ALSA: seq: Fix racy pool initializations
"Linux Kernel Mailing List" <[email protected]> Fri, 16 Feb 2018 17:15:05 +0000 (UTC)
| Newsgroups | gmane.linux.kernel.commits.head |
|---|---|
| Message-ID | <[email protected]> |
Web: https://git.kernel.org/torvalds/c/d15d662e89fc667b90cd294b0eb45694e33144da Commit: d15d662e89fc667b90cd294b0eb45694e33144da Parent: 1dcb1859dd6752131b85c78fb3e440bd1d2cf361 Refname: refs/heads/master Author: Takashi Iwai <[email protected]> AuthorDate: Mon Feb 12 15:20:51 2018 +0100 Committer: Takashi Iwai <[email protected]> CommitDate: Wed Feb 14 10:39:08 2018 +0100 ALSA: seq: Fix racy pool initializations ALSA sequencer core initializes the event pool on demand by invoking snd_seq_pool_init() when the first write happens and the pool is empty. Meanwhile user can reset the pool size manually via ioctl concurrently, and this may lead to UAF or out-of-bound accesses since the function tries to vmalloc / vfree the buffer. A simple fix is to just wrap the snd_seq_pool_init() call with the recently introduced client->ioctl_mutex; as the calls for snd_seq_pool_init() from other side are always protected with this mutex, we can avoid the race. Reported-by: 范龙飞 <[email protected]> Cc: <[email protected]> Signed-off-by: Takashi Iwai <[email protected]> --- sound/core/seq/seq_clientmgr.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/sound/core/seq/seq_clientmgr.c b/sound/core/seq/seq_clientmgr.c index 60db32785f62..04d4db44fae5 100644 --- a/sound/core/seq/seq_clientmgr.c +++ b/sound/core/seq/seq_clientmgr.c @@ -1003,7 +1003,7 @@ static ssize_t snd_seq_write(struct file *file, const char __user *buf, { struct snd_seq_client *client = file->private_data; int written = 0, len; - int err = -EINVAL; + int err; struct snd_seq_event event; if (!(snd_seq_file_flags(file) & SNDRV_SEQ_LFLG_OUTPUT)) @@ -1018,11 +1018,15 @@ static ssize_t snd_seq_write(struct file *file, const char __user *buf, /* allocate the pool now if the pool is not allocated yet */ if (client->pool->size > 0 && !snd_seq_write_pool_allocated(client)) { - if (snd_seq_pool_init(client->pool) < 0) + mutex_lock(&client->ioctl_mutex); + err = snd_seq_pool_init(client->pool); + mutex_unlock(&client->ioctl_mutex); + if (err < 0) return -ENOMEM; } /* only process whole events */ + err = -EINVAL; while (count >= sizeof(struct snd_seq_event)) { /* Read in the event header from the user */ len = sizeof(event); -- To unsubscribe from this list: send the line "unsubscribe git-commits-head" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html