Re: KASAN: use-after-free Read in line6_submit_audio_in_all_urbs

[Takashi Iwai <tiwai@suse.de>]

On Fri, 26 Jun 2020 20:26:05 +0200,
Alan Stern wrote:
> 
> On Fri, Jun 26, 2020 at 10:18:12AM -0700, syzbot wrote:
> > Hello,
> > 
> > syzbot found the following crash on:
> > 
> > HEAD commit:    fb574682 usbip: tools: fix module name in man page
> > git tree:       https://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb.git usb-testing
> > console output: https://syzkaller.appspot.com/x/log.txt?x=156560b1100000
> > kernel config:  https://syzkaller.appspot.com/x/.config?x=63b40b2ae167bad6
> > dashboard link: https://syzkaller.appspot.com/bug?extid=145012a46658ac00fc9e
> > compiler:       gcc (GCC) 10.1.0-syz 20200507
> > 
> > Unfortunately, I don't have any reproducer for this crash yet.
> > 
> > IMPORTANT: if you fix the bug, please add the following tag to the commit:
> > Reported-by: syzbot+145012a46658ac00fc9e@syzkaller.appspotmail.com
> > 
> > snd_usb_toneport 5-1:0.0: URB in #0 submission failed (-19)
> > snd_usb_toneport 5-1:0.0: URB in #0 submission failed (-19)
> > snd_usb_toneport 5-1:0.0: URB in #0 submission failed (-19)
> > snd_usb_toneport 5-1:0.0: URB in #0 submission failed (-19)
> > ==================================================================
> > BUG: KASAN: use-after-free in line6_submit_audio_in_all_urbs+0x10b/0x120 sound/usb/line6/capture.c:72
> > Read of size 8 at addr ffff8881cffb1800 by task kworker/1:5/3257
> > 
> > CPU: 1 PID: 3257 Comm: kworker/1:5 Not tainted 5.8.0-rc1-syzkaller #0
> > Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
> > Workqueue: events line6_startup_work
> > Call Trace:
> >  __dump_stack lib/dump_stack.c:77 [inline]
> >  dump_stack+0xf6/0x16e lib/dