[PATCH 6.6.y] ALSA: pcm: oss: Fix setup list UAF on proc write error

[PATCH 6.6.y] ALSA: pcm: oss: Fix setup list UAF on proc write error

[Artem Dinaburg]

From: =?utf-8?q?C=C3=A1ssio_Gabriel?= <cassiogabrielcontato@gmail.com>

[ Upstream commit 4cc54bdd54b337e77115be5b55577d1c58608eae ]

snd_pcm_oss_proc_write() links a newly allocated setup entry into the
OSS setup list before duplicating the task name. If the task-name
allocation fails, the error path frees the already linked entry and
leaves setup_list pointing at freed memory.

A later OSS device open can then walk the stale list entry in
snd_pcm_oss_look_for_setup() and dereference freed memory.

Allocate the task name and initialize the setup entry before publishing
the entry on setup_list. Also fetch the initial proc read iterator only
after taking setup_mutex, so all setup_list traversal follows the same
list lifetime rules.

Reported-by: syzbot+8e498074a794999eb41c@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/all/6a1062b7.170a0220.35b2b7.0003.GAE@google.com
Closes: https://syzkaller.appspot.com/bug?extid=8e498074a794999eb41c
Fixes: 060d77b9c04a ("[ALSA] Fix / clean up PCM-OSS setup hooks")
Signed-off-by: Cássio Gabriel <cassiogabrielcontato@gmail.com>
Link: https://patch.msgid.link/20260522-alsa-pcm-oss-setup-uaf-v1-1-40bdcc4d17e8@gmail.com
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Assisted-by: LLM
[ Backport to 6.6.y: Keep the target's explicit setup_mutex unlocks on
  allocation failure and before continuing; upstream releases that mutex
  through its scoped guard. ]
Signed-off-by: Artem Dinaburg <artem@trailofbits.com>
---
Hi,

Please consider this backport of the fix for CVE-2026-64001 to 6.6.y.

The code changes are identical to upstream, except explicit mutex unlocks
replace upstream scoped cleanup; see the bracketed backport note above.

The complete fix is already present in 6.12.y, 6.18.y and 7.2.y.

Build-tested only as part of a cumulative 6.6.y series: x86_64 objects
(allmodconfig, CONFIG_WERROR=y); no runtime testing.

Thanks,
Artem Dinaburg

 sound/core/oss/pcm_oss.c |   21 +++++++++++++--------
 1 file changed, 13 insertions(+), 8 deletions(-)

diff --git a/sound/core/oss/pcm_oss.c b/sound/core/oss/pcm_oss.c
index 23ecbd70ccf653aa70a2b08aecf781a194afcc66..25f7e585bc7b93db16b024c1f3915b9895e1c82f 100644
--- a/sound/core/oss/pcm_oss.c
+++ b/sound/core/oss/pcm_oss.c
@@ -2964,8 +2964,10 @@ static void snd_pcm_oss_proc_read(struct snd_info_entry *entry,
 				  struct snd_info_buffer *buffer)
 {
 	struct snd_pcm_str *pstr = entry->private_data;
-	struct snd_pcm_oss_setup *setup = pstr->oss.setup_list;
+	struct snd_pcm_oss_setup *setup;
+
 	mutex_lock(&pstr->oss.setup_mutex);
+	setup = pstr->oss.setup_list;
 	while (setup) {
 		snd_iprintf(buffer, "%s %u %u%s%s%s%s%s%s\n",
 			    setup->task_name,
@@ -3053,6 +3055,14 @@ static void snd_pcm_oss_proc_write(struct snd_info_entry *entry,
 				mutex_unlock(&pstr->oss.setup_mutex);
 				return;
 			}
+			template.task_name = kstrdup(task_name, GFP_KERNEL);
+			if (!template.task_name) {
+				kfree(setup);
+				buffer->error = -ENOMEM;
+				mutex_unlock(&pstr->oss.setup_mutex);
+				return;
+			}
+			*setup = template;
 			if (pstr->oss.setup_list == NULL)
 				pstr->oss.setup_list = setup;
 			else {
@@ -3060,13 +3070,8 @@ static void snd_pcm_oss_proc_write(struct snd_info_entry *entry,
 				     setup1->next; setup1 = setup1->next);
 				setup1->next = setup;
 			}
-			template.task_name = kstrdup(task_name, GFP_KERNEL);
-			if (! template.task_name) {
-				kfree(setup);
-				buffer->error = -ENOMEM;
-				mutex_unlock(&pstr->oss.setup_mutex);
-				return;
-			}
+			mutex_unlock(&pstr->oss.setup_mutex);
+			continue;
 		}
 		*setup = template;
 		mutex_unlock(&pstr->oss.setup_mutex);