Skip to contentSkip to conversation

media: redrat3: fix UAF in probe error path leaving rc device registered

PATCH In review Rik van Riel 2 participants 4 messages Aug 9 – Aug 13

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 media: redrat3: fix UAF in probe error path leaving rc device registered

redrat3 stores its device name and phys path inside the main rr3 structure, and the rc device's device_name and input_phys point into rr3. When the RC device is registered, it keeps those pointers.

  KASAN: slab-use-after-free in string_nocheck lib/vsprintf.c:648
  Read of size 1 at addr ffff888051fda758 by task udevd/7464
  Call Trace:
   string_nocheck lib/vsprintf.c:648 [inline]
   string+0x216/0x2d0 lib/vsprintf.c:730
   vsnprintf+0x74a/0xef0 lib/vsprintf.c:2945
   vscnprintf+0x41/0x90 lib/vsprintf.c:3014
   sysfs_emit+0x10e/0x180 fs/sysfs/file.c:761
   input_dev_show_name+0x58/0x70 drivers/input/input.c:1282

  Allocated by task 10:
   redrat3_dev_probe+0x477/0x2570 drivers/media/rc/redrat3.c:1023
  Freed by task 10:
   redrat3_delete drivers/media/rc/redrat3.c:466 [inline]
   redrat3_dev_probe+0x1bf4/0x2570 drivers/media/rc/redrat3.c:1124

Syzkaller triggers this via usb probing. It probes the RedRat3 USB interface, which calls redrat3_dev_probe() in redrat3.c. That function allocates rr3, then builds an rc device whose name points into rr3 via redrat3_init_rc_dev() in redrat3.c, and registers it with rc_register_device().

When the detector enable fails after the RC device is registered, the probe jumps to the led_free path. That path unregisters the LED but does not unregister the RC device before freeing rr3 via redrat3_delete() in redrat3.c. The RC device still holds device_name = rr3->name which is now freed.

Later udevd reads /sys/.../input device name via sysfs_emit() in file.c, which calls input_dev_show_name() in input.c, which emits dev->name which is the freed rr3->name.

When the detector enable fails after the RC device is registered, the error path must not leave the RC device registered. Make the probe error path clean up the same way redrat3_dev_disconnect() does.

This change should be safe because the RC device is fully registered at this point and its teardown via rc_unregister_device() is protected by the input device mutex, and rr3 is still alive during unregister so device_name remains valid until after unregister. The URBs are killed before rc_free, so no completion can run after rc is freed. No new lock ordering is introduced.

Reported-by [email protected]
Fixes 2154be651b90 ("[media] redrat3: new rc-core IR transceiver device driver")
Assisted-by Hermes:muse-spark-1.2 syzkaller

---
v2: fix the Sashiko bug https://sashiko.dev/#/patchset/20260808174213.1d68336e%40fangorn

 drivers/media/rc/redrat3.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/media/rc/redrat3.c +4 -0
index 3f828a564e19..565c490b7051 100644
@@ -975,6 +975,7 @@ static int redrat3_dev_probe(struct usb_interface *intf,
 	struct device *dev = &intf->dev;
 	struct usb_host_interface *uhi;
 	struct redrat3_dev *rr3;
+	struct rc_dev *rc;
 	struct usb_endpoint_descriptor *ep;
 	struct usb_endpoint_descriptor *ep_narrow = NULL;
 	struct usb_endpoint_descriptor *ep_wide = NULL;
@@ -1119,9 +1120,12 @@ static int redrat3_dev_probe(struct usb_interface *intf,
 	return 0;
 
 led_free:
+	rc_unregister_device(rr3->rc);
 	led_classdev_unregister(&rr3->led);
 redrat_free:
+	rc = rr3->rc;
 	redrat3_delete(rr3, rr3->udev);
+	rc_free_device(rc);
 
 no_endpoints:
 	return retval;
Signature · 1 line
2.55.0
maintainer
#1 lore raw

Hoi Rik,

Quoted 47 lines · Rik van Riel

redrat3 stores its device name and phys path inside the main rr3 structure, and the rc device's device_name and input_phys point into rr3. When the RC device is registered, it keeps those pointers.

  KASAN: slab-use-after-free in string_nocheck lib/vsprintf.c:648
  Read of size 1 at addr ffff888051fda758 by task udevd/7464
  Call Trace:
   string_nocheck lib/vsprintf.c:648 [inline]
   string+0x216/0x2d0 lib/vsprintf.c:730
   vsnprintf+0x74a/0xef0 lib/vsprintf.c:2945
   vscnprintf+0x41/0x90 lib/vsprintf.c:3014
   sysfs_emit+0x10e/0x180 fs/sysfs/file.c:761
   input_dev_show_name+0x58/0x70 drivers/input/input.c:1282

  Allocated by task 10:
   redrat3_dev_probe+0x477/0x2570 drivers/media/rc/redrat3.c:1023
  Freed by task 10:
   redrat3_delete drivers/media/rc/redrat3.c:466 [inline]
   redrat3_dev_probe+0x1bf4/0x2570 drivers/media/rc/redrat3.c:1124

Syzkaller triggers this via usb probing. It probes the RedRat3 USB interface, which calls redrat3_dev_probe() in redrat3.c. That function allocates rr3, then builds an rc device whose name points into rr3 via redrat3_init_rc_dev() in redrat3.c, and registers it with rc_register_device().

