctl->media_entity.function = MEDIA_ENT_F_AUDIO_CAPTURE;
>>>> +		mctl->media_pad.flags = MEDIA_PAD_FL_SINK;
>>>> +		mixer_pad = 2;
>>>> +	}
>>>> +	mctl->media_entity.name = pcm->name;
>>>> +	media_entity_pads_init(&mctl->media_entity, 1, &mctl->media_pad);
>>>> +	ret =  media_device_register_entity(mctl->media_dev,
>>>> +					    &mctl->media_entity);
>>>> +	if (ret)
>>>> +		goto err1;
>>>
>>> Could we give this label a meaningful name instead of a number?
>>> goto free_mctl;
>>
>> I do see other places where numbered labels are used.
>> Names might help with code readability.
>>
>> register_entity_fail probably makes more sense as a
>> label than free_mctl. In any case, I can address the
>> labels in a follow-on patch.
>>
>>>
>>>> +
>>>> +	mctl->intf_devnode = media_devnode_create(mdev, intf_type, 0,
>>>> +						  MAJOR(pcm_dev->devt),
>>>> +						  MINOR(pcm_dev->devt));
>>>> +	if (!mctl->intf_devnode) {
>>>> +		ret = -ENOMEM;
>>>> +		goto err2;
>>>
>>> goto unregister_device;
>>>
>>>> +	}
>>>> +	mctl->intf_link = media_create_intf_link(&mctl->media_entity,
>>>> +						 &mctl->intf_devnode->intf,
>>>> +						 MEDIA_LNK_FL_ENABLED);
>>>> +	if (!mctl->intf_link) {
>>>> +		ret = -ENOMEM;
>>>> +		goto err3;
>>>
>>> goto delete_devnode;
>>>
>>>> +	}
>>>> +
>>>> +	/* create link between mixer and audio */
>>>> +	media_device_for_each_entity(entity, mdev) {
>>>> +		switch (entity->function) {
>>>> +		case MEDIA_ENT_F_AUDIO_MIXER:
>>>> +			ret = media_create_pad_link(entity, mixer_pad,
>>>> +						    &mctl->media_entity, 0,
>>>> +						    MEDIA_LNK_FL_ENABLED);
>>>> +			if (ret)
>>>> +				goto err4;
>>>
>>> This is a bit weird because we're inside a loop.  Shouldn't we call
>>> media_entity_remove_links() or something if this is the second time
>>> through the loop?
>>
>> Links are removed from media_device_unregister_entity()
>> which is called in the error path.
>>
>>>
>>> I don't understand this.  The kernel has the media_entity_cleanup() stub
>>> function which is supposed to do this but it hasn't been implemented
>>> yet?
>>>
>>
>> Please see above. Links are removed when entity is
>> unregistered. media_entity_cleanup() is a stub. It
>> isn't intended for removing links.
> 
> Shuah,
> 
> Please address the issue that Dan is pointing on a separate
> patch.
> 

Mauro,

Sent it a couple of hours ago. 

https://lkml.org/lkml/2016/3/3/601
https://www.mail-archive.com/linux-media@vger.kernel.org/msg95141.html

thanks,
-- Shuah


-- 
Shuah Khan
Sr. Linux Kernel Developer
Open Source Innovation Group
Samsung Research America (Silicon Valley)
shuahkh@osg.samsung.com | (970) 217-8978

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Mauro Carvalho Chehab]

Em Wed,  2 Mar 2016 09:50:31 -0700
Shuah Khan <shuahkh@osg.samsung.com> escreveu:

> Change ALSA driver to use Media Controller API to
> share media resources with DVB and V4L2 drivers
> on a AU0828 media device. Media Controller specific
> initialization is done after sound card is registered.
> ALSA creates Media interface and entity function graph
> nodes for Control, Mixer, PCM Playback, and PCM Capture
> devices.
> 
> snd_usb_hw_params() will call Media Controller enable
> source handler interface to request the media resource.
> If resource request is granted, it will release it from
> snd_usb_hw_free(). If resource is busy, -EBUSY is returned.
> 
> Media specific cleanup is done in usb_audio_disconnect().
> 
> Signed-off-by: Shuah Khan <shuahkh@osg.samsung.com>
> Acked-by: Takashi Iwai <tiwai@suse.de>

Shuah, by looking at the produced graphs:
	https://mchehab.fedorapeople.org/mc-next-gen/au0828_test/

I'm noticing the lack of ALSA I/O entities in the diagram
(MEDIA_ENT_F_AUDIO_CAPTURE). The mixer there is not connected.

Could you please check what's happening?

Those are the relevant dmesg data:

[   19.017276] usbcore: registered new interface driver snd-usb-audio
[  230.706102] Linux video capture interface: v2.00
[  230.856983] au0828: au0828 driver loaded
[  231.230612] au0828: i2c bus registered
[  231.822006] tveeprom 5-0050: Hauppauge model 72001, rev E1H3, serial# 4035199481
[  231.822991] tveeprom 5-0050: MAC address is 00:0d:fe:84:41:f9
[  231.823955] tveeprom 5-0050: tuner model is Xceive XC5000C (idx 173, type 88)
[  231.824782] tveeprom 5-0050: TV standards NTSC(M) ATSC/DVB Digital (eeprom 0x88)
[  231.825272] tveeprom 5-0050: audio processor is AU8522 (idx 44)
[  231.825276] tveeprom 5-0050: decoder processor is AU8522 (idx 42)
[  231.825280] tveeprom 5-0050: has no radio, has IR receiver, has no IR transmitter
[  231.825283] au0828: hauppauge_eeprom: hauppauge eeprom: model=72001
[  231.857567] au8522 5-0047: creating new instance
[  231.857879] au8522_decoder creating new instance...
[  231.910525] tuner 5-0061: Setting mode_mask to 0x06
[  231.910532] tuner 5-0061: tuner 0x61: Tuner type absent
[  231.910535] tuner 5-0061: Tuner -1 found with type(s) Radio TV.
[  231.911343] tuner 5-0061: Calling set_type_addr for type=88, addr=0x61, mode=0x04, config=ffffffffa0e64100
[  231.911347] tuner 5-0061: defining GPIO callback
[  231.934896] xc5000 5-0061: creating new instance
[  231.954664] xc5000: Successfully identified at address 0x61
[  231.954987] xc5000: Firmware has not been loaded previously
[  231.955327] tuner 5-0061: type set to Xceive XC5000
[  231.955330] tuner 5-0061: au0828 tuner I2C addr 0xc2 with type 88 used for 0x04
[  233.622614] au8522 5-0047: attaching existing instance
[  233.636061] xc5000 5-0061: attaching existing instance
[  233.645775] xc5000: Successfully identified at address 0x61
[  233.646034] xc5000: Firmware has not been loaded previously
[  233.646461] DVB: registering new adapter (au0828)
[  233.647575] usb 2-3.3: DVB: registering adapter 0 frontend 0 (Auvitek AU8522 QAM/8VSB Frontend)...
[  233.648344] dvb_create_media_entity: media entity 'Auvitek AU8522 QAM/8VSB Frontend' registered.
[  233.656105] dvb_create_media_entity: media entity 'dvb-demux' registered.
[  234.166644] IR keymap rc-hauppauge not found
[  234.166962] Registered IR keymap rc-empty
[  234.178055] input: au0828 IR (Hauppauge HVR950Q) as /devices/pci0000:00/0000:00:14.0/usb2/2-3/2-3.3/rc/rc0/input14
[  234.203311] rc rc0: au0828 IR (Hauppauge HVR950Q) as /devices/pci0000:00/0000:00:14.0/usb2/2-3/2-3.3/rc/rc0
[  234.207828] au0828: Remote controller au0828 IR (Hauppauge HVR950Q) initalized
[  234.208270] au0828: Registered device AU0828 [Hauppauge HVR950Q]
[  234.212073] usbcore: registered new interface driver au0828
[  234.230371] lirc_dev: IR Remote Control driver registered, major 243 
[  234.257555] rc rc0: lirc_dev: driver ir-lirc-codec (au0828-input) registered at minor = 0
[  234.257960] IR LIRC bridge handler initialized

(as au0828 is blacklisted, snd-usb-audio was probed first)

====xxxx====

I'm also getting some other weird behavior when removing/reinserting
the modules a few times. OK, officially we don't support it, but,
as devices can be bind/unbind all the times, removing modules is
a way to simulate such things. Also, I use it a lot while testing
USB drivers ;)

This one is after removing both the media drivers and snd-usb-audio, 
and then modprobing snd-usb-audio:

[ 1839.510587] BUG: sleeping function called from invalid context at mm/slub.c:1289
[ 1839.510881] in_atomic(): 1, irqs_disabled(): 0, pid: 3479, name: modprobe
[ 1839.511157] 4 locks held by modprobe/3479:
[ 1839.511415]  #0:  (&dev->mutex){......}, at: [<ffffffff81ce8933>] __driver_attach+0xa3/0x160
[ 1839.512381]  #1:  (&dev->mutex){......}, at: [<ffffffff81ce8941>] __driver_attach+0xb1/0x160
[ 1839.512388]  #2:  (register_mutex#5){+.+.+.}, at: [<ffffffffa10596c7>] usb_audio_probe+0x257/0x1c90 [snd_usb_audio]
[ 1839.512401]  #3:  (&(&mdev->lock)->rlock){+.+.+.}, at: [<ffffffffa0e6051b>] media_device_register_entity+0x1cb/0x700 [media]
[ 1839.512412] CPU: 2 PID: 3479 Comm: modprobe Not tainted 4.5.0-rc3+ #49
[ 1839.512415] Hardware name:                  /NUC5i7RYB, BIOS RYBDWi35.86A.0350.2015.0812.1722 08/12/2015
[ 1839.512417]  0000000000000000 ffff8803b3f6f288 ffffffff81933901 ffff8803c4bae000
[ 1839.512422]  ffff8803c4bae5c8 ffff8803b3f6f2b0 ffffffff811c6af5 ffff8803c4bae000
[ 1839.512427]  ffffffff8285d7f6 0000000000000509 ffff8803b3f6f2f0 ffffffff811c6ce5
[ 1839.512432] Call Trace:
[ 1839.512436]  [<ffffffff81933901>] dump_stack+0x85/0xc4
[ 1839.512440]  [<ffffffff811c6af5>] ___might_sleep+0x245/0x3a0
[ 1839.512443]  [<ffffffff811c6ce5>] __might_sleep+0x95/0x1a0
[ 1839.512446]  [<ffffffff8155aade>] kmem_cache_alloc_trace+0x20e/0x300
[ 1839.512451]  [<ffffffffa0e66e3d>] ? media_add_link+0x4d/0x140 [media]
[ 1839.512455]  [<ffffffffa0e66e3d>] media_add_link+0x4d/0x140 [media]
[ 1839.512459]  [<ffffffffa0e69931>] media_create_pad_link+0xa1/0x600 [media]
[ 1839.512463]  [<ffffffffa0fe11b3>] au0828_media_graph_notify+0x173/0x360 [au0828]
[ 1839.512467]  [<ffffffffa0e68a6a>] ? media_gobj_create+0x1ba/0x480 [media]
[ 1839.512471]  [<ffffffffa0e606fb>] media_device_register_entity+0x3ab/0x700 [media]
[ 1839.512475]  [<ffffffffa0e60350>] ? media_device_release_devres+0x10/0x10 [media]
[ 1839.512479]  [<ffffffff8124068d>] ? trace_hardirqs_on+0xd/0x10
[ 1839.512482]  [<ffffffff8159f992>] ? create_object+0x5a2/0x940
[ 1839.512485]  [<ffffffff8159f3f0>] ? kmemleak_disable+0x90/0x90
[ 1839.512489]  [<ffffffff8159f3f0>] ? kmemleak_disable+0x90/0x90
[ 1839.512497]  [<ffffffffa109722c>] ? media_snd_mixer_init+0x16c/0x500 [snd_usb_audio]
[ 1839.512501]  [<ffffffff8155f696>] ? kasan_unpoison_shadow+0x36/0x50
[ 1839.512504]  [<ffffffff8155f70e>] ? kasan_kmalloc+0x5e/0x70
[ 1839.512511]  [<ffffffffa1097352>] media_snd_mixer_init+0x292/0x500 [snd_usb_audio]
[ 1839.512519]  [<ffffffffa1097693>] media_snd_device_create+0xd3/0x2e0 [snd_usb_audio]
[ 1839.512525]  [<ffffffffa1059b05>] usb_audio_probe+0x695/0x1c90 [snd_usb_audio]
[ 1839.512529]  [<ffffffff8128b217>] ? debug_lockdep_rcu_enabled+0x77/0x90
[ 1839.512536]  [<ffffffffa1059470>] ? snd_usb_create_stream+0x4e0/0x4e0 [snd_usb_audio]
[ 1839.512540]  [<ffffffff812404fa>] ? trace_hardirqs_on_caller+0x40a/0x590
[ 1839.512543]  [<ffffffff8124068d>] ? trace_hardirqs_on+0xd/0x10
[ 1839.512546]  [<ffffffff81dbb1dd>] usb_probe_interface+0x45d/0x940
[ 1839.512550]  [<ffffffff81ce7e7a>] driver_probe_device+0x21a/0xc30
[ 1839.512553]  [<ffffffff81ce8890>] ? driver_probe_device+0xc30/0xc30
[ 1839.512556]  [<ffffffff81ce89b1>] __driver_attach+0x121/0x160
[ 1839.512559]  [<ffffffff81ce21bf>] bus_for_each_dev+0x11f/0x1a0
[ 1839.512562]  [<ffffffff81ce20a0>] ? subsys_dev_iter_exit+0x10/0x10
[ 1839.512565]  [<ffffffff822e7be7>] ? _raw_spin_unlock+0x27/0x40
[ 1839.512568]  [<ffffffff81ce6cdd>] driver_attach+0x3d/0x50
[ 1839.512571]  [<ffffffff81ce5df9>] bus_add_driver+0x4c9/0x770
[ 1839.512575]  [<ffffffff81cea39c>] driver_register+0x18c/0x3b0
[ 1839.512578]  [<ffffffff81250f92>] ? __raw_spin_lock_init+0x32/0x100
[ 1839.512581]  [<ffffffff81db6e98>] usb_register_driver+0x1f8/0x440
[ 1839.512584]  [<ffffffff810021a1>] ? do_one_initcall+0x131/0x300
[ 1839.512587]  [<ffffffffa10d8000>] ? 0xffffffffa10d8000
[ 1839.512594]  [<ffffffffa10d801e>] usb_audio_driver_init+0x1e/0x1000 [snd_usb_audio]
[ 1839.512597]  [<ffffffff810021b1>] do_one_initcall+0x141/0x300
[ 1839.512600]  [<ffffffff81002070>] ? try_to_run_init_process+0x40/0x40
[ 1839.512603]  [<ffffffff8155f696>] ? kasan_unpoison_shadow+0x36/0x50
[ 1839.512606]  [<ffffffff8155f696>] ? kasan_unpoison_shadow+0x36/0x50
[ 1839.512609]  [<ffffffff8155f7a7>] ? __asan_register_globals+0x87/0xa0
[ 1839.512613]  [<ffffffff8144d8eb>] do_init_module+0x1d0/0x5ad
[ 1839.512617]  [<ffffffff812f27b6>] load_module+0x6666/0x9ba0
[ 1839.512620]  [<ffffffff812e9e20>] ? symbol_put_addr+0x50/0x50
[ 1839.512627]  [<ffffffff812ec150>] ? module_frob_arch_sections+0x20/0x20
[ 1839.512630]  [<ffffffff815be010>] ? open_exec+0x50/0x50
[ 1839.512634]  [<ffffffff8116716b>] ? ns_capable+0x5b/0xd0
[ 1839.512637]  [<ffffffff812f5fe8>] SyS_finit_module+0x108/0x130
[ 1839.512640]  [<ffffffff812f5ee0>] ? SyS_init_module+0x1f0/0x1f0
[ 1839.512643]  [<ffffffff81004044>] ? lockdep_sys_exit_thunk+0x12/0x14
[ 1839.512647]  [<ffffffff822e8636>] entry_SYSCALL_64_fastpath+0x16/0x76
[ 1839.514027] usbcore: registered new interface driver snd-usb-audio

And an extra kmemleak issue:

unreferenced object 0xffff8803bcbacf30 (size 128):
  comm "modprobe", pid 2625, jiffies 4294950803 (age 1715.788s)
  hex dump (first 32 bytes):
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
  backtrace:
    [<ffffffff822cf76e>] kmemleak_alloc+0x4e/0xb0
    [<ffffffff8155ab06>] kmem_cache_alloc_trace+0x236/0x300
    [<ffffffff81937eb2>] ida_pre_get+0x1a2/0x250
    [<ffffffff81938005>] ida_simple_get+0xa5/0x170
    [<ffffffffa0f1469a>] 0xffffffffa0f1469a
    [<ffffffffa0fe6951>] au0828_dvb_register+0xe21/0x12c0 [au0828]
    [<ffffffffa0fd1d92>] vb2_thread+0xbb2/0xc10 [videobuf2_core]
    [<ffffffff81dbb1dd>] usb_probe_interface+0x45d/0x940
    [<ffffffff81ce7e7a>] driver_probe_device+0x21a/0xc30
    [<ffffffff81ce89b1>] __driver_attach+0x121/0x160
    [<ffffffff81ce21bf>] bus_for_each_dev+0x11f/0x1a0
    [<ffffffff81ce6cdd>] driver_attach+0x3d/0x50
    [<ffffffff81ce5df9>] bus_add_driver+0x4c9/0x770
    [<ffffffff81cea39c>] driver_register+0x18c/0x3b0
    [<ffffffff81db6e98>] usb_register_driver+0x1f8/0x440
    [<ffffffffa0fb80b7>] au8522_get_frontend+0xb7/0x160 [au8522_dig]

I did one extra module removal/reinsert, and this time, I got lots
of bugs. Full dmesg is at:
	http://pastebin.com/LEX5LD5K

The Kernel hanged after a while with (at the serial console - with
kind of explain why the message is truncated):

[ 2117.282666] kasan: GPF could be caused by NULL-ptr deref or user memory accessgeneral protection fault: 0000 [#2] SMP KASAN
[ 2117.284074] Modules linked in: videobuf2_v4l2 videobuf2_core tveeprom dvb_core rc_core v4l2_common videodev snd_usb_audio media cpufreq_powersave cpufreq_conservative cpufreq_userspace cpufreq_stats parport_pc ppdev lp parport snd_hda_codec_hdmi intel_rapl x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel iTCO_wdt iTCO_vendor_support kvm irqbypass crct10dif_pclmul crc32_

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Mauro Carvalho Chehab]

Em Wed,  2 Mar 2016 09:50:31 -0700
Shuah Khan <shuahkh@osg.samsung.com> escreveu:

> Change ALSA driver to use Media Controller API to
> share media resources with DVB and V4L2 drivers
> on a AU0828 media device. Media Controller specific
> initialization is done after sound card is registered.
> ALSA creates Media interface and entity function graph
> nodes for Control, Mixer, PCM Playback, and PCM Capture
> devices.
> 
> snd_usb_hw_params() will call Media Controller enable
> source handler interface to request the media resource.
> If resource request is granted, it will release it from
> snd_usb_hw_free(). If resource is busy, -EBUSY is returned.
> 
> Media specific cleanup is done in usb_audio_disconnect().
> 
> Signed-off-by: Shuah Khan <shuahkh@osg.samsung.com>
> Acked-by: Takashi Iwai <tiwai@suse.de>

Shuah, by looking at the produced graphs:
	https://mchehab.fedorapeople.org/mc-next-gen/au0828_test/

I'm noticing the lack of ALSA I/O entities in the diagram
(MEDIA_ENT_F_AUDIO_CAPTURE). The mixer there is not connected.

Could you please check what's happening?

Those are the relevant dmesg data:

[   19.017276] usbcore: registered new interface driver snd-usb-audio
[  230.706102] Linux video capture interface: v2.00
[  230.856983] au0828: au0828 driver loaded
[  231.230612] au0828: i2c bus registered
[  231.822006] tveeprom 5-0050: Hauppauge model 72001, rev E1H3, serial# 4035199481
[  231.822991] tveeprom 5-0050: MAC address is 00:0d:fe:84:41:f9
[  231.823955] tveeprom 5-0050: tuner model is Xceive XC5000C (idx 173, type 88)
[  231.824782] tveeprom 5-0050: TV standards NTSC(M) ATSC/DVB Digital (eeprom 0x88)
[  231.825272] tveeprom 5-0050: audio processor is AU8522 (idx 44)
[  231.825276] tveeprom 5-0050: decoder processor is AU8522 (idx 42)
[  231.825280] tveeprom 5-0050: has no radio, has IR receiver, has no IR transmitter
[  231.825283] au0828: hauppauge_eeprom: hauppauge eeprom: model=72001
[  231.857567] au8522 5-0047: creating new instance
[  231.857879] au8522_decoder creating new instance...
[  231.910525] tuner 5-0061: Setting mode_mask to 0x06
[  231.910532] tuner 5-0061: tuner 0x61: Tuner type absent
[  231.910535] tuner 5-0061: Tuner -1 found with type(s) Radio TV.
[  231.911343] tuner 5-0061: Calling set_type_addr for type=88, addr=0x61, mode=0x04, config=ffffffffa0e64100
[  231.911347] tuner 5-0061: defining GPIO callback
[  231.934896] xc5000 5-0061: creating new instance
[  231.954664] xc5000: Successfully identified at address 0x61
[  231.954987] xc5000: Firmware has not been loaded previously
[  231.955327] tuner 5-0061: type set to Xceive XC5000
[  231.955330] tuner 5-0061: au0828 tuner I2C addr 0xc2 with type 88 used for 0x04
[  233.622614] au8522 5-0047: attaching existing instance
[  233.636061] xc5000 5-0061: attaching existing instance
[  233.645775] xc5000: Successfully identified at address 0x61
[  233.646034] xc5000: Firmware has not been loaded previously
[  233.646461] DVB: registering new adapter (au0828)
[  233.647575] usb 2-3.3: DVB: registering adapter 0 frontend 0 (Auvitek AU8522 QAM/8VSB Frontend)...
[  233.648344] dvb_create_media_entity: media entity 'Auvitek AU8522 QAM/8VSB Frontend' registered.
[  233.656105] dvb_create_media_entity: media entity 'dvb-demux' registered.
[  234.166644] IR keymap rc-hauppauge not found
[  234.166962] Registered IR keymap rc-empty
[  234.178055] input: au0828 IR (Hauppauge HVR950Q) as /devices/pci0000:00/0000:00:14.0/usb2/2-3/2-3.3/rc/rc0/input14
[  234.203311] rc rc0: au0828 IR (Hauppauge HVR950Q) as /devices/pci0000:00/0000:00:14.0/usb2/2-3/2-3.3/rc/rc0
[  234.207828] au0828: Remote controller au0828 IR (Hauppauge HVR950Q) initalized
[  234.208270] au0828: Registered device AU0828 [Hauppauge HVR950Q]
[  234.212073] usbcore: registered new interface driver au0828
[  234.230371] lirc_dev: IR Remote Control driver registered, major 243 
[  234.257555] rc rc0: lirc_dev: driver ir-lirc-codec (au0828-input) registered at minor = 0
[  234.257960] IR LIRC bridge handler initialized

(as au0828 is blacklisted, snd-usb-audio was probed first)

====xxxx====

I'm also getting some other weird behavior when removing/reinserting
the modules a few times. OK, officially we don't support it, but,
as devices can be bind/unbind all the times, removing modules is
a way to simulate such things. Also, I use it a lot while testing
USB drivers ;)

This one is after removing both the media drivers and snd-usb-audio, 
and then modprobing snd-usb-audio:

[ 1839.510587] BUG: sleeping function called from invalid context at mm/slub.c:1289
[ 1839.510881] in_atomic(): 1, irqs_disabled(): 0, pid: 3479, name: modprobe
[ 1839.511157] 4 locks held by modprobe/3479:
[ 1839.511415]  #0:  (&dev->mutex){......}, at: [<ffffffff81ce8933>] __driver_attach+0xa3/0x160
[ 1839.512381]  #1:  (&dev->mutex){......}, at: [<ffffffff81ce8941>] __driver_attach+0xb1/0x160
[ 1839.512388]  #2:  (register_mutex#5){+.+.+.}, at: [<ffffffffa10596c7>] usb_audio_probe+0x257/0x1c90 [snd_usb_audio]
[ 1839.512401]  #3:  (&(&mdev->lock)->rlock){+.+.+.}, at: [<ffffffffa0e6051b>] media_device_register_entity+0x1cb/0x700 [media]
[ 1839.512412] CPU: 2 PID: 3479 Comm: modprobe Not tainted 4.5.0-rc3+ #49
[ 1839.512415] Hardware name:                  /NUC5i7RYB, BIOS RYBDWi35.86A.0350.2015.0812.1722 08/12/2015
[ 1839.512417]  0000000000000000 ffff8803b3f6f288 ffffffff81933901 ffff8803c4bae000
[ 1839.512422]  ffff8803c4bae5c8 ffff8803b3f6f2b0 ffffffff811c6af5 ffff8803c4bae000
[ 1839.512427]  ffffffff8285d7f6 0000000000000509 ffff8803b3f6f2f0 ffffffff811c6ce5
[ 1839.512432] Call Trace:
[ 1839.512436]  [<ffffffff81933901>] dump_stack+0x85/0xc4
[ 1839.512440]  [<ffffffff811c6af5>] ___might_sleep+0x245/0x3a0
[ 1839.512443]  [<ffffffff811c6ce5>] __might_sleep+0x95/0x1a0
[ 1839.512446]  [<ffffffff8155aade>] kmem_cache_alloc_trace+0x20e/0x300
[ 1839.512451]  [<ffffffffa0e66e3d>] ? media_add_link+0x4d/0x140 [media]
[ 1839.512455]  [<ffffffffa0e66e3d>] media_add_link+0x4d/0x140 [media]
[ 1839.512459]  [<ffffffffa0e69931>] media_create_pad_link+0xa1/0x600 [media]
[ 1839.512463]  [<ffffffffa0fe11b3>] au0828_media_graph_notify+0x173/0x360 [au0828]
[ 1839.512467]  [<ffffffffa0e68a6a>] ? media_gobj_create+0x1ba/0x480 [media]
[ 1839.512471]  [<ffffffffa0e606fb>] media_device_register_entity+0x3ab/0x700 [media]
[ 1839.512475]  [<ffffffffa0e60350>] ? media_device_release_devres+0x10/0x10 [media]
[ 1839.512479]  [<ffffffff8124068d>] ? trace_hardirqs_on+0xd/0x10
[ 1839.512482]  [<ffffffff8159f992>] ? create_object+0x5a2/0x940
[ 1839.512485]  [<ffffffff8159f3f0>] ? kmemleak_disable+0x90/0x90
[ 1839.512489]  [<ffffffff8159f3f0>] ? kmemleak_disable+0x90/0x90
[ 1839.512497]  [<ffffffffa109722c>] ? media_snd_mixer_init+0x16c/0x500 [snd_usb_audio]
[ 1839.512501]  [<ffffffff8155f696>] ? kasan_unpoison_shadow+0x36/0x50
[ 1839.512504]  [<ffffffff8155f70e>] ? kasan_kmalloc+0x5e/0x70
[ 1839.512511]  [<ffffffffa1097352>] media_snd_mixer_init+0x292/0x500 [snd_usb_audio]
[ 1839.512519]  [<ffffffffa1097693>] media_snd_device_create+0xd3/0x2e0 [snd_usb_audio]
[ 1839.512525]  [<ffffffffa1059b05>] usb_audio_probe+0x695/0x1c90 [snd_usb_audio]
[ 1839.512529]  [<ffffffff8128b217>] ? debug_lockdep_rcu_enabled+0x77/0x90
[ 1839.512536]  [<ffffffffa1059470>] ? snd_usb_create_stream+0x4e0/0x4e0 [snd_usb_audio]
[ 1839.512540]  [<ffffffff812404fa>] ? trace_hardirqs_on_caller+0x40a/0x590
[ 1839.512543]  [<ffffffff8124068d>] ? trace_hardirqs_on+0xd/0x10
[ 1839.512546]  [<ffffffff81dbb1dd>] usb_probe_interface+0x45d/0x940
[ 1839.512550]  [<ffffffff81ce7e7a>] driver_probe_device+0x21a/0xc30
[ 1839.512553]  [<ffffffff81ce8890>] ? driver_probe_device+0xc30/0xc30
[ 1839.512556]  [<ffffffff81ce89b1>] __driver_attach+0x121/0x160
[ 1839.512559]  [<ffffffff81ce21bf>] bus_for_each_dev+0x11f/0x1a0
[ 1839.512562]  [<ffffffff81ce20a0>] ? subsys_dev_iter_exit+0x10/0x10
[ 1839.512565]  [<ffffffff822e7be7>] ? _raw_spin_unlock+0x27/0x40
[ 1839.512568]  [<ffffffff81ce6cdd>] driver_attach+0x3d/0x50
[ 1839.512571]  [<ffffffff81ce5df9>] bus_add_driver+0x4c9/0x770
[ 1839.512575]  [<ffffffff81cea39c>] driver_register+0x18c/0x3b0
[ 1839.512578]  [<ffffffff81250f92>] ? __raw_spin_lock_init+0x32/0x100
[ 1839.512581]  [<ffffffff81db6e98>] usb_register_driver+0x1f8/0x440
[ 1839.512584]  [<ffffffff810021a1>] ? do_one_initcall+0x131/0x300
[ 1839.512587]  [<ffffffffa10d8000>] ? 0xffffffffa10d8000
[ 1839.512594]  [<ffffffffa10d801e>] usb_audio_driver_init+0x1e/0x1000 [snd_usb_audio]
[ 1839.512597]  [<ffffffff810021b1>] do_one_initcall+0x141/0x300
[ 1839.512600]  [<ffffffff81002070>] ? try_to_run_init_process+0x40/0x40
[ 1839.512603]  [<ffffffff8155f696>] ? kasan_unpoison_shadow+0x36/0x50
[ 1839.512606]  [<ffffffff8155f696>] ? kasan_unpoison_shadow+0x36/0x50
[ 1839.512609]  [<ffffffff8155f7a7>] ? __asan_register_globals+0x87/0xa0
[ 1839.512613]  [<ffffffff8144d8eb>] do_init_module+0x1d0/0x5ad
[ 1839.512617]  [<ffffffff812f27b6>] load_module+0x6666/0x9ba0
[ 1839.512620]  [<ffffffff812e9e20>] ? symbol_put_addr+0x50/0x50
[ 1839.512627]  [<ffffffff812ec150>] ? module_frob_arch_sections+0x20/0x20
[ 1839.512630]  [<ffffffff815be010>] ? open_exec+0x50/0x50
[ 1839.512634]  [<ffffffff8116716b>] ? ns_capable+0x5b/0xd0
[ 1839.512637]  [<ffffffff812f5fe8>] SyS_finit_module+0x108/0x130
[ 1839.512640]  [<ffffffff812f5ee0>] ? SyS_init_module+0x1f0/0x1f0
[ 1839.512643]  [<ffffffff81004044>] ? lockdep_sys_exit_thunk+0x12/0x14
[ 1839.512647]  [<ffffffff822e8636>] entry_SYSCALL_64_fastpath+0x16/0x76
[ 1839.514027] usbcore: registered new interface driver snd-usb-audio

And an extra kmemleak issue:

unreferenced object 0xffff8803bcbacf30 (size 128):
  comm "modprobe", pid 2625, jiffies 4294950803 (age 1715.788s)
  hex dump (first 32 bytes):
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
  backtrace:
    [<ffffffff822cf76e>] kmemleak_alloc+0x4e/0xb0
    [<ffffffff8155ab06>] kmem_cache_alloc_trace+0x236/0x300
    [<ffffffff81937eb2>] ida_pre_get+0x1a2/0x250
    [<ffffffff81938005>] ida_simple_get+0xa5/0x170
    [<ffffffffa0f1469a>] 0xffffffffa0f1469a
    [<ffffffffa0fe6951>] au0828_dvb_register+0xe21/0x12c0 [au0828]
    [<ffffffffa0fd1d92>] vb2_thread+0xbb2/0xc10 [videobuf2_core]
    [<ffffffff81dbb1dd>] usb_probe_interface+0x45d/0x940
    [<ffffffff81ce7e7a>] driver_probe_device+0x21a/0xc30
    [<ffffffff81ce89b1>] __driver_attach+0x121/0x160
    [<ffffffff81ce21bf>] bus_for_each_dev+0x11f/0x1a0
    [<ffffffff81ce6cdd>] driver_attach+0x3d/0x50
    [<ffffffff81ce5df9>] bus_add_driver+0x4c9/0x770
    [<ffffffff81cea39c>] driver_register+0x18c/0x3b0
    [<ffffffff81db6e98>] usb_register_driver+0x1f8/0x440
    [<ffffffffa0fb80b7>] au8522_get_frontend+0xb7/0x160 [au8522_dig]

I did one extra module removal/reinsert, and this time, I got lots
of bugs. Full dmesg is at:
	http://pastebin.com/LEX5LD5K

The Kernel hanged after a while with (at the serial console - with
kind of explain why the message is truncated):

[ 2117.282666] kasan: GPF could be caused by NULL-ptr deref or user memory accessgeneral protection fault: 0000 [#2] SMP KASAN
[ 2117.284074] Modules linked in: videobuf2_v4l2 videobuf2_core tveeprom dvb_core rc_core v4l2_common videodev snd_usb_audio media cpufreq_powersave cpufreq_conservative cpufreq_userspace cpufreq_stats parport_pc ppdev lp parport snd_hda_codec_hdmi intel_rapl x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel iTCO_wdt iTCO_vendor_support kvm irqbypass crct10dif_pclmul crc32_

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Shuah Khan]

On 03/05/2016 03:00 AM, Mauro Carvalho Chehab wrote:
> Em Wed,  2 Mar 2016 09:50:31 -0700
> Shuah Khan <shuahkh@osg.samsung.com> escreveu:
> 
>> Change ALSA driver to use Media Controller API to
>> share media resources with DVB and V4L2 drivers
>> on a AU0828 media device. Media Controller specific
>> initialization is done after sound card is registered.
>> ALSA creates Media interface and entity function graph
>> nodes for Control, Mixer, PCM Playback, and PCM Capture
>> devices.
>>
>> snd_usb_hw_params() will call Media Controller enable
>> source handler interface to request the media resource.
>> If resource request is granted, it will release it from
>> snd_usb_hw_free(). If resource is busy, -EBUSY is returned.
>>
>> Media specific cleanup is done in usb_audio_disconnect().
>>
>> Signed-off-by: Shuah Khan <shuahkh@osg.samsung.com>
>> Acked-by: Takashi Iwai <tiwai@suse.de>
> 
> Shuah, by looking at the produced graphs:
> 	https://mchehab.fedorapeople.org/mc-next-gen/au0828_test/
> 
> I'm noticing the lack of ALSA I/O entities in the diagram
> (MEDIA_ENT_F_AUDIO_CAPTURE). The mixer there is not connected.
> 
> Could you please check what's happening?

Mauro,

Did you pull in this patch that fixes the graph problem:

https://www.mail-archive.com/linux-media@vger.kernel.org/msg95047.html

If you haven't, could you please pull this in.


> 
> Those are the relevant dmesg data:
> 
> [   19.017276] usbcore: registered new interface driver snd-usb-audio
> [  230.706102] Linux video capture interface: v2.00
> [  230.856983] au0828: au0828 driver loaded
> [  231.230612] au0828: i2c bus registered
> [  231.822006] tveeprom 5-0050: Hauppauge model 72001, rev E1H3, serial# 4035199481
> [  231.822991] tveeprom 5-0050: MAC address is 00:0d:fe:84:41:f9
> [  231.823955] tveeprom 5-0050: tuner model is Xceive XC5000C (idx 173, type 88)
> [  231.824782] tveeprom 5-0050: TV standards NTSC(M) ATSC/DVB Digital (eeprom 0x88)
> [  231.825272] tveeprom 5-0050: audio processor is AU8522 (idx 44)
> [  231.825276] tveeprom 5-0050: decoder processor is AU8522 (idx 42)
> [  231.825280] tveeprom 5-0050: has no radio, has IR receiver, has no IR transmitter
> [  231.825283] au0828: hauppauge_eeprom: hauppauge eeprom: model=72001
> [  231.857567] au8522 5-0047: creating new instance
> [  231.857879] au8522_decoder creating new instance...
> [  231.910525] tuner 5-0061: Setting mode_mask to 0x06
> [  231.910532] tuner 5-0061: tuner 0x61: Tuner type absent
> [  231.910535] tuner 5-0061: Tuner -1 found with type(s) Radio TV.
> [  231.911343] tuner 5-0061: Calling set_type_addr for type=88, addr=0x61, mode=0x04, config=ffffffffa0e64100
> [  231.911347] tuner 5-0061: defining GPIO callback
> [  231.934896] xc5000 5-0061: creating new instance
> [  231.954664] xc5000: Successfully identified at address 0x61
> [  231.954987] xc5000: Firmware has not been loaded previously
> [  231.955327] tuner 5-0061: type set to Xceive XC5000
> [  231.955330] tuner 5-0061: au0828 tuner I2C addr 0xc2 with type 88 used for 0x04
> [  233.622614] au8522 5-0047: attaching existing instance
> [  233.636061] xc5000 5-0061: attaching existing instance
> [  233.645775] xc5000: Successfully identified at address 0x61
> [  233.646034] xc5000: Firmware has not been loaded previously
> [  233.646461] DVB: registering new adapter (au0828)
> [  233.647575] usb 2-3.3: DVB: registering adapter 0 frontend 0 (Auvitek AU8522 QAM/8VSB Frontend)...
> [  233.648344] dvb_create_media_entity: media entity 'Auvitek AU8522 QAM/8VSB Frontend' registered.
> [  233.656105] dvb_create_media_entity: media entity 'dvb-demux' registered.
> [  234.166644] IR keymap rc-hauppauge not found
> [  234.166962] Registered IR keymap rc-empty
> [  234.178055] input: au0828 IR (Hauppauge HVR950Q) as /devices/pci0000:00/0000:00:14.0/usb2/2-3/2-3.3/rc/rc0/input14
> [  234.203311] rc rc0: au0828 IR (Hauppauge HVR950Q) as /devices/pci0000:00/0000:00:14.0/usb2/2-3/2-3.3/rc/rc0
> [  234.207828] au0828: Remote controller au0828 IR (Hauppauge HVR950Q) initalized
> [  234.208270] au0828: Registered device AU0828 [Hauppauge HVR950Q]
> [  234.212073] usbcore: registered new interface driver au0828
> [  234.230371] lirc_dev: IR Remote Control driver registered, major 243 
> [  234.257555] rc rc0: lirc_dev: driver ir-lirc-codec (au0828-input) registered at minor = 0
> [  234.257960] IR LIRC bridge handler initialized
> 
> (as au0828 is blacklisted, snd-usb-audio was probed first)
> 
> ====xxxx====
> 
> I'm also getting some other weird behavior when removing/reinserting
> the modules a few times. OK, officially we don't support it, but,
> as devices can be bind/unbind all the times, removing modules is
> a way to simulate such things. Also, I use it a lot while testing
> USB drivers ;)

I will do more testing on this. I didn't see any problems when
I fixed the graph patch above - I will do more on the latest
media branch today.

thanks,
-- Shuah

> 
> This one is after removing both the media drivers and snd-usb-audio, 
> and then modprobing snd-usb-audio:
> 
> [ 1839.510587] BUG: sleeping function called from invalid context at mm/slub.c:1289
> [ 1839.510881] in_atomic(): 1, irqs_disabled(): 0, pid: 3479, name: modprobe
> [ 1839.511157] 4 locks held by modprobe/3479:
> [ 1839.511415]  #0:  (&dev->mutex){......}, at: [<ffffffff81ce8933>] __driver_attach+0xa3/0x160
> [ 1839.512381]  #1:  (&dev->mutex){......}, at: [<ffffffff81ce8941>] __driver_attach+0xb1/0x160
> [ 1839.512388]  #2:  (register_mutex#5){+.+.+.}, at: [<ffffffffa10596c7>] usb_audio_probe+0x257/0x1c90 [snd_usb_audio]
> [ 1839.512401]  #3:  (&(&mdev->lock)->rlock){+.+.+.}, at: [<ffffffffa0e6051b>] media_device_register_entity+0x1cb/0x700 [media]
> [ 1839.512412] CPU: 2 PID: 3479 Comm: modprobe Not tainted 4.5.0-rc3+ #49
> [ 1839.512415] Hardware name:                  /NUC5i7RYB, BIOS RYBDWi35.86A.0350.2015.0812.1722 08/12/2015
> [ 1839.512417]  0000000000000000 ffff8803b3f6f288 ffffffff81933901 ffff8803c4bae000
> [ 1839.512422]  ffff8803c4bae5c8 ffff8803b3f6f2b0 ffffffff811c6af5 ffff8803c4bae000
> [ 1839.512427]  ffffffff8285d7f6 0000000000000509 ffff8803b3f6f2f0 ffffffff811c6ce5
> [ 1839.512432] Call Trace:
> [ 1839.512436]  [<ffffffff81933901>] dump_stack+0x85/0xc4
> [ 1839.512440]  [<ffffffff811c6af5>] ___might_sleep+0x245/0x3a0
> [ 1839.512443]  [<ffffffff811c6ce5>] __might_sleep+0x95/0x1a0
> [ 1839.512446]  [<ffffffff8155aade>] kmem_cache_alloc_trace+0x20e/0x300
> [ 1839.512451]  [<ffffffffa0e66e3d>] ? media_add_link+0x4d/0x140 [media]
> [ 1839.512455]  [<ffffffffa0e66e3d>] media_add_link+0x4d/0x140 [media]
> [ 1839.512459]  [<ffffffffa0e69931>] media_create_pad_link+0xa1/0x600 [media]
> [ 1839.512463]  [<ffffffffa0fe11b3>] au0828_media_graph_notify+0x173/0x360 [au0828]
> [ 1839.512467]  [<ffffffffa0e68a6a>] ? media_gobj_create+0x1ba/0x480 [media]
> [ 1839.512471]  [<ffffffffa0e606fb>] media_device_register_entity+0x3ab/0x700 [media]
> [ 1839.512475]  [<ffffffffa0e60350>] ? media_device_release_devres+0x10/0x10 [media]
> [ 1839.512479]  [<ffffffff8124068d>] ? trace_hardirqs_on+0xd/0x10
> [ 1839.512482]  [<ffffffff8159f992>] ? create_object+0x5a2/0x940
> [ 1839.512485]  [<ffffffff8159f3f0>] ? kmemleak_disable+0x90/0x90
> [ 1839.512489]  [<ffffffff8159f3f0>] ? kmemleak_disable+0x90/0x90
> [ 1839.512497]  [<ffffffffa109722c>] ? media_snd_mixer_init+0x16c/0x500 [snd_usb_audio]
> [ 1839.512501]  [<ffffffff8155f696>] ? kasan_unpoison_shadow+0x36/0x50
> [ 1839.512504]  [<ffffffff8155f70e>] ? kasan_kmalloc+0x5e/0x70
> [ 1839.512511]  [<ffffffffa1097352>] media_snd_mixer_init+0x292/0x500 [snd_usb_audio]
> [ 1839.512519]  [<ffffffffa1097693>] media_snd_device_create+0xd3/0x2e0 [snd_usb_audio]
> [ 1839.512525]  [<ffffffffa1059b05>] usb_audio_probe+0x695/0x1c90 [snd_usb_audio]
> [ 1839.512529]  [<ffffffff8128b217>] ? debug_lockdep_rcu_enabled+0x77/0x90
> [ 1839.512536]  [<ffffffffa1059470>] ? snd_usb_create_stream+0x4e0/0x4e0 [snd_usb_audio]
> [ 1839.512540]  [<ffffffff812404fa>] ? trace_hardirqs_on_caller+0x40a/0x590
> [ 1839.512543]  [<ffffffff8124068d>] ? trace_hardirqs_on+0xd/0x10
> [ 1839.512546]  [<ffffffff81dbb1dd>] usb_probe_interface+0x45d/0x940
> [ 1839.512550]  [<ffffffff81ce7e7a>] driver_probe_device+0x21a/0xc30
> [ 1839.512553]  [<ffffffff81ce8890>] ? driver_probe_device+0xc30/0xc30
> [ 1839.512556]  [<ffffffff81ce89b1>] __driver_attach+0x121/0x160
> [ 1839.512559]  [<ffffffff81ce21bf>] bus_for_each_dev+0x11f/0x1a0
> [ 1839.512562]  [<ffffffff81ce20a0>] ? subsys_dev_iter_exit+0x10/0x10
> [ 1839.512565]  [<ffffffff822e7be7>] ? _raw_spin_unlock+0x27/0x40
> [ 1839.512568]  [<ffffffff81ce6cdd>] driver_attach+0x3d/0x50
> [ 1839.512571]  [<ffffffff81ce5df9>] bus_add_driver+0x4c9/0x770
> [ 1839.512575]  [<ffffffff81cea39c>] driver_register+0x18c/0x3b0
> [ 1839.512578]  [<ffffffff81250f92>] ? __raw_spin_lock_init+0x32/0x100
> [ 1839.512581]  [<ffffffff81db6e98>] usb_register_driver+0x1f8/0x440
> [ 1839.512584]  [<ffffffff810021a1>] ? do_one_initcall+0x131/0x300
> [ 1839.512587]  [<ffffffffa10d8000>] ? 0xffffffffa10d8000
> [ 1839.512594]  [<ffffffffa10d801e>] usb_audio_driver_init+0x1e/0x1000 [snd_usb_audio]
> [ 1839.512597]  [<ffffffff810021b1>] do_one_initcall+0x141/0x300
> [ 1839.512600]  [<ffffffff81002070>] ? try_to_run_init_process+0x40/0x40
> [ 1839.512603]  [<ffffffff8155f696>] ? kasan_unpoison_shadow+0x36/0x50
> [ 1839.512606]  [<ffffffff8155f696>] ? kasan_unpoison_shadow+0x36/0x50
> [ 1839.512609]  [<ffffffff8155f7a7>] ? __asan_register_globals+0x87/0xa0
> [ 1839.512613]  [<ffffffff8144d8eb>] do_init_module+0x1d0/0x5ad
> [ 1839.512617]  [<ffffffff812f27b6>] load_module+0x6666/0x9ba0
> [ 1839.512620]  [<ffffffff812e9e20>] ? symbol_put_addr+0x50/0x50
> [ 1839.512627]  [<ffffffff812ec150>] ? module_frob_arch_sections+0x20/0x20
> [ 1839.512630]  [<ffffffff815be010>] ? open_exec+0x50/0x50
> [ 1839.512634]  [<ffffffff8116716b>] ? ns_capable+0x5b/0xd0
> [ 1839.512637]  [<ffffffff812f5fe8>] SyS_finit_module+0x108/0x130
> [ 1839.512640]  [<ffffffff812f5ee0>] ? SyS_init_module+0x1f0/0x1f0
> [ 1839.512643]  [<ffffffff81004044>] ? lockdep_sys_exit_thunk+0x12/0x14
> [ 1839.512647]  [<ffffffff822e8636>] entry_SYSCALL_64_fastpath+0x16/0x76
> [ 1839.514027] usbcore: registered new interface driver snd-usb-audio
> 
> And an extra kmemleak issue:
> 
> unreferenced object 0xffff8803bcbacf30 (size 128):
>   comm "modprobe", pid 2625, jiffies 4294950803 (age 1715.788s)
>   hex dump (first 32 bytes):
>     00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
>     00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
>   backtrace:
>     [<ffffffff822cf76e>] kmemleak_alloc+0x4e/0xb0
>     [<ffffffff8155ab06>] kmem_cache_alloc_trace+0x236/0x300
>     [<ffffffff81937eb2>] ida_pre_get+0x1a2/0x250
>     [<ffffffff81938005>] ida_simple_get+0xa5/0x170
>     [<ffffffffa0f1469a>] 0xffffffffa0f1469a
>     [<ffffffffa0fe6951>] au0828_dvb_register+0xe21/0x12c0 [au0828]
>     [<ffffffffa0fd1d92>] vb2_thread+0xbb2/0xc10 [videobuf2_core]
>     [<ffffffff81dbb1dd>] usb_probe_interface+0x45d/0x940
>     [<ffffffff81ce7e7a>] driver_probe_device+0x21a/0xc30
>     [<ffffffff81ce89b1>] __driver_attach+0x121/0x160
>     [<ffffffff81ce21bf>] bus_for_each_dev+0x11f/0x1a0
>     [<ffffffff81ce6cdd>] driver_attach+0x3d/0x50
>     [<ffffffff81ce5df9>] bus_add_driver+0x4c9/0x770
>     [<ffffffff81cea39c>] driver_register+0x18c/0x3b0
>     [<ffffffff81db6e98>] usb_register_driver+0x1f8/0x440
>     [<ffffffffa0fb80b7>] au8522_get_frontend+0xb7/0x160 [au8522_dig]
> 
> I did one extra module removal/reinsert, and this time, I got lots
> of bugs. Full dmesg is at:
> 	http://pastebin.com/LEX5LD5K
> 
> The Kernel hanged after a while with (at the serial console - with
> kind of explain why the message is truncated):
> 
> [ 2117.282666] kasan: GPF could be caused by NULL-ptr deref or user memory accessgeneral protection fault: 0000 [#2] SMP KASAN
> [ 2117.284074] Modules linked in: videobuf2_v4l2 videobuf2_core tveeprom dvb_core rc_core v4l2_common videodev snd_usb_audio media cpufreq_powersave cpufreq_conservative cpufreq_userspace cpufreq_stats parport_pc ppdev lp parport snd_hda_codec_hdmi intel_rapl x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel iTCO_wdt iTCO_vendor_support kvm irqbypass crct10dif_pclmul crc32_
> 
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-media" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 


-- 
Shuah Khan
Sr. Linux Kernel Developer
Open Source Innovation Group
Samsung Research America (Silicon Valley)
shuahkh@osg.samsung.com | (970) 217-8978

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Shuah Khan]

On 03/05/2016 03:00 AM, Mauro Carvalho Chehab wrote:
> Em Wed,  2 Mar 2016 09:50:31 -0700
> Shuah Khan <shuahkh@osg.samsung.com> escreveu:
> 
>> Change ALSA driver to use Media Controller API to
>> share media resources with DVB and V4L2 drivers
>> on a AU0828 media device. Media Controller specific
>> initialization is done after sound card is registered.
>> ALSA creates Media interface and entity function graph
>> nodes for Control, Mixer, PCM Playback, and PCM Capture
>> devices.
>>
>> snd_usb_hw_params() will call Media Controller enable
>> source handler interface to request the media resource.
>> If resource request is granted, it will release it from
>> snd_usb_hw_free(). If resource is busy, -EBUSY is returned.
>>
>> Media specific cleanup is done in usb_audio_disconnect().
>>
>> Signed-off-by: Shuah Khan <shuahkh@osg.samsung.com>
>> Acked-by: Takashi Iwai <tiwai@suse.de>
> 
> Shuah, by looking at the produced graphs:
> 	https://mchehab.fedorapeople.org/mc-next-gen/au0828_test/
> 
> I'm noticing the lack of ALSA I/O entities in the diagram
> (MEDIA_ENT_F_AUDIO_CAPTURE). The mixer there is not connected.
> 
> Could you please check what's happening?

