From: Takashi Iwai <tiwai@suse.de>
To: linux-sound@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Subject: [PATCH] ALSA: core: Serialize snd_refcount put and sync
Date: Thu, 15 Oct 2026 13:22:41 +0200 [thread overview]
Message-ID: <20261015112256.3902-1-tiwai@suse.de> (raw)
When a thread tries to sync-and-destroy the resource with
snd_refcount_sync(), it might lead to a UAF in a very slight time
window. Namely, snd_refount_put() calls wake_up() after
atomic_dec_and_test(), but wait_event() in snd_refcount_sync() checks
the condition via atomic_test() without serialization, hence the
thread that called snd_refcount_sync() can continue immediately to
release the resource before another thread calls wake_up().
For avoiding the race above, introduce the spinlock and take it at
snd_refcount_put() and snd_refcount_sync() in order to serialize the
operations.
Fixes: dcca9b6064c3 ("ALSA: Add simple refcount helper functions")
Reported-by: Sashiko <sashiko-bot@kernel.org>
Closes: https://sashiko.dev/#/patchset/20261015071838.31082-1-tiwai%40suse.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
include/sound/core.h | 1 +
sound/core/misc.c | 5 ++++-
2 files changed, 5 insertions(+), 1 deletion(-)
diff --git a/include/sound/core.h b/include/sound/core.h
index daa91a46a16b..657b12e1e95b 100644
--- a/include/sound/core.h
+++ b/include/sound/core.h
@@ -84,6 +84,7 @@ struct snd_device {
struct snd_refcount {
atomic_t count;
wait_queue_head_t waiter;
+ spinlock_t lock;
};
void snd_refcount_init(struct snd_refcount *ref);
diff --git a/sound/core/misc.c b/sound/core/misc.c
index 066fb9ecdcdc..6fe199e76e83 100644
--- a/sound/core/misc.c
+++ b/sound/core/misc.c
@@ -183,11 +183,13 @@ void snd_refcount_init(struct snd_refcount *ref)
{
atomic_set(&ref->count, 0);
init_waitqueue_head(&ref->waiter);
+ spin_lock_init(&ref->lock);
}
EXPORT_SYMBOL_GPL(snd_refcount_init);
void snd_refcount_put(struct snd_refcount *ref)
{
+ guard(spinlock_irqsave)(&ref->lock);
if (atomic_dec_and_test(&ref->count))
wake_up(&ref->waiter);
}
@@ -195,6 +197,7 @@ EXPORT_SYMBOL_GPL(snd_refcount_put);
void snd_refcount_sync(struct snd_refcount *ref)
{
- wait_event(ref->waiter, !atomic_read(&ref->count));
+ guard(spinlock_irq)(&ref->lock);
+ wait_event_lock_irq(ref->waiter, !atomic_read(&ref->count), ref->lock);
}
EXPORT_SYMBOL_GPL(snd_refcount_sync);
--
2.55.0
reply other threads:[~2026-10-15 11:23 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
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=20261015112256.3902-1-tiwai@suse.de \
--to=tiwai@suse.de \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-sound@vger.kernel.org \
/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
all inboxes | Powered by JetHome®