mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
* [PATCH v3] ALSA: line6: Kill URBs before releasing the PCM buffer
@ 2026-11-21 10:40 YuanZhe Liu
  2026-11-22  8:16 ` Takashi Iwai
  0 siblings, 1 reply; 3+ messages in thread
From: YuanZhe Liu @ 2026-11-21 10:40 UTC (permalink / raw)
  To: Takashi Iwai, Jaroslav Kysela
  Cc: linux-sound, linux-kernel, stable, Bin Yu, Cheng Wen

line6_buffer_release() frees the stream buffer after
line6_wait_clear_audio_urbs() gave up waiting for the URBs that are still
active.  The buffer is still referenced by those URBs, and the completion
callback of such an URB resubmits it from submit_audio_out_urb() with
line6pcm->out.buffer already set to NULL, which ends up in a NULL pointer
dereference in the memset() in submit_audio_out_urb().

While the URBs remain queued, the host controller can also write the
received isochronous data into the freed buffer.  Such a write does not go
through the KASAN checks, so it silently corrupts the freed slab object;
the corruption was observed as a bogus KASAN quarantine entry that later
led to a general protection fault in qlist_free_all().

The active_urbs bitmap cannot prove that the URBs are gone:
line6_unlink_audio_urbs() is called outside the stream lock and only marks
the URBs that are active at that moment, so an URB resubmitted by a
completion callback right after the scan is never unlinked and keeps being
resubmitted after the stream was stopped.

Set the unlink bit for every currently active URB under the stream lock,
so that the completion callbacks cannot resubmit them, and then kill all
URBs synchronously.  usb_kill_urb() waits for the giveback, hence the
buffer is no longer referenced by any URB when it is freed.  Drop the
unlink marks afterwards: line6_unlink_audio_urbs() can leave a stale mark
on an URB that completed and was not resubmitted, and such a mark makes
line6_stream_start() skip submitting the URBs.

Fixes: 63e20df1e5b2 ("ALSA: line6: Reorganize PCM stream handling")
Cc: stable@vger.kernel.org
Co-developed-by: Bin Yu <byu@xidian.edu.cn>
Signed-off-by: Bin Yu <byu@xidian.edu.cn>
Co-developed-by: Cheng Wen <wencheng@xidian.edu.cn>
Signed-off-by: Cheng Wen <wencheng@xidian.edu.cn>
Signed-off-by: YuanZhe Liu <25031212351@stu.xidian.edu.cn>
---
v3:
- Kill the URBs unconditionally instead of rechecking active_urbs outside
  the stream lock, and clear the unlink marks after the kills.

v2:
- Rebased on sound/for-next.  v1 was generated against v7.3.1 and no longer
  applied, because line6_buffer_release() gained the prev_fbuf clearing
  block in the meantime.  No code change.
---
 sound/usb/line6/pcm.c | 32 +++++++++++++++++++++++++++++++-
 1 file changed, 31 insertions(+), 1 deletion(-)

diff --git a/sound/usb/line6/pcm.c b/sound/usb/line6/pcm.c
index d17baf7761c6..4be3b41a98a2 100644
--- a/sound/usb/line6/pcm.c
+++ b/sound/usb/line6/pcm.c
@@ -134,6 +134,36 @@ static void line6_wait_clear_audio_urbs(struct snd_line6_pcm *line6pcm,
 			"timeout: still %d active urbs..\n", alive);
 }
 
+/*
+ * Mark all currently active URBs for unlinking and kill them.
+ *
+ * The unlink bit is set under the stream lock, so the completion
+ * callbacks cannot resubmit these URBs any longer.  usb_kill_urb() then
+ * waits for the giveback of every URB, hence the stream buffer is no
+ * longer referenced by the host controller.  The unlink marks are
+ * cleared afterwards, so that a later stream start can submit the URBs
+ * again.  This function must be called from a sleepable context.
+ */
+static void line6_kill_audio_urbs(struct snd_line6_pcm *line6pcm,
+				  struct line6_pcm_stream *pcms)
+{
+	int i;
+
+	scoped_guard(spinlock_irqsave, &pcms->lock) {
+		for (i = 0; i < line6pcm->line6->iso_buffers; i++)
+			if (test_bit(i, &pcms->active_urbs))
+				set_bit(i, &pcms->unlink_urbs);
+	}
+
+	for (i = 0; i < line6pcm->line6->iso_buffers; i++)
+		if (pcms->urbs[i])
+			usb_kill_urb(pcms->urbs[i]);
+
+	scoped_guard(spinlock_irqsave, &pcms->lock) {
+		pcms->unlink_urbs = 0;
+	}
+}
+
 static inline struct line6_pcm_stream *
 get_stream(struct snd_line6_pcm *line6pcm, int direction)
 {
@@ -172,7 +202,7 @@ static void line6_buffer_release(struct snd_line6_pcm *line6pcm,
 {
 	clear_bit(type, &pstr->opened);
 	if (!pstr->opened) {
-		line6_wait_clear_audio_urbs(line6pcm, pstr);
+		line6_kill_audio_urbs(line6pcm, pstr);
 		if (pstr == &line6pcm->in) {
 			/* prev_fbuf points into the capture buffer we are about
 			 * to free; drop it while no capture URB is in flight.
-- 
2.34.1


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH v3] ALSA: line6: Kill URBs before releasing the PCM buffer
  2026-11-21 10:40 [PATCH v3] ALSA: line6: Kill URBs before releasing the PCM buffer YuanZhe Liu
@ 2026-11-22  8:16 ` Takashi Iwai
  2026-11-22  9:04   ` [PATCH v4] ALSA: line6: Stop URBs synchronously via sync_stop callback YuanZhe Liu
  0 siblings, 1 reply; 3+ messages in thread
From: Takashi Iwai @ 2026-11-22  8:16 UTC (permalink / raw)
  To: YuanZhe Liu
  Cc: Takashi Iwai, Jaroslav Kysela, linux-sound, linux-kernel, stable,
	Bin Yu, Cheng Wen

On Sat, 21 Nov 2026 11:40:00 +0100,
YuanZhe Liu wrote:
> 
> line6_buffer_release() frees the stream buffer after
> line6_wait_clear_audio_urbs() gave up waiting for the URBs that are still
> active.  The buffer is still referenced by those URBs, and the completion
> callback of such an URB resubmits it from submit_audio_out_urb() with
> line6pcm->out.buffer already set to NULL, which ends up in a NULL pointer
> dereference in the memset() in submit_audio_out_urb().
> 
> While the URBs remain queued, the host controller can also write the
> received isochronous data into the freed buffer.  Such a write does not go
> through the KASAN checks, so it silently corrupts the freed slab object;
> the corruption was observed as a bogus KASAN quarantine entry that later
> led to a general protection fault in qlist_free_all().
> 
> The active_urbs bitmap cannot prove that the URBs are gone:
> line6_unlink_audio_urbs() is called outside the stream lock and only marks
> the URBs that are active at that moment, so an URB resubmitted by a
> completion callback right after the scan is never unlinked and keeps being
> resubmitted after the stream was stopped.
> 
> Set the unlink bit for every currently active URB under the stream lock,
> so that the completion callbacks cannot resubmit them, and then kill all
> URBs synchronously.  usb_kill_urb() waits for the giveback, hence the
> buffer is no longer referenced by any URB when it is freed.  Drop the
> unlink marks afterwards: line6_unlink_audio_urbs() can leave a stale mark
> on an URB that completed and was not resubmitted, and such a mark makes
> line6_stream_start() skip submitting the URBs.
> 
> Fixes: 63e20df1e5b2 ("ALSA: line6: Reorganize PCM stream handling")
> Cc: stable@vger.kernel.org
> Co-developed-by: Bin Yu <byu@xidian.edu.cn>
> Signed-off-by: Bin Yu <byu@xidian.edu.cn>
> Co-developed-by: Cheng Wen <wencheng@xidian.edu.cn>
> Signed-off-by: Cheng Wen <wencheng@xidian.edu.cn>
> Signed-off-by: YuanZhe Liu <25031212351@stu.xidian.edu.cn>

Thanks for the patch.

But, this kind of synchronization should be rather done in sync_stop
PCM ops that seems missing in the current LINE6 driver.
 
I guess line6_wait_clear_audio_urbs() could be replaced with
usb_kill_urb() as you've done, and move it to sync_stop callback.
Then it can be dropped from snd_line6_prepare() and
line6_pcm_acquire().  line6_pcm_disconnect() still might need it,
though.


Takashi

> v3:
> - Kill the URBs unconditionally instead of rechecking active_urbs outside
>   the stream lock, and clear the unlink marks after the kills.
> 
> v2:
> - Rebased on sound/for-next.  v1 was generated against v7.3.1 and no longer
>   applied, because line6_buffer_release() gained the prev_fbuf clearing
>   block in the meantime.  No code change.
> ---
>  sound/usb/line6/pcm.c | 32 +++++++++++++++++++++++++++++++-
>  1 file changed, 31 insertions(+), 1 deletion(-)
> 
> diff --git a/sound/usb/line6/pcm.c b/sound/usb/line6/pcm.c
> index d17baf7761c6..4be3b41a98a2 100644
> --- a/sound/usb/line6/pcm.c
> +++ b/sound/usb/line6/pcm.c
> @@ -134,6 +134,36 @@ static void line6_wait_clear_audio_urbs(struct snd_line6_pcm *line6pcm,
>  			"timeout: still %d active urbs..\n", alive);
>  }
>  
> +/*
> + * Mark all currently active URBs for unlinking and kill them.
> + *
> + * The unlink bit is set under the stream lock, so the completion
> + * callbacks cannot resubmit these URBs any longer.  usb_kill_urb() then
> + * waits for the giveback of every URB, hence the stream buffer is no
> + * longer referenced by the host controller.  The unlink marks are
> + * cleared afterwards, so that a later stream start can submit the URBs
> + * again.  This function must be called from a sleepable context.
> + */
> +static void line6_kill_audio_urbs(struct snd_line6_pcm *line6pcm,
> +				  struct line6_pcm_stream *pcms)
> +{
> +	int i;
> +
> +	scoped_guard(spinlock_irqsave, &pcms->lock) {
> +		for (i = 0; i < line6pcm->line6->iso_buffers; i++)
> +			if (test_bit(i, &pcms->active_urbs))
> +				set_bit(i, &pcms->unlink_urbs);
> +	}
> +
> +	for (i = 0; i < line6pcm->line6->iso_buffers; i++)
> +		if (pcms->urbs[i])
> +			usb_kill_urb(pcms->urbs[i]);
> +
> +	scoped_guard(spinlock_irqsave, &pcms->lock) {
> +		pcms->unlink_urbs = 0;
> +	}
> +}
> +
>  static inline struct line6_pcm_stream *
>  get_stream(struct snd_line6_pcm *line6pcm, int direction)
>  {
> @@ -172,7 +202,7 @@ static void line6_buffer_release(struct snd_line6_pcm *line6pcm,
>  {
>  	clear_bit(type, &pstr->opened);
>  	if (!pstr->opened) {
> -		line6_wait_clear_audio_urbs(line6pcm, pstr);
> +		line6_kill_audio_urbs(line6pcm, pstr);
>  		if (pstr == &line6pcm->in) {
>  			/* prev_fbuf points into the capture buffer we are about
>  			 * to free; drop it while no capture URB is in flight.
> -- 
> 2.34.1
> 

^ permalink raw reply	[flat|nested] 3+ messages in thread

* [PATCH v4] ALSA: line6: Stop URBs synchronously via sync_stop callback
  2026-11-22  8:16 ` Takashi Iwai
@ 2026-11-22  9:04   ` YuanZhe Liu
  0 siblings, 0 replies; 3+ messages in thread
From: YuanZhe Liu @ 2026-11-22  9:04 UTC (permalink / raw)
  To: Takashi Iwai, Jaroslav Kysela
  Cc: linux-sound, linux-kernel, stable, Bin Yu, Cheng Wen

line6_buffer_release() frees the stream buffer after
line6_wait_clear_audio_urbs() gave up waiting for the URBs that are still
active.  The buffer is still referenced by those URBs, and the completion
callback of such an URB resubmits it from submit_audio_out_urb() with
line6pcm->out.buffer already set to NULL, which ends up in a NULL pointer
dereference in the memset() in submit_audio_out_urb().

While the URBs remain queued, the host controller can also write the
received isochronous data into the freed buffer.  Such a write does not go
through the KASAN checks, so it silently corrupts the freed slab object;
the corruption was observed as a bogus KASAN quarantine entry that later
led to a general protection fault in qlist_free_all().

The active_urbs bitmap cannot prove that the URBs are gone:
line6_unlink_audio_urbs() is called outside the stream lock and only marks
the URBs that are active at that moment, so an URB resubmitted by a
completion callback right after the scan is never unlinked and keeps being
resubmitted after the stream was stopped.

Stop the URBs synchronously instead of polling for them: mark all
currently active URBs for unlinking under the stream lock, so that the
completion callbacks cannot resubmit them, and kill them with
usb_kill_urb(), which waits for the giveback.  Expose this through the
sync_stop PCM op, which ALSA core invokes before hw_free(), hw_params()
and prepare() once the stream was stopped, so the buffer is guaranteed to
be unused when it is released.

Replace line6_wait_clear_audio_urbs() with this helper accordingly, and
drop its now redundant calls from snd_line6_prepare() and
line6_pcm_acquire().  line6_pcm_disconnect() keeps stopping the URBs
synchronously, since the device may be gone already.

The monitor and impulse streams are started via line6_pcm_acquire() and
released via line6_pcm_release() without going through the PCM ops, so
line6_buffer_release() still stops the URBs before it frees the buffer.

Fixes: 63e20df1e5b2 ("ALSA: line6: Reorganize PCM stream handling")
Cc: stable@vger.kernel.org
Co-developed-by: Bin Yu <byu@xidian.edu.cn>
Signed-off-by: Bin Yu <byu@xidian.edu.cn>
Co-developed-by: Cheng Wen <wencheng@xidian.edu.cn>
Signed-off-by: Cheng Wen <wencheng@xidian.edu.cn>
Signed-off-by: YuanZhe Liu <25031212351@stu.xidian.edu.cn>
---
v4:
- Add the sync_stop PCM op and drop the waits from snd_line6_prepare()
  and line6_pcm_acquire(), as suggested by Takashi Iwai.

v3:
- Kill the URBs unconditionally instead of rechecking active_urbs outside
  the stream lock, and clear the unlink marks after the kills.

v2:
- Rebased on sound/for-next.  v1 was generated against v7.3.1 and no longer
  applied, because line6_buffer_release() gained the prev_fbuf clearing
  block in the meantime.  No code change.
---
 sound/usb/line6/capture.c  |  1 +
 sound/usb/line6/pcm.c      | 66 +++++++++++++++++++++-----------------
 sound/usb/line6/pcm.h      |  1 +
 sound/usb/line6/playback.c |  1 +
 4 files changed, 40 insertions(+), 29 deletions(-)

diff --git a/sound/usb/line6/capture.c b/sound/usb/line6/capture.c
index 626274e503c7..fa98c26cf77d 100644
--- a/sound/usb/line6/capture.c
+++ b/sound/usb/line6/capture.c
@@ -284,6 +284,7 @@ const struct snd_pcm_ops snd_line6_capture_ops = {
 	.hw_free = snd_line6_hw_free,
 	.prepare = snd_line6_prepare,
 	.trigger = snd_line6_trigger,
+	.sync_stop = snd_line6_sync_stop,
 	.pointer = snd_line6_pointer,
 };
 
diff --git a/sound/usb/line6/pcm.c b/sound/usb/line6/pcm.c
index d17baf7761c6..dc850aa2ebd0 100644
--- a/sound/usb/line6/pcm.c
+++ b/sound/usb/line6/pcm.c
@@ -109,29 +109,33 @@ static void line6_unlink_audio_urbs(struct snd_line6_pcm *line6pcm,
 }
 
 /*
-	Wait until unlinking of all currently active URBs has been finished.
-*/
-static void line6_wait_clear_audio_urbs(struct snd_line6_pcm *line6pcm,
-					struct line6_pcm_stream *pcms)
+ * Mark all currently active URBs for unlinking and kill them.
+ *
+ * The unlink bit is set under the stream lock, so the completion
+ * callbacks cannot resubmit these URBs any longer.  usb_kill_urb() then
+ * waits for the giveback of every URB, hence the stream buffer is no
+ * longer referenced by the host controller.  The unlink marks are
+ * cleared afterwards, so that a later stream start can submit the URBs
+ * again.  This function must be called from a sleepable context.
+ */
+static void line6_kill_audio_urbs(struct snd_line6_pcm *line6pcm,
+				  struct line6_pcm_stream *pcms)
 {
-	int timeout = HZ;
 	int i;
-	int alive;
 
-	do {
-		alive = 0;
-		for (i = 0; i < line6pcm->line6->iso_buffers; i++) {
+	scoped_guard(spinlock_irqsave, &pcms->lock) {
+		for (i = 0; i < line6pcm->line6->iso_buffers; i++)
 			if (test_bit(i, &pcms->active_urbs))
-				alive++;
-		}
-		if (!alive)
-			break;
-		set_current_state(TASK_UNINTERRUPTIBLE);
-		schedule_timeout(1);
-	} while (--timeout > 0);
-	if (alive)
-		dev_err(line6pcm->line6->ifcdev,
-			"timeout: still %d active urbs..\n", alive);
+				set_bit(i, &pcms->unlink_urbs);
+	}
+
+	for (i = 0; i < line6pcm->line6->iso_buffers; i++)
+		if (pcms->urbs[i])
+			usb_kill_urb(pcms->urbs[i]);
+
+	scoped_guard(spinlock_irqsave, &pcms->lock) {
+		pcms->unlink_urbs = 0;
+	}
 }
 
 static inline struct line6_pcm_stream *
@@ -172,7 +176,7 @@ static void line6_buffer_release(struct snd_line6_pcm *line6pcm,
 {
 	clear_bit(type, &pstr->opened);
 	if (!pstr->opened) {
-		line6_wait_clear_audio_urbs(line6pcm, pstr);
+		line6_kill_audio_urbs(line6pcm, pstr);
 		if (pstr == &line6pcm->in) {
 			/* prev_fbuf points into the capture buffer we are about
 			 * to free; drop it while no capture URB is in flight.
@@ -337,8 +341,6 @@ int line6_pcm_acquire(struct snd_line6_pcm *line6pcm, int type, bool start)
 		ret = line6_buffer_acquire(line6pcm, pstr, dir, type);
 		if (ret < 0)
 			goto error;
-		if (!pstr->running)
-			line6_wait_clear_audio_urbs(line6pcm, pstr);
 	}
 	if (start) {
 		for (dir = 0; dir < 2; dir++) {
@@ -510,10 +512,8 @@ static int snd_line6_new_pcm(struct usb_line6 *line6, struct snd_pcm **pcm_ret)
 */
 void line6_pcm_disconnect(struct snd_line6_pcm *line6pcm)
 {
-	line6_unlink_audio_urbs(line6pcm, &line6pcm->out);
-	line6_unlink_audio_urbs(line6pcm, &line6pcm->in);
-	line6_wait_clear_audio_urbs(line6pcm, &line6pcm->out);
-	line6_wait_clear_audio_urbs(line6pcm, &line6pcm->in);
+	line6_kill_audio_urbs(line6pcm, &line6pcm->out);
+	line6_kill_audio_urbs(line6pcm, &line6pcm->in);
 }
 
 /*
@@ -592,15 +592,23 @@ int line6_init_pcm(struct usb_line6 *line6,
 }
 EXPORT_SYMBOL_GPL(line6_init_pcm);
 
+/* common PCM sync_stop callback */
+int snd_line6_sync_stop(struct snd_pcm_substream *substream)
+{
+	struct snd_line6_pcm *line6pcm = snd_pcm_substream_chip(substream);
+	struct line6_pcm_stream *pstr = get_stream(line6pcm, substream->stream);
+
+	guard(mutex)(&line6pcm->state_mutex);
+	line6_kill_audio_urbs(line6pcm, pstr);
+	return 0;
+}
+
 /* prepare pcm callback */
 int snd_line6_prepare(struct snd_pcm_substream *substream)
 {
 	struct snd_line6_pcm *line6pcm = snd_pcm_substream_chip(substream);
-	struct line6_pcm_stream *pstr = get_stream(line6pcm, substream->stream);
 
 	guard(mutex)(&line6pcm->state_mutex);
-	if (!pstr->running)
-		line6_wait_clear_audio_urbs(line6pcm, pstr);
 
 	if (!test_and_set_bit(LINE6_FLAG_PREPARED, &line6pcm->flags)) {
 		line6pcm->out.count = 0;
diff --git a/sound/usb/line6/pcm.h b/sound/usb/line6/pcm.h
index a15913bf2a7a..21e550948cd8 100644
--- a/sound/usb/line6/pcm.h
+++ b/sound/usb/line6/pcm.h
@@ -183,6 +183,7 @@ extern int line6_init_pcm(struct usb_line6 *line6,
 			  struct line6_pcm_properties *properties);
 extern int snd_line6_trigger(struct snd_pcm_substream *substream, int cmd);
 extern int snd_line6_prepare(struct snd_pcm_substream *substream);
+extern int snd_line6_sync_stop(struct snd_pcm_substream *substream);
 extern int snd_line6_hw_params(struct snd_pcm_substream *substream,
 			       struct snd_pcm_hw_params *hw_params);
 extern int snd_line6_hw_free(struct snd_pcm_substream *substream);
diff --git a/sound/usb/line6/playback.c b/sound/usb/line6/playback.c
index e93a16f6958e..10e273f4b77a 100644
--- a/sound/usb/line6/playback.c
+++ b/sound/usb/line6/playback.c
@@ -414,6 +414,7 @@ const struct snd_pcm_ops snd_line6_playback_ops = {
 	.hw_free = snd_line6_hw_free,
 	.prepare = snd_line6_prepare,
 	.trigger = snd_line6_trigger,
+	.sync_stop = snd_line6_sync_stop,
 	.pointer = snd_line6_pointer,
 };
 
-- 
2.34.1


^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2026-11-22  9:05 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2026-11-21 10:40 [PATCH v3] ALSA: line6: Kill URBs before releasing the PCM buffer YuanZhe Liu
2026-11-22  8:16 ` Takashi Iwai
2026-11-22  9:04   ` [PATCH v4] ALSA: line6: Stop URBs synchronously via sync_stop callback YuanZhe Liu

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®