Mauro,

Did you pull in this patch that fixes the graph problem:

https://www.mail-archive.com/linux-media@vger.kernel.org/msg95047.html

If you haven't, could you please pull this in.


> 
> Those are the relevant dmesg data:
> 
> [   19.017276] usbcore: registered new interface driver snd-usb-audio
> [  230.706102] Linux video capture interface: v2.00
> [  230.856983] au0828: au0828 driver loaded
> [  231.230612] au0828: i2c bus registered
> [  231.822006] tveeprom 5-0050: Hauppauge model 72001, rev E1H3, serial# 4035199481
> [  231.822991] tveeprom 5-0050: MAC address is 00:0d:fe:84:41:f9
> [  231.823955] tveeprom 5-0050: tuner model is Xceive XC5000C (idx 173, type 88)
> [  231.824782] tveeprom 5-0050: TV standards NTSC(M) ATSC/DVB Digital (eeprom 0x88)
> [  231.825272] tveeprom 5-0050: audio processor is AU8522 (idx 44)
> [  231.825276] tveeprom 5-0050: decoder processor is AU8522 (idx 42)
> [  231.825280] tveeprom 5-0050: has no radio, has IR receiver, has no IR transmitter
> [  231.825283] au0828: hauppauge_eeprom: hauppauge eeprom: model=72001
> [  231.857567] au8522 5-0047: creating new instance
> [  231.857879] au8522_decoder creating new instance...
> [  231.910525] tuner 5-0061: Setting mode_mask to 0x06
> [  231.910532] tuner 5-0061: tuner 0x61: Tuner type absent
> [  231.910535] tuner 5-0061: Tuner -1 found with type(s) Radio TV.
> [  231.911343] tuner 5-0061: Calling set_type_addr for type=88, addr=0x61, mode=0x04, config=ffffffffa0e64100
> [  231.911347] tuner 5-0061: defining GPIO callback
> [  231.934896] xc5000 5-0061: creating new instance
> [  231.954664] xc5000: Successfully identified at address 0x61
> [  231.954987] xc5000: Firmware has not been loaded previously
> [  231.955327] tuner 5-0061: type set to Xceive XC5000
> [  231.955330] tuner 5-0061: au0828 tuner I2C addr 0xc2 with type 88 used for 0x04
> [  233.622614] au8522 5-0047: attaching existing instance
> [  233.636061] xc5000 5-0061: attaching existing instance
> [  233.645775] xc5000: Successfully identified at address 0x61
> [  233.646034] xc5000: Firmware has not been loaded previously
> [  233.646461] DVB: registering new adapter (au0828)
> [  233.647575] usb 2-3.3: DVB: registering adapter 0 frontend 0 (Auvitek AU8522 QAM/8VSB Frontend)...
> [  233.648344] dvb_create_media_entity: media entity 'Auvitek AU8522 QAM/8VSB Frontend' registered.
> [  233.656105] dvb_create_media_entity: media entity 'dvb-demux' registered.
> [  234.166644] IR keymap rc-hauppauge not found
> [  234.166962] Registered IR keymap rc-empty
> [  234.178055] input: au0828 IR (Hauppauge HVR950Q) as /devices/pci0000:00/0000:00:14.0/usb2/2-3/2-3.3/rc/rc0/input14
> [  234.203311] rc rc0: au0828 IR (Hauppauge HVR950Q) as /devices/pci0000:00/0000:00:14.0/usb2/2-3/2-3.3/rc/rc0
> [  234.207828] au0828: Remote controller au0828 IR (Hauppauge HVR950Q) initalized
> [  234.208270] au0828: Registered device AU0828 [Hauppauge HVR950Q]
> [  234.212073] usbcore: registered new interface driver au0828
> [  234.230371] lirc_dev: IR Remote Control driver registered, major 243 
> [  234.257555] rc rc0: lirc_dev: driver ir-lirc-codec (au0828-input) registered at minor = 0
> [  234.257960] IR LIRC bridge handler initialized
> 
> (as au0828 is blacklisted, snd-usb-audio was probed first)
> 
> ====xxxx====
> 
> I'm also getting some other weird behavior when removing/reinserting
> the modules a few times. OK, officially we don't support it, but,
> as devices can be bind/unbind all the times, removing modules is
> a way to simulate such things. Also, I use it a lot while testing
> USB drivers ;)

I will do more testing on this. I didn't see any problems when
I fixed the graph patch above - I will do more on the latest
media branch today.

thanks,
-- Shuah

> 
> This one is after removing both the media drivers and snd-usb-audio, 
> and then modprobing snd-usb-audio:
> 
> [ 1839.510587] BUG: sleeping function called from invalid context at mm/slub.c:1289
> [ 1839.510881] in_atomic(): 1, irqs_disabled(): 0, pid: 3479, name: modprobe
> [ 1839.511157] 4 locks held by modprobe/3479:
> [ 1839.511415]  #0:  (&dev->mutex){......}, at: [<ffffffff81ce8933>] __driver_attach+0xa3/0x160
> [ 1839.512381]  #1:  (&dev->mutex){......}, at: [<ffffffff81ce8941>] __driver_attach+0xb1/0x160
> [ 1839.512388]  #2:  (register_mutex#5){+.+.+.}, at: [<ffffffffa10596c7>] usb_audio_probe+0x257/0x1c90 [snd_usb_audio]
> [ 1839.512401]  #3:  (&(&mdev->lock)->rlock){+.+.+.}, at: [<ffffffffa0e6051b>] media_device_register_entity+0x1cb/0x700 [media]
> [ 1839.512412] CPU: 2 PID: 3479 Comm: modprobe Not tainted 4.5.0-rc3+ #49
> [ 1839.512415] Hardware name:                  /NUC5i7RYB, BIOS RYBDWi35.86A.0350.2015.0812.1722 08/12/2015
> [ 1839.512417]  0000000000000000 ffff8803b3f6f288 ffffffff81933901 ffff8803c4bae000
> [ 1839.512422]  ffff8803c4bae5c8 ffff8803b3f6f2b0 ffffffff811c6af5 ffff8803c4bae000
> [ 1839.512427]  ffffffff8285d7f6 0000000000000509 ffff8803b3f6f2f0 ffffffff811c6ce5
> [ 1839.512432] Call Trace:
> [ 1839.512436]  [<ffffffff81933901>] dump_stack+0x85/0xc4
> [ 1839.512440]  [<ffffffff811c6af5>] ___might_sleep+0x245/0x3a0
> [ 1839.512443]  [<ffffffff811c6ce5>] __might_sleep+0x95/0x1a0
> [ 1839.512446]  [<ffffffff8155aade>] kmem_cache_alloc_trace+0x20e/0x300
> [ 1839.512451]  [<ffffffffa0e66e3d>] ? media_add_link+0x4d/0x140 [media]
> [ 1839.512455]  [<ffffffffa0e66e3d>] media_add_link+0x4d/0x140 [media]
> [ 1839.512459]  [<ffffffffa0e69931>] media_create_pad_link+0xa1/0x600 [media]
> [ 1839.512463]  [<ffffffffa0fe11b3>] au0828_media_graph_notify+0x173/0x360 [au0828]
> [ 1839.512467]  [<ffffffffa0e68a6a>] ? media_gobj_create+0x1ba/0x480 [media]
> [ 1839.512471]  [<ffffffffa0e606fb>] media_device_register_entity+0x3ab/0x700 [media]
> [ 1839.512475]  [<ffffffffa0e60350>] ? media_device_release_devres+0x10/0x10 [media]
> [ 1839.512479]  [<ffffffff8124068d>] ? trace_hardirqs_on+0xd/0x10
> [ 1839.512482]  [<ffffffff8159f992>] ? create_object+0x5a2/0x940
> [ 1839.512485]  [<ffffffff8159f3f0>] ? kmemleak_disable+0x90/0x90
> [ 1839.512489]  [<ffffffff8159f3f0>] ? kmemleak_disable+0x90/0x90
> [ 1839.512497]  [<ffffffffa109722c>] ? media_snd_mixer_init+0x16c/0x500 [snd_usb_audio]
> [ 1839.512501]  [<ffffffff8155f696>] ? kasan_unpoison_shadow+0x36/0x50
> [ 1839.512504]  [<ffffffff8155f70e>] ? kasan_kmalloc+0x5e/0x70
> [ 1839.512511]  [<ffffffffa1097352>] media_snd_mixer_init+0x292/0x500 [snd_usb_audio]
> [ 1839.512519]  [<ffffffffa1097693>] media_snd_device_create+0xd3/0x2e0 [snd_usb_audio]
> [ 1839.512525]  [<ffffffffa1059b05>] usb_audio_probe+0x695/0x1c90 [snd_usb_audio]
> [ 1839.512529]  [<ffffffff8128b217>] ? debug_lockdep_rcu_enabled+0x77/0x90
> [ 1839.512536]  [<ffffffffa1059470>] ? snd_usb_create_stream+0x4e0/0x4e0 [snd_usb_audio]
> [ 1839.512540]  [<ffffffff812404fa>] ? trace_hardirqs_on_caller+0x40a/0x590
> [ 1839.512543]  [<ffffffff8124068d>] ? trace_hardirqs_on+0xd/0x10
> [ 1839.512546]  [<ffffffff81dbb1dd>] usb_probe_interface+0x45d/0x940
> [ 1839.512550]  [<ffffffff81ce7e7a>] driver_probe_device+0x21a/0xc30
> [ 1839.512553]  [<ffffffff81ce8890>] ? driver_probe_device+0xc30/0xc30
> [ 1839.512556]  [<ffffffff81ce89b1>] __driver_attach+0x121/0x160
> [ 1839.512559]  [<ffffffff81ce21bf>] bus_for_each_dev+0x11f/0x1a0
> [ 1839.512562]  [<ffffffff81ce20a0>] ? subsys_dev_iter_exit+0x10/0x10
> [ 1839.512565]  [<ffffffff822e7be7>] ? _raw_spin_unlock+0x27/0x40
> [ 1839.512568]  [<ffffffff81ce6cdd>] driver_attach+0x3d/0x50
> [ 1839.512571]  [<ffffffff81ce5df9>] bus_add_driver+0x4c9/0x770
> [ 1839.512575]  [<ffffffff81cea39c>] driver_register+0x18c/0x3b0
> [ 1839.512578]  [<ffffffff81250f92>] ? __raw_spin_lock_init+0x32/0x100
> [ 1839.512581]  [<ffffffff81db6e98>] usb_register_driver+0x1f8/0x440
> [ 1839.512584]  [<ffffffff810021a1>] ? do_one_initcall+0x131/0x300
> [ 1839.512587]  [<ffffffffa10d8000>] ? 0xffffffffa10d8000
> [ 1839.512594]  [<ffffffffa10d801e>] usb_audio_driver_init+0x1e/0x1000 [snd_usb_audio]
> [ 1839.512597]  [<ffffffff810021b1>] do_one_initcall+0x141/0x300
> [ 1839.512600]  [<ffffffff81002070>] ? try_to_run_init_process+0x40/0x40
> [ 1839.512603]  [<ffffffff8155f696>] ? kasan_unpoison_shadow+0x36/0x50
> [ 1839.512606]  [<ffffffff8155f696>] ? kasan_unpoison_shadow+0x36/0x50
> [ 1839.512609]  [<ffffffff8155f7a7>] ? __asan_register_globals+0x87/0xa0
> [ 1839.512613]  [<ffffffff8144d8eb>] do_init_module+0x1d0/0x5ad
> [ 1839.512617]  [<ffffffff812f27b6>] load_module+0x6666/0x9ba0
> [ 1839.512620]  [<ffffffff812e9e20>] ? symbol_put_addr+0x50/0x50
> [ 1839.512627]  [<ffffffff812ec150>] ? module_frob_arch_sections+0x20/0x20
> [ 1839.512630]  [<ffffffff815be010>] ? open_exec+0x50/0x50
> [ 1839.512634]  [<ffffffff8116716b>] ? ns_capable+0x5b/0xd0
> [ 1839.512637]  [<ffffffff812f5fe8>] SyS_finit_module+0x108/0x130
> [ 1839.512640]  [<ffffffff812f5ee0>] ? SyS_init_module+0x1f0/0x1f0
> [ 1839.512643]  [<ffffffff81004044>] ? lockdep_sys_exit_thunk+0x12/0x14
> [ 1839.512647]  [<ffffffff822e8636>] entry_SYSCALL_64_fastpath+0x16/0x76
> [ 1839.514027] usbcore: registered new interface driver snd-usb-audio
> 
> And an extra kmemleak issue:
> 
> unreferenced object 0xffff8803bcbacf30 (size 128):
>   comm "modprobe", pid 2625, jiffies 4294950803 (age 1715.788s)
>   hex dump (first 32 bytes):
>     00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
>     00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
>   backtrace:
>     [<ffffffff822cf76e>] kmemleak_alloc+0x4e/0xb0
>     [<ffffffff8155ab06>] kmem_cache_alloc_trace+0x236/0x300
>     [<ffffffff81937eb2>] ida_pre_get+0x1a2/0x250
>     [<ffffffff81938005>] ida_simple_get+0xa5/0x170
>     [<ffffffffa0f1469a>] 0xffffffffa0f1469a
>     [<ffffffffa0fe6951>] au0828_dvb_register+0xe21/0x12c0 [au0828]
>     [<ffffffffa0fd1d92>] vb2_thread+0xbb2/0xc10 [videobuf2_core]
>     [<ffffffff81dbb1dd>] usb_probe_interface+0x45d/0x940
>     [<ffffffff81ce7e7a>] driver_probe_device+0x21a/0xc30
>     [<ffffffff81ce89b1>] __driver_attach+0x121/0x160
>     [<ffffffff81ce21bf>] bus_for_each_dev+0x11f/0x1a0
>     [<ffffffff81ce6cdd>] driver_attach+0x3d/0x50
>     [<ffffffff81ce5df9>] bus_add_driver+0x4c9/0x770
>     [<ffffffff81cea39c>] driver_register+0x18c/0x3b0
>     [<ffffffff81db6e98>] usb_register_driver+0x1f8/0x440
>     [<ffffffffa0fb80b7>] au8522_get_frontend+0xb7/0x160 [au8522_dig]
> 
> I did one extra module removal/reinsert, and this time, I got lots
> of bugs. Full dmesg is at:
> 	http://pastebin.com/LEX5LD5K
> 
> The Kernel hanged after a while with (at the serial console - with
> kind of explain why the message is truncated):
> 
> [ 2117.282666] kasan: GPF could be caused by NULL-ptr deref or user memory accessgeneral protection fault: 0000 [#2] SMP KASAN
> [ 2117.284074] Modules linked in: videobuf2_v4l2 videobuf2_core tveeprom dvb_core rc_core v4l2_common videodev snd_usb_audio media cpufreq_powersave cpufreq_conservative cpufreq_userspace cpufreq_stats parport_pc ppdev lp parport snd_hda_codec_hdmi intel_rapl x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel iTCO_wdt iTCO_vendor_support kvm irqbypass crct10dif_pclmul crc32_
> 
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-media" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 


