From: Takashi Iwai <tiwai@suse.de>
To: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Cc: alsa-devel@alsa-project.org, Mark Brown <broonie@kernel.org>,
Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Subject: Re: [PATCH v2 1/2] ALSA: add snd_card_disconnect_sync()
Date: Wed, 11 Oct 2017 15:13:07 +0200 [thread overview]
Message-ID: <s5hpo9t96ak.wl-tiwai@suse.de> (raw)
In-Reply-To: <s5hzi8x98rb.wl-tiwai@suse.de>
On Wed, 11 Oct 2017 14:19:52 +0200,
Takashi Iwai wrote:
>
> > At least, it's better for us to add enough comments to inform it to
> > developers.
>
> Well, I thought it well informed about the disconnection procedure.
> Maybe the term "lower level driver" needs to be more explicit.
I updated the patch description and added a couple of notes for
explaining your concerns.
thanks,
Takashi
-- 8< --
From: Takashi Iwai <tiwai@suse.de>
Subject: [PATCH] ALSA: add snd_card_disconnect_sync()
In case of user unbind ALSA driver during playing back / capturing,
each driver needs to stop and remove it correctly. One note here is
that we can't cancel from remove function in such case, because
unbind operation doesn't check return value from remove function.
So, we *must* stop and remove in this case.
For this purpose, we need to sync (= wait) until the all top-level
operations are canceled at remove function.
For example, snd_card_free() processes the disconnection procedure at
first, then waits for the completion. That's how the hot-unplug works
safely. It's implemented, at least, in the top-level driver removal.
Now for the lower level driver, we need a similar strategy. Notify to
the toplevel for hot-unplug (disconnect in ALSA), and sync with the
stop operation, then continue the rest of its own remove procedure.
This patch adds snd_card_disconnect_sync(), and driver can use it from
remove function.
Note: the "lower level" driver here refers to a middle layer driver
(e.g. ASoC components) that can be unbound freely during operation.
Most of legacy ALSA helper drivers don't have such a problem because
they can't be unbound.
Note#2: snd_card_disconnect_sync() merely calls snd_card_disconnect()
and syncs with closing all pending files. It takes only the files
opened by user-space into account, and doesn't care about object
refcounts. (The latter is handled by snd_card_free() completion call,
BTW.) Also, the function doesn't free resources by itself.
Tested-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
include/sound/core.h | 2 ++
sound/core/init.c | 32 ++++++++++++++++++++++++++++++++
2 files changed, 34 insertions(+)
diff --git a/include/sound/core.h b/include/sound/core.h
index 4104a9d1001f..5f181b875c2f 100644
--- a/include/sound/core.h
+++ b/include/sound/core.h
@@ -133,6 +133,7 @@ struct snd_card {
struct device card_dev; /* cardX object for sysfs */
const struct attribute_group *dev_groups[4]; /* assigned sysfs attr */
bool registered; /* card_dev is registered? */
+ wait_queue_head_t remove_sleep;
#ifdef CONFIG_PM
unsigned int power_state; /* power state */
@@ -240,6 +241,7 @@ int snd_card_new(struct device *parent, int idx, const char *xid,
struct snd_card **card_ret);
int snd_card_disconnect(struct snd_card *card);
+void snd_card_disconnect_sync(struct snd_card *card);
int snd_card_free(struct snd_card *card);
int snd_card_free_when_closed(struct snd_card *card);
void snd_card_set_id(struct snd_card *card, const char *id);
diff --git a/sound/core/init.c b/sound/core/init.c
index 32ebe2f6bc59..168ae03d3a1c 100644
--- a/sound/core/init.c
+++ b/sound/core/init.c
@@ -255,6 +255,7 @@ int snd_card_new(struct device *parent, int idx, const char *xid,
#ifdef CONFIG_PM
init_waitqueue_head(&card->power_sleep);
#endif
+ init_waitqueue_head(&card->remove_sleep);
device_initialize(&card->card_dev);
card->card_dev.parent = parent;
@@ -452,6 +453,35 @@ int snd_card_disconnect(struct snd_card *card)
}
EXPORT_SYMBOL(snd_card_disconnect);
+/**
+ * snd_card_disconnect_sync - disconnect card and wait until files get closed
+ * @card: card object to disconnect
+ *
+ * This calls snd_card_disconnect() for disconnecting all belonging components
+ * and waits until all pending files get closed.
+ * It assures that all accesses from user-space finished so that the driver
+ * can release its resources gracefully.
+ */
+void snd_card_disconnect_sync(struct snd_card *card)
+{
+ int err;
+
+ err = snd_card_disconnect(card);
+ if (err < 0) {
+ dev_err(card->dev,
+ "snd_card_disconnect error (%d), skipping sync\n",
+ err);
+ return;
+ }
+
+ spin_lock_irq(&card->files_lock);
+ wait_event_lock_irq(card->remove_sleep,
+ list_empty(&card->files_list),
+ card->files_lock);
+ spin_unlock_irq(&card->files_lock);
+}
+EXPORT_SYMBOL_GPL(snd_card_disconnect_sync);
+
static int snd_card_do_free(struct snd_card *card)
{
#if IS_ENABLED(CONFIG_SND_MIXER_OSS)
@@ -957,6 +987,8 @@ int snd_card_file_remove(struct snd_card *card, struct file *file)
break;
}
}
+ if (list_empty(&card->files_list))
+ wake_up_all(&card->remove_sleep);
spin_unlock(&card->files_lock);
if (!found) {
dev_err(card->dev, "card file remove problem (%p)\n", file);
--
2.14.2
next prev parent reply other threads:[~2017-10-11 13:13 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-10-11 10:16 [PATCH v2 0/2] Add snd_card_disconnect_sync() helper Takashi Iwai
2017-10-11 10:16 ` [PATCH v2 1/2] ALSA: add snd_card_disconnect_sync() Takashi Iwai
2017-10-11 11:22 ` Takashi Sakamoto
2017-10-11 12:19 ` Takashi Iwai
2017-10-11 13:13 ` Takashi Iwai [this message]
2017-10-11 10:16 ` [PATCH v2 2/2] ALSA: pcm: Forcibly stop at disconnect callback Takashi Iwai
2017-10-13 7:39 ` [PATCH v2 0/2] Add snd_card_disconnect_sync() helper Kuninori Morimoto
2017-10-13 7:44 ` Takashi Iwai
2017-10-13 9:11 ` Kuninori Morimoto
2017-10-13 9:43 ` Takashi Iwai
2017-10-13 17:11 ` Mark Brown
2017-10-16 2:26 ` Kuninori Morimoto
2017-10-16 15:37 ` Takashi Iwai
2017-10-17 0:59 ` Kuninori Morimoto
2017-10-17 10:37 ` Takashi Iwai
2017-10-17 20:05 ` Mark Brown
2017-10-18 2:08 ` Kuninori Morimoto
2017-10-18 6:10 ` Takashi Iwai
2017-10-18 7:13 ` Kuninori Morimoto
2017-10-17 14:25 ` Takashi Sakamoto
2017-10-17 14:34 ` Takashi Iwai
2017-10-17 15:16 ` Takashi Sakamoto
2017-10-17 16:09 ` Takashi Iwai
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=s5hpo9t96ak.wl-tiwai@suse.de \
--to=tiwai@suse.de \
--cc=alsa-devel@alsa-project.org \
--cc=broonie@kernel.org \
--cc=kuninori.morimoto.gx@renesas.com \
--cc=o-takashi@sakamocchi.jp \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox