On Wed, 4 Sep 2019, Alan Stern wrote: > The syzbot fuzzer provoked a general protection fault in the > hid-prodikeys driver: > > kasan: CONFIG_KASAN_INLINE enabled > kasan: GPF could be caused by NULL-ptr deref or user memory access > general protection fault: 0000 [#1] SMP KASAN > CPU: 0 PID: 12 Comm: kworker/0:1 Not tainted 5.3.0-rc5+ #28 > Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS > Google 01/01/2011 > Workqueue: usb_hub_wq hub_event > RIP: 0010:pcmidi_submit_output_report drivers/hid/hid-prodikeys.c:300 [inline] > RIP: 0010:pcmidi_set_operational drivers/hid/hid-prodikeys.c:558 [inline] > RIP: 0010:pcmidi_snd_initialise drivers/hid/hid-prodikeys.c:686 [inline] > RIP: 0010:pk_probe+0xb51/0xfd0 drivers/hid/hid-prodikeys.c:836 > Code: 0f 85 50 04 00 00 48 8b 04 24 4c 89 7d 10 48 8b 58 08 e8 b2 53 e4 fc > 48 8b 54 24 20 48 b8 00 00 00 00 00 fc ff df 48 c1 ea 03 <80> 3c 02 00 0f > 85 13 04 00 00 48 ba 00 00 00 00 00 fc ff df 49 8b > > The problem is caused by the fact that pcmidi_get_output_report() will > return an error if the HID device doesn't provide the right sort of > output report, but pcmidi_set_operational() doesn't bother to check > the return code and assumes the function call always succeeds. > > This patch adds the missing check and aborts the probe operation if > necessary. > > Reported-and-tested-by: syzbot+1088533649dafa1c9004@xxxxxxxxxxxxxxxxxxxxxxxxx > Signed-off-by: Alan Stern <stern@xxxxxxxxxxxxxxxxxxx> > CC: <stable@xxxxxxxxxxxxxxx> Applied, thanks a lot Alan. -- Jiri Kosina SUSE Labs