-- 
Shuah Khan
Sr. Linux Kernel Developer
Open Source Innovation Group
Samsung Research America (Silicon Valley)
shuahkh@osg.samsung.com | (970) 217-8978

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Shuah Khan]

On 03/07/2016 07:16 AM, Shuah Khan wrote:
> On 03/05/2016 03:00 AM, Mauro Carvalho Chehab wrote:
>> Em Wed,  2 Mar 2016 09:50:31 -0700
>> Shuah Khan <shuahkh@osg.samsung.com> escreveu:
>>
>>> Change ALSA driver to use Media Controller API to
>>> share media resources with DVB and V4L2 drivers
>>> on a AU0828 media device. Media Controller specific
>>> initialization is done after sound card is registered.
>>> ALSA creates Media interface and entity function graph
>>> nodes for Control, Mixer, PCM Playback, and PCM Capture
>>> devices.
>>>
>>> snd_usb_hw_params() will call Media Controller enable
>>> source handler interface to request the media resource.
>>> If resource request is granted, it will release it from
>>> snd_usb_hw_free(). If resource is busy, -EBUSY is returned.
>>>
>>> Media specific cleanup is done in usb_audio_disconnect().
>>>
>>> Signed-off-by: Shuah Khan <shuahkh@osg.samsung.com>
>>> Acked-by: Takashi Iwai <tiwai@suse.de>
>>
>> Shuah, by looking at the produced graphs:
>> 	https://mchehab.fedorapeople.org/mc-next-gen/au0828_test/
>>
>> I'm noticing the lack of ALSA I/O entities in the diagram
>> (MEDIA_ENT_F_AUDIO_CAPTURE). The mixer there is not connected.
>>
>> Could you please check what's happening?
> 
> Mauro,
> 
> Did you pull in this patch that fixes the graph problem:
> 
> https://www.mail-archive.com/linux-media@vger.kernel.org/msg95047.html
> 
> If you haven't, could you please pull this in.
> 

You do have the above patch in the latest linux_media
git. I couldn't reproduce the problem you are seeing.

Here is what I did.
- Blacklisted au0828 and forcing snd_usb_audio
  probe first. Then I generated a graph that shows just
  the audio nodes.
- And then did a modprobe and generated full graph.

https://drive.google.com/folderview?id=0B0NIL0BQg-AlejFpR19Cb1RGYVk&usp=drive_web

I see that in you graph, mixer gets connected to decoder, but the rest
of the audio nodes are missing from the graph.

MEDIA_ENT_F_AUDIO_CAPTURE gets created from  snd_usb_pcm_open()
and gets deleted from free_substream(). Is it possible, for
some reason, either snd_usb_pcm_open() fails or free_substream()
is called?


thanks,
-- Shuah


-- 
Shuah Khan
Sr. Linux Kernel Developer
Open Source Innovation Group
Samsung Research America (Silicon Valley)
shuahkh@osg.samsung.com | (970) 217-8978

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Shuah Khan]

On 03/07/2016 07:16 AM, Shuah Khan wrote:
> On 03/05/2016 03:00 AM, Mauro Carvalho Chehab wrote:
>> Em Wed,  2 Mar 2016 09:50:31 -0700
>> Shuah Khan <shuahkh@osg.samsung.com> escreveu:
>>
>>> Change ALSA driver to use Media Controller API to
>>> share media resources with DVB and V4L2 drivers
>>> on a AU0828 media device. Media Controller specific
>>> initialization is done after sound card is registered.
>>> ALSA creates Media interface and entity function graph
>>> nodes for Control, Mixer, PCM Playback, and PCM Capture
>>> devices.
>>>
>>> snd_usb_hw_params() will call Media Controller enable
>>> source handler interface to request the media resource.
>>> If resource request is granted, it will release it from
>>> snd_usb_hw_free(). If resource is busy, -EBUSY is returned.
>>>
>>> Media specific cleanup is done in usb_audio_disconnect().
>>>
>>> Signed-off-by: Shuah Khan <shuahkh@osg.samsung.com>
>>> Acked-by: Takashi Iwai <tiwai@suse.de>
>>
>> Shuah, by looking at the produced graphs:
>> 	https://mchehab.fedorapeople.org/mc-next-gen/au0828_test/
>>
>> I'm noticing the lack of ALSA I/O entities in the diagram
>> (MEDIA_ENT_F_AUDIO_CAPTURE). The mixer there is not connected.
>>
>> Could you please check what's happening?
> 
> Mauro,
> 
> Did you pull in this patch that fixes the graph problem:
> 
> https://www.mail-archive.com/linux-media@vger.kernel.org/msg95047.html
> 
> If you haven't, could you please pull this in.
> 

You do have the above patch in the latest linux_media
git. I couldn't reproduce the problem you are seeing.

Here is what I did.
- Blacklisted au0828 and forcing snd_usb_audio
  probe first. Then I generated a graph that shows just
  the audio nodes.
- And then did a modprobe and generated full graph.

https://drive.google.com/folderview?id=0B0NIL0BQg-AlejFpR19Cb1RGYVk&usp=drive_web

I see that in you graph, mixer gets connected to decoder, but the rest
of the audio nodes are missing from the graph.

MEDIA_ENT_F_AUDIO_CAPTURE gets created from  snd_usb_pcm_open()
and gets deleted from free_substream(). Is it possible, for
some reason, either snd_usb_pcm_open() fails or free_substream()
is called?


thanks,
-- Shuah


-- 
Shuah Khan
Sr. Linux Kernel Developer
Open Source Innovation Group
Samsung Research America (Silicon Valley)
shuahkh@osg.samsung.com | (970) 217-8978

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Shuah Khan]

On 03/05/2016 03:00 AM, Mauro Carvalho Chehab wrote:
> Em Wed,  2 Mar 2016 09:50:31 -0700
> Shuah Khan <shuahkh@osg.samsung.com> escreveu:
> 
>> Change ALSA driver to use Media Controller API to
>> share media resources with DVB and V4L2 drivers
>> on a AU0828 media device. Media Controller specific
>> initialization is done after sound card is registered.
>> ALSA creates Media interface and entity function graph
>> nodes for Control, Mixer, PCM Playback, and PCM Capture
>> devices.
>>
>> snd_usb_hw_params() will call Media Controller enable
>> source handler interface to request the media resource.
>> If resource request is granted, it will release it from
>> snd_usb_hw_free(). If resource is busy, -EBUSY is returned.
>>
>> Media specific cleanup is done in usb_audio_disconnect().
>>
>> Signed-off-by: Shuah Khan <shuahkh@osg.samsung.com>
>> Acked-by: Takashi Iwai <tiwai@suse.de>
> 
> Shuah, by looking at the produced graphs:
> 	https://mchehab.fedorapeople.org/mc-next-gen/au0828_test/
> 
> I'm noticing the lack of ALSA I/O entities in the diagram
> (MEDIA_ENT_F_AUDIO_CAPTURE). The mixer there is not connected.
> 
> Could you please check what's happening?
> 
> Those are the relevant dmesg data:
> 
> [   19.017276] usbcore: registered new interface driver snd-usb-audio
> [  230.706102] Linux video capture interface: v2.00
> [  230.856983] au0828: au0828 driver loaded
> [  231.230612] au0828: i2c bus registered
> [  231.822006] tveeprom 5-0050: Hauppauge model 72001, rev E1H3, serial# 4035199481
> [  231.822991] tveeprom 5-0050: MAC address is 00:0d:fe:84:41:f9
> [  231.823955] tveeprom 5-0050: tuner model is Xceive XC5000C (idx 173, type 88)
> [  231.824782] tveeprom 5-0050: TV standards NTSC(M) ATSC/DVB Digital (eeprom 0x88)
> [  231.825272] tveeprom 5-0050: audio processor is AU8522 (idx 44)
> [  231.825276] tveeprom 5-0050: decoder processor is AU8522 (idx 42)
> [  231.825280] tveeprom 5-0050: has no radio, has IR receiver, has no IR transmitter
> [  231.825283] au0828: hauppauge_eeprom: hauppauge eeprom: model=72001
> [  231.857567] au8522 5-0047: creating new instance
> [  231.857879] au8522_decoder creating new instance...
> [  231.910525] tuner 5-0061: Setting mode_mask to 0x06
> [  231.910532] tuner 5-0061: tuner 0x61: Tuner type absent
> [  231.910535] tuner 5-0061: Tuner -1 found with type(s) Radio TV.
> [  231.911343] tuner 5-0061: Calling set_type_addr for type=88, addr=0x61, mode=0x04, config=ffffffffa0e64100
> [  231.911347] tuner 5-0061: defining GPIO callback
> [  231.934896] xc5000 5-0061: creating new instance
> [  231.954664] xc5000: Successfully identified at address 0x61
> [  231.954987] xc5000: Firmware has not been loaded previously
> [  231.955327] tuner 5-0061: type set to Xceive XC5000
> [  231.955330] tuner 5-0061: au0828 tuner I2C addr 0xc2 with type 88 used for 0x04
> [  233.622614] au8522 5-0047: attaching existing instance
> [  233.636061] xc5000 5-0061: attaching existing instance
> [  233.645775] xc5000: Successfully identified at address 0x61
> [  233.646034] xc5000: Firmware has not been loaded previously
> [  233.646461] DVB: registering new adapter (au0828)
> [  233.647575] usb 2-3.3: DVB: registering adapter 0 frontend 0 (Auvitek AU8522 QAM/8VSB Frontend)...
> [  233.648344] dvb_create_media_entity: media entity 'Auvitek AU8522 QAM/8VSB Frontend' registered.
> [  233.656105] dvb_create_media_entity: media entity 'dvb-demux' registered.
> [  234.166644] IR keymap rc-hauppauge not found
> [  234.166962] Registered IR keymap rc-empty
> [  234.178055] input: au0828 IR (Hauppauge HVR950Q) as /devices/pci0000:00/0000:00:14.0/usb2/2-3/2-3.3/rc/rc0/input14
> [  234.203311] rc rc0: au0828 IR (Hauppauge HVR950Q) as /devices/pci0000:00/0000:00:14.0/usb2/2-3/2-3.3/rc/rc0
> [  234.207828] au0828: Remote controller au0828 IR (Hauppauge HVR950Q) initalized
> [  234.208270] au0828: Registered device AU0828 [Hauppauge HVR950Q]
> [  234.212073] usbcore: registered new interface driver au0828
> [  234.230371] lirc_dev: IR Remote Control driver registered, major 243 
> [  234.257555] rc rc0: lirc_dev: driver ir-lirc-codec (au0828-input) registered at minor = 0
> [  234.257960] IR LIRC bridge handler initialized
> 
> (as au0828 is blacklisted, snd-usb-audio was probed first)
> 
> ====xxxx====
> 
> I'm also getting some other weird behavior when removing/reinserting
> the modules a few times. OK, officially we don't support it, but,
> as devices can be bind/unbind all the times, removing modules is
> a way to simulate such things. Also, I use it a lot while testing
> USB drivers ;)
> 
> This one is after removing both the media drivers and snd-usb-audio, 
> and then modprobing snd-usb-audio:
> 

I did see some issues when I did the following sequence:

- blacklisted au0828 and snd-usb-audio was probed first
  graph is good just with audio entities as expected
- modprobed au0828 - graph looks good.
- rmmod au0828 - no problems seen in dmesg
- modprobe au0828 - problems kasan reports bad access etc.
  http://pastebin.com/FFqNzx9G

Here is what's going on after each step:

blacklisted au0828 and snd-usb-audio was probed first
1. snd-usb-audio creates media device and registers it
   Creates its graph etc.
2. modprobed au0828
   au0828 finds the media device created and registered.
   Adds its graph
3. rmmod au0828
   Even though there are no problems reported, at this
   time media device is unregistered, and /dev/mediaX is
   removed. We still have snd_usb-audio and media device.
   As media device is a device resource on usb device parent,
   it will still be there, but no way to access the device
   itself, because it is no longer registered.
4. modprobe au0828
   At this point, au0828 finds the media device as it still
   there, registers it and adds its graph. No audio graph
   present at this time.

