[PATCH 3.2 135/199] ALSA: seq: Don't handle loop timeout at snd_seq_pool_done()

From: Ben Hutchings
Date: Fri Mar 10 2017 - 06:54:44 EST


3.2.87-rc1 review patch. If anyone has any objections, please let me know.

------------------

From: Takashi Iwai <tiwai@xxxxxxx>

commit 37a7ea4a9b81f6a864c10a7cb0b96458df5310a3 upstream.

snd_seq_pool_done() syncs with closing of all opened threads, but it
aborts the wait loop with a timeout, and proceeds to the release
resource even if not all threads have been closed. The timeout was 5
seconds, and if you run a crazy stuff, it can exceed easily, and may
result in the access of the invalid memory address -- this is what
syzkaller detected in a bug report.

As a fix, let the code graduate from naiveness, simply remove the loop
timeout.

BugLink: http://lkml.kernel.org/r/CACT4Y+YdhDV2H5LLzDTJDVF-qiYHUHhtRaW4rbb4gUhTCQB81w@xxxxxxxxxxxxxx
Reported-by: Dmitry Vyukov <dvyukov@xxxxxxxxxx>
Signed-off-by: Takashi Iwai <tiwai@xxxxxxx>
[bwh: Backported to 3.2: deleted log statement is slightly different]
Signed-off-by: Ben Hutchings <ben@xxxxxxxxxxxxxxx>
---
sound/core/seq/seq_memory.c | 9 +--------
1 file changed, 1 insertion(+), 8 deletions(-)

--- a/sound/core/seq/seq_memory.c
+++ b/sound/core/seq/seq_memory.c
@@ -421,7 +421,6 @@ int snd_seq_pool_done(struct snd_seq_poo
{
unsigned long flags;
struct snd_seq_event_cell *ptr;
- int max_count = 5 * HZ;

if (snd_BUG_ON(!pool))
return -EINVAL;
@@ -434,14 +433,8 @@ int snd_seq_pool_done(struct snd_seq_poo
if (waitqueue_active(&pool->output_sleep))
wake_up(&pool->output_sleep);

- while (atomic_read(&pool->counter) > 0) {
- if (max_count == 0) {
- snd_printk(KERN_WARNING "snd_seq_pool_done timeout: %d cells remain\n", atomic_read(&pool->counter));
- break;
- }
+ while (atomic_read(&pool->counter) > 0)
schedule_timeout_uninterruptible(1);
- max_count--;
- }

/* release all resources */
spin_lock_irqsave(&pool->lock, flags);