When the detector enable fails after the RC device is registered, the probe jumps to the led_free path. That path unregisters the LED but does not unregister the RC device before freeing rr3 via redrat3_delete() in redrat3.c. The RC device still holds device_name = rr3->name which is now freed.

Later udevd reads /sys/.../input device name via sysfs_emit() in file.c, which calls input_dev_show_name() in input.c, which emits dev->name which is the freed rr3->name.

When the detector enable fails after the RC device is registered, the error path must not leave the RC device registered. Make the probe error path clean up the same way redrat3_dev_disconnect() does.

This change should be safe because the RC device is fully registered at this point and its teardown via rc_unregister_device() is protected by the input device mutex, and rr3 is still alive during unregister so device_name remains valid until after unregister. The URBs are killed before rc_free, so no completion can run after rc is freed. No new lock ordering is introduced.

Quoted 7 lines
Reported-by [email protected]
Fixes 2154be651b90 ("[media] redrat3: new rc-core IR transceiver device driver")
Assisted-by Hermes:muse-spark-1.2 syzkaller

The commit message is very good. Your patch is very similar to a patch I wrote earlier (which still has to be reviewed, I can't merge my own patches without review into the media-committer tree):

https://lkml.org/lkml/2026/7/29/1730

I don't really have a preference for which one to merge.

Sean

Quoted 34 lines

---
v2: fix the Sashiko bug https://sashiko.dev/#/patchset/20260808174213.1d68336e%40fangorn

 drivers/media/rc/redrat3.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/media/rc/redrat3.c +4 -0
index 3f828a564e19..565c490b7051 100644
@@ -975,6 +975,7 @@ static int redrat3_dev_probe(struct usb_interface *intf,
 	struct device *dev = &intf->dev;
 	struct usb_host_interface *uhi;
 	struct redrat3_dev *rr3;
+	struct rc_dev *rc;
 	struct usb_endpoint_descriptor *ep;
 	struct usb_endpoint_descriptor *ep_narrow = NULL;
 	struct usb_endpoint_descriptor *ep_wide = NULL;
@@ -1119,9 +1120,12 @@ static int redrat3_dev_probe(struct usb_interface *intf,
 	return 0;
 
 led_free:
+	rc_unregister_device(rr3->rc);
 	led_classdev_unregister(&rr3->led);
 redrat_free:
+	rc = rr3->rc;
 	redrat3_delete(rr3, rr3->udev);
+	rc_free_device(rc);
 
 no_endpoints:
 	return retval;
Signature · 1 line
2.55.0
#2 lore raw
Quoted 7 lines · Sean Young
Quoted 6 lines

https://lore.kernel.org/all/[email protected]/

Fixes 2154be651b90 ("[media] redrat3: new rc-core IR transceiver

device driver")

Assisted-by Hermes:muse-spark-1.2 syzkaller
Quoted 8 lines

The commit message is very good. Your patch is very similar to a patch
I wrote earlier (which still has to be reviewed, I can't merge my own patches without review into the media-committer tree):

https://lkml.org/lkml/2026/7/29/1730

I don't really have a preference for which one to merge.

If it's any consolation, Sashiko suggests there
is another race condition left, pre-existing in
the code.

https://sashiko.dev/#/patchset/20260808232031.500f781c%40fangorn

I would be happy to review patches to fix that.

Signature · 1 line
All Rights Reversed.
maintainer
#3 lore raw
Quoted 17 lines · Rik van Riel
Quoted 7 lines · Sean Young
Quoted 6 lines

https://lore.kernel.org/all/[email protected]/

Fixes 2154be651b90 ("[media] redrat3: new rc-core IR transceiver

device driver")

Assisted-by Hermes:muse-spark-1.2 syzkaller
Quoted 8 lines

The commit message is very good. Your patch is very similar to a patch
I wrote earlier (which still has to be reviewed, I can't merge my own patches without review into the media-committer tree):

https://lkml.org/lkml/2026/7/29/1730

I don't really have a preference for which one to merge.

Quoted 7 lines

If it's any consolation, Sashiko suggests there
is another race condition left, pre-existing in
the code.

https://sashiko.dev/#/patchset/20260808232031.500f781c%40fangorn

I would be happy to review patches to fix that.

Thank you. I've think I've got all them covered. The first series is here:

	https://lore.kernel.org/all/[email protected]/

I've got another series almost ready to go (mostly locking issues).

For more than 10 years I've been working on rc-core (on and off) and I convinced myself it was pretty solid. It turns out that I was deluded and Sashiko found lots of embarrasing problems. It's a shame that I can't prompt Sashiko to find all bugs in rc-core. I have used claude, and it found problems sashiko missed (and vice versa). 19 patches so far for problems llms found, which is a lot for a small subsystem.

Sean