Please note that the media device will not be deleted until
the last put on the parent usb struct device. So even when
both snd-usb-audio, and au0828 modules are removed, media
device is still there without its graph and associated devnode
(/dev/mediax is removed).

This isn't bad, however, media_device could still have
stale information.

e.g: enable/disable handlers - when au0828 is removed,
these are no longer valid. Could be cleaned up in
media device unregister just like entity_notify() handler
gets deleted from media device unregister. At the moment,
either driver can call unregister and same cleanup happens.

I will send a patch to do enable/disable hanlder cleanup
in unregister path.

However, the root of the problem is media device is
still there without its graph and associated devnode
(/dev/mediax is removed) when any one of the drivers
is removed. This leaves the remaining drivers in a
degenerate state.

The problem can be solved with some handshaking at
unregister time. We could add a callback for each
if the drivers to handle media device unregister.
However, that would add delays in device removal path
when all the drivers exit. I think it will be hard to
handle all the corner cases without adding run-time overhead.

Any thoughts on whether we want to unofficially support
being able to remove individual drivers?

thanks,
-- Shuah

-- 
Shuah Khan
Sr. Linux Kernel Developer
Open Source Innovation Group
Samsung Research America (Silicon Valley)
shuahkh@osg.samsung.com | (970) 217-8978

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Shuah Khan]

On 03/05/2016 03:00 AM, Mauro Carvalho Chehab wrote:
> Em Wed,  2 Mar 2016 09:50:31 -0700
> Shuah Khan <shuahkh@osg.samsung.com> escreveu:
> 
>> Change ALSA driver to use Media Controller API to
>> share media resources with DVB and V4L2 drivers
>> on a AU0828 media device. Media Controller specific
>> initialization is done after sound card is registered.
>> ALSA creates Media interface and entity function graph
>> nodes for Control, Mixer, PCM Playback, and PCM Capture
>> devices.
>>
>> snd_usb_hw_params() will call Media Controller enable
>> source handler interface to request the media resource.
>> If resource request is granted, it will release it from
>> snd_usb_hw_free(). If resource is busy, -EBUSY is returned.
>>
>> Media specific cleanup is done in usb_audio_disconnect().
>>
>> Signed-off-by: Shuah Khan <shuahkh@osg.samsung.com>
>> Acked-by: Takashi Iwai <tiwai@suse.de>
> 
> Shuah, by looking at the produced graphs:
> 	https://mchehab.fedorapeople.org/mc-next-gen/au0828_test/
> 
> I'm noticing the lack of ALSA I/O entities in the diagram
> (MEDIA_ENT_F_AUDIO_CAPTURE). The mixer there is not connected.
> 
> Could you please check what's happening?
> 
> Those are the relevant dmesg data:
> 
> [   19.017276] usbcore: registered new interface driver snd-usb-audio
> [  230.706102] Linux video capture interface: v2.00
> [  230.856983] au0828: au0828 driver loaded
> [  231.230612] au0828: i2c bus registered
> [  231.822006] tveeprom 5-0050: Hauppauge model 72001, rev E1H3, serial# 4035199481
> [  231.822991] tveeprom 5-0050: MAC address is 00:0d:fe:84:41:f9
> [  231.823955] tveeprom 5-0050: tuner model is Xceive XC5000C (idx 173, type 88)
> [  231.824782] tveeprom 5-0050: TV standards NTSC(M) ATSC/DVB Digital (eeprom 0x88)
> [  231.825272] tveeprom 5-0050: audio processor is AU8522 (idx 44)
> [  231.825276] tveeprom 5-0050: decoder processor is AU8522 (idx 42)
> [  231.825280] tveeprom 5-0050: has no radio, has IR receiver, has no IR transmitter
> [  231.825283] au0828: hauppauge_eeprom: hauppauge eeprom: model=72001
> [  231.857567] au8522 5-0047: creating new instance
> [  231.857879] au8522_decoder creating new instance...
> [  231.910525] tuner 5-0061: Setting mode_mask to 0x06
> [  231.910532] tuner 5-0061: tuner 0x61: Tuner type absent
> [  231.910535] tuner 5-0061: Tuner -1 found with type(s) Radio TV.
> [  231.911343] tuner 5-0061: Calling set_type_addr for type=88, addr=0x61, mode=0x04, config=ffffffffa0e64100
> [  231.911347] tuner 5-0061: defining GPIO callback
> [  231.934896] xc5000 5-0061: creating new instance
> [  231.954664] xc5000: Successfully identified at address 0x61
> [  231.954987] xc5000: Firmware has not been loaded previously
> [  231.955327] tuner 5-0061: type set to Xceive XC5000
> [  231.955330] tuner 5-0061: au0828 tuner I2C addr 0xc2 with type 88 used for 0x04
> [  233.622614] au8522 5-0047: attaching existing instance
> [  233.636061] xc5000 5-0061: attaching existing instance
> [  233.645775] xc5000: Successfully identified at address 0x61
> [  233.646034] xc5000: Firmware has not been loaded previously
> [  233.646461] DVB: registering new adapter (au0828)
> [  233.647575] usb 2-3.3: DVB: registering adapter 0 frontend 0 (Auvitek AU8522 QAM/8VSB Frontend)...
> [  233.648344] dvb_create_media_entity: media entity 'Auvitek AU8522 QAM/8VSB Frontend' registered.
> [  233.656105] dvb_create_media_entity: media entity 'dvb-demux' registered.
> [  234.166644] IR keymap rc-hauppauge not found
> [  234.166962] Registered IR keymap rc-empty
> [  234.178055] input: au0828 IR (Hauppauge HVR950Q) as /devices/pci0000:00/0000:00:14.0/usb2/2-3/2-3.3/rc/rc0/input14
> [  234.203311] rc rc0: au0828 IR (Hauppauge HVR950Q) as /devices/pci0000:00/0000:00:14.0/usb2/2-3/2-3.3/rc/rc0
> [  234.207828] au0828: Remote controller au0828 IR (Hauppauge HVR950Q) initalized
> [  234.208270] au0828: Registered device AU0828 [Hauppauge HVR950Q]
> [  234.212073] usbcore: registered new interface driver au0828
> [  234.230371] lirc_dev: IR Remote Control driver registered, major 243 
> [  234.257555] rc rc0: lirc_dev: driver ir-lirc-codec (au0828-input) registered at minor = 0
> [  234.257960] IR LIRC bridge handler initialized
> 
> (as au0828 is blacklisted, snd-usb-audio was probed first)
> 
> ====xxxx====
> 
> I'm also getting some other weird behavior when removing/reinserting
> the modules a few times. OK, officially we don't support it, but,
> as devices can be bind/unbind all the times, removing modules is
> a way to simulate such things. Also, I use it a lot while testing
> USB drivers ;)
> 
> This one is after removing both the media drivers and snd-usb-audio, 
> and then modprobing snd-usb-audio:
> 

I did see some issues when I did the following sequence:

- blacklisted au0828 and snd-usb-audio was probed first
  graph is good just with audio entities as expected
- modprobed au0828 - graph looks good.
- rmmod au0828 - no problems seen in dmesg
- modprobe au0828 - problems kasan reports bad access etc.
  http://pastebin.com/FFqNzx9G

Here is what's going on after each step:

blacklisted au0828 and snd-usb-audio was probed first
1. snd-usb-audio creates media device and registers it
   Creates its graph etc.
2. modprobed au0828
   au0828 finds the media device created and registered.
   Adds its graph
3. rmmod au0828
   Even though there are no problems reported, at this
   time media device is unregistered, and /dev/mediaX is
   removed. We still have snd_usb-audio and media device.
   As media device is a device resource on usb device parent,
   it will still be there, but no way to access the device
   itself, because it is no longer registered.
4. modprobe au0828
   At this point, au0828 finds the media device as it still
   there, registers it and adds its graph. No audio graph
   present at this time.

Please note that the media device will not be deleted until
the last put on the parent usb struct device. So even when
both snd-usb-audio, and au0828 modules are removed, media
device is still there without its graph and associated devnode
(/dev/mediax is removed).

This isn't bad, however, media_device could still have
stale information.

e.g: enable/disable handlers - when au0828 is removed,
these are no longer valid. Could be cleaned up in
media device unregister just like entity_notify() handler
gets deleted from media device unregister. At the moment,
either driver can call unregister and same cleanup happens.

I will send a patch to do enable/disable hanlder cleanup
in unregister path.

However, the root of the problem is media device is
still there without its graph and associated devnode
(/dev/mediax is removed) when any one of the drivers
is removed. This leaves the remaining drivers in a
degenerate state.

The problem can be solved with some handshaking at
unregister time. We could add a callback for each
if the drivers to handle media device unregister.
However, that would add delays in device removal path
when all the drivers exit. I think it will be hard to
handle all the corner cases without adding run-time overhead.

Any thoughts on whether we want to unofficially support
being able to remove individual drivers?

thanks,
-- Shuah

-- 
Shuah Khan
Sr. Linux Kernel Developer
Open Source Innovation Group
Samsung Research America (Silicon Valley)
shuahkh@osg.samsung.com | (970) 217-8978

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Shuah Khan]

On 03/05/2016 03:00 AM, Mauro Carvalho Chehab wrote:
> Em Wed,  2 Mar 2016 09:50:31 -0700
> Shuah Khan <shuahkh@osg.samsung.com> escreveu:
> 
>> Change ALSA driver to use Media Controller API to
>> share media resources with DVB and V4L2 drivers
>> on a AU0828 media device. Media Controller specific
>> initialization is done after sound card is registered.
>> ALSA creates Media interface and entity function graph
>> nodes for Control, Mixer, PCM Playback, and PCM Capture
>> devices.
>>
>> snd_usb_hw_params() will call Media Controller enable
>> source handler interface to request the media resource.
>> If resource request is granted, it will release it from
>> snd_usb_hw_free(). If resource is busy, -EBUSY is returned.
>>
>> Media specific cleanup is done in usb_audio_disconnect().
>>
>> Signed-off-by: Shuah Khan <shuahkh@osg.samsung.com>
>> Acked-by: Takashi Iwai <tiwai@suse.de>
> 
> Shuah, by looking at the produced graphs:
> 	https://mchehab.fedorapeople.org/mc-next-gen/au0828_test/
> 
> I'm noticing the lack of ALSA I/O entities in the diagram
> (MEDIA_ENT_F_AUDIO_CAPTURE). The mixer there is not connected.
> 
> Could you please check what's happening?
> 
> Those are the relevant dmesg data:
> 
> [   19.017276] usbcore: registered new interface driver snd-usb-audio
> [  230.706102] Linux video capture interface: v2.00
> [  230.856983] au0828: au0828 driver loaded
> [  231.230612] au0828: i2c bus registered
> [  231.822006] tveeprom 5-0050: Hauppauge model 72001, rev E1H3, serial# 4035199481
> [  231.822991] tveeprom 5-0050: MAC address is 00:0d:fe:84:41:f9
> [  231.823955] tveeprom 5-0050: tuner model is Xceive XC5000C (idx 173, type 88)
> [  231.824782] tveeprom 5-0050: TV standards NTSC(M) ATSC/DVB Digital (eeprom 0x88)
> [  231.825272] tveeprom 5-0050: audio processor is AU8522 (idx 44)
> [  231.825276] tveeprom 5-0050: decoder processor is AU8522 (idx 42)
> [  231.825280] tveeprom 5-0050: has no radio, has IR receiver, has no IR transmitter
> [  231.825283] au0828: hauppauge_eeprom: hauppauge eeprom: model=72001
> [  231.857567] au8522 5-0047: creating new instance
> [  231.857879] au8522_decoder creating new instance...
> [  231.910525] tuner 5-0061: Setting mode_mask to 0x06
> [  231.910532] tuner 5-0061: tuner 0x61: Tuner type absent
> [  231.910535] tuner 5-0061: Tuner -1 found with type(s) Radio TV.
> [  231.911343] tuner 5-0061: Calling set_type_addr for type=88, addr=0x61, mode=0x04, config=ffffffffa0e64100
> [  231.911347] tuner 5-0061: defining GPIO callback
> [  231.934896] xc5000 5-0061: creating new instance
> [  231.954664] xc5000: Successfully identified at address 0x61
> [  231.954987] xc5000: Firmware has not been loaded previously
> [  231.955327] tuner 5-0061: type set to Xceive XC5000
> [  231.955330] tuner 5-0061: au0828 tuner I2C addr 0xc2 with type 88 used for 0x04
> [  233.622614] au8522 5-0047: attaching existing instance
> [  233.636061] xc5000 5-0061: attaching existing instance
> [  233.645775] xc5000: Successfully identified at address 0x61
> [  233.646034] xc5000: Firmware has not been loaded previously
> [  233.646461] DVB: registering new adapter (au0828)
> [  233.647575] usb 2-3.3: DVB: registering adapter 0 frontend 0 (Auvitek AU8522 QAM/8VSB Frontend)...
> [  233.648344] dvb_create_media_entity: media entity 'Auvitek AU8522 QAM/8VSB Frontend' registered.
> [  233.656105] dvb_create_media_entity: media entity 'dvb-demux' registered.
> [  234.166644] IR keymap rc-hauppauge not found
> [  234.166962] Registered IR keymap rc-empty
> [  234.178055] input: au0828 IR (Hauppauge HVR950Q) as /devices/pci0000:00/0000:00:14.0/usb2/2-3/2-3.3/rc/rc0/input14
> [  234.203311] rc rc0: au0828 IR (Hauppauge HVR950Q) as /devices/pci0000:00/0000:00:14.0/usb2/2-3/2-3.3/rc/rc0
> [  234.207828] au0828: Remote controller au0828 IR (Hauppauge HVR950Q) initalized
> [  234.208270] au0828: Registered device AU0828 [Hauppauge HVR950Q]
> [  234.212073] usbcore: registered new interface driver au0828
> [  234.230371] lirc_dev: IR Remote Control driver registered, major 243 
> [  234.257555] rc rc0: lirc_dev: driver ir-lirc-codec (au0828-input) registered at minor = 0
> [  234.257960] IR LIRC bridge handler initialized
> 
> (as au0828 is blacklisted, snd-usb-audio was probed first)
> 
> ====xxxx====
> 
> I'm also getting some other weird behavior when removing/reinserting
> the modules a few times. OK, officially we don't support it, but,
> as devices can be bind/unbind all the times, removing modules is
> a way to simulate such things. Also, I use it a lot while testing
> USB drivers ;)
> 
> This one is after removing both the media drivers and snd-usb-audio, 
> and then modprobing snd-usb-audio:
> 

I did see some issues when I did the following sequence:

- blacklisted au0828 and snd-usb-audio was probed first
  graph is good just with audio entities as expected
- modprobed au0828 - graph looks good.
- rmmod au0828 - no problems seen in dmesg
- modprobe au0828 - problems kasan reports bad access etc.
  http://pastebin.com/FFqNzx9G

Here is what's going on after each step:

blacklisted au0828 and snd-usb-audio was probed first
1. snd-usb-audio creates media device and registers it
   Creates its graph etc.
2. modprobed au0828
   au0828 finds the media device created and registered.
   Adds its graph
3. rmmod au0828
   Even though there are no problems reported, at this
   time media device is unregistered, and /dev/mediaX is
   removed. We still have snd_usb-audio and media device.
   As media device is a device resource on usb device parent,
   it will still be there, but no way to access the device
   itself, because it is no longer registered.
4. modprobe au0828
   At this point, au0828 finds the media device as it still
   there, registers it and adds its graph. No audio graph
   present at this time.

Please note that the media device will not be deleted until
the last put on the parent usb struct device. So even when
both snd-usb-audio, and au0828 modules are removed, media
device is still there without its graph and associated devnode
(/dev/mediax is removed).

This isn't bad, however, media_device could still have
stale information.

e.g: enable/disable handlers - when au0828 is removed,
these are no longer valid. Could be cleaned up in
media device unregister just like entity_notify() handler
gets deleted from media device unregister. At the moment,
either driver can call unregister and same cleanup happens.

I will send a patch to do enable/disable handler cleanup
in unregister path.

However, the root of the problem is media device is
still there without its graph and associated devnode
(/dev/mediax is removed) when any one of the drivers
is removed. This leaves the remaining drivers in a
degenerate state.

The problem can be solved with some handshaking at
unregister time. We could add a callback for each
if the drivers to handle media device unregister.
However, that would add delays in device removal path
when all the drivers exit. I think it will be hard to
handle all the corner cases without adding run-time overhead.

Any thoughts on whether we want to unofficially support
being able to remove individual drivers?

thanks,
-- Shuah

-- 
Shuah Khan
Sr. Linux Kernel Developer
Open Source Innovation Group
Samsung Research America (Silicon Valley)
shuahkh@osg.samsung.com | (970) 217-8978

Re: [PATCH v5 22/22] sound/usb: Use Media Controller API to share media resources

[Shuah Khan]

On 03/05/2016 03:00 AM, Mauro Carvalho Chehab wrote:
> Em Wed,  2 Mar 2016 09:50:31 -0700
> Shuah Khan <shuahkh@osg.samsung.com> escreveu:
> 
>> Change ALSA driver to use Media Controller API to
>> share media resources with DVB and V4L2 drivers
>> on a AU0828 media device. Media Controller specific
>> initialization is done after sound card is registered.
>> ALSA creates Media interface and entity function graph
>> nodes for Control, Mixer, PCM Playback, and PCM Capture
>> devices.
>>
>> snd_usb_hw_params() will call Media Controller enable
>> source handler interface to request the media resource.
>> If resource request is granted, it will release it from
>> snd_usb_hw_free(). If resource is busy, -EBUSY is returned.
>>
>> Media specific cleanup is done in usb_audio_disconnect().
>>
>> Signed-off-by: Shuah Khan <shuahkh@osg.samsung.com>
>> Acked-by: Takashi Iwai <tiwai@suse.de>
> 
> Shuah, by looking at the produced graphs:
> 	https://mchehab.fedorapeople.org/mc-next-gen/au0828_test/
> 
> I'm noticing the lack of ALSA I/O entities in the diagram
> (MEDIA_ENT_F_AUDIO_CAPTURE). The mixer there is not connected.
> 
> Could you please check what's happening?
> 
> Those are the relevant dmesg data:
> 
> [   19.017276] usbcore: registered new interface driver snd-usb-audio
> [  230.706102] Linux video capture interface: v2.00
> [  230.856983] au0828: au0828 driver loaded
> [  231.230612] au0828: i2c bus registered
> [  231.822006] tveeprom 5-0050: Hauppauge model 72001, rev E1H3, serial# 4035199481
> [  231.822991] tveeprom 5-0050: MAC address is 00:0d:fe:84:41:f9
> [  231.823955] tveeprom 5-0050: tuner model is Xceive XC5000C (idx 173, type 88)
> [  231.824782] tveeprom 5-0050: TV standards NTSC(M) ATSC/DVB Digital (eeprom 0x88)
> [  231.825272] tveeprom 5-0050: audio processor is AU8522 (idx 44)
> [  231.825276] tveeprom 5-0050: decoder processor is AU8522 (idx 42)
> [  231.825280] tveeprom 5-0050: has no radio, has IR receiver, has no IR transmitter
> [  231.825283] au0828: hauppauge_eeprom: hauppauge eeprom: model=72001
> [  231.857567] au8522 5-0047: creating new instance
> [  231.857879] au8522_decoder creating new instance...
> [  231.910525] tuner 5-0061: Setting mode_mask to 0x06
> [  231.910532] tuner 5-0061: tuner 0x61: Tuner type absent
> [  231.910535] tuner 5-0061: Tuner -1 found with type(s) Radio TV.
> [  231.911343] tuner 5-0061: Calling set_type_addr for type=88, addr=0x61, mode=0x04, config=ffffffffa0e64100
> [  231.911347] tuner 5-0061: defining GPIO callback
> [  231.934896] xc5000 5-0061: creating new instance
> [  231.954664] xc5000: Successfully identified at address 0x61
> [  231.954987] xc5000: Firmware has not been loaded previously
> [  231.955327] tuner 5-0061: type set to Xceive XC5000
> [  231.955330] tuner 5-0061: au0828 tuner I2C addr 0xc2 with type 88 used for 0x04
> [  233.622614] au8522 5-0047: attaching existing instance
> [  233.636061] xc5000 5-0061: attaching existing instance
> [  233.645775] xc5000: Successfully identified at address 0x61
> [  233.646034] xc5000: Firmware has not been loaded previously
> [  233.646461] DVB: registering new adapter (au0828)
> [  233.647575] usb 2-3.3: DVB: registering adapter 0 frontend 0 (Auvitek AU8522 QAM/8VSB Frontend)...
> [  233.648344] dvb_create_media_entity: media entity 'Auvitek AU8522 QAM/8VSB Frontend' registered.
> [  233.656105] dvb_create_media_entity: media entity 'dvb-demux' registered.
> [  234.166644] IR keymap rc-hauppauge not found
> [  234.166962] Registered IR keymap rc-empty
> [  234.178055] input: au0828 IR (Hauppauge HVR950Q) as /devices/pci0000:00/0000:00:14.0/usb2/2-3/2-3.3/rc/rc0/input14
> [  234.203311] rc rc0: au0828 IR (Hauppauge HVR950Q) as /devices/pci0000:00/0000:00:14.0/usb2/2-3/2-3.3/rc/rc0
> [  234.207828] au0828: Remote controller au0828 IR (Hauppauge HVR950Q) initalized
> [  234.208270] au0828: Registered device AU0828 [Hauppauge HVR950Q]
> [  234.212073] usbcore: registered new interface driver au0828
> [  234.230371] lirc_dev: IR Remote Control driver registered, major 243 
> [  234.257555] rc rc0: lirc_dev: driver ir-lirc-codec (au0828-input) registered at minor = 0
> [  234.257960] IR LIRC bridge handler initialized
> 
> (as au0828 is blacklisted, snd-usb-audio was probed first)
> 
> ====xxxx====
> 
> I'm also getting some other weird behavior when removing/reinserting
> the modules a few times. OK, officially we don't support it, but,
> as devices can be bind/unbind all the times, removing modules is
> a way to simulate such things. Also, I use it a lot while testing
> USB drivers ;)
> 
> This one is after removing both the media drivers and snd-usb-audio, 
> and then modprobing snd-usb-audio:
> 

I did see some issues when I did the following sequence:

- blacklisted au0828 and snd-usb-audio was probed first
  graph is good just with audio entities as expected
- modprobed au0828 - graph looks good.
- rmmod au0828 - no problems seen in dmesg
- modprobe au0828 - problems kasan reports bad access etc.
  http://pastebin.com/FFqNzx9G

Here is what's going on after each step:

blacklisted au0828 and snd-usb-audio was probed first
1. snd-usb-audio creates media device and registers it
   Creates its graph etc.
2. modprobed au0828
   au0828 finds the media device created and registered.
   Adds its graph
3. rmmod au0828
   Even though there are no problems reported, at this
   time media device is unregistered, and /dev/mediaX is
   removed. We still have snd_usb-audio and media device.
   As media device is a device resource on usb device parent,
   it will still be there, but no way to access the device
   itself, because it is no longer registered.
4. modprobe au0828
   At this point, au0828 finds the media device as it still
   there, registers it and adds its graph. No audio graph
   present at this time.

Please note that the media device will not be deleted until
the last put on the parent usb struct device. So even when
both snd-usb-audio, and au0828 modules are removed, media
device is still there without its graph and associated devnode
(/dev/mediax is removed).

This isn't bad, however, media_device could still have
stale information.

e.g: enable/disable handlers - when au0828 is removed,
these are no longer valid. Could be cleaned up in
media device unregister just like entity_notify() handler
gets deleted from media device unregister. At the moment,
either driver can call unregister and same cleanup happens.

I will send a patch to do enable/disable handler cleanup
in unregister path.

However, the root of the problem is media device is
still there without its graph and associated devnode
(/dev/mediax is removed) when any one of the drivers
is removed. This leaves the remaining drivers in a
degenerate state.

The problem can be solved with some handshaking at
unregister time. We could add a callback for each
if the drivers to handle media device unregister.
However, that would add delays in device removal path
when all the drivers exit. I think it will be hard to
handle all the corner cases without adding run-time overhead.

Any thoughts on whether we want to unofficially support
being able to remove individual drivers?

thanks,
-- Shuah

-- 
Shuah Khan
Sr. Linux Kernel Developer
Open Source Innovation Group
Samsung Research America (Silicon Valley)
shuahkh@osg.samsung.com | (970) 217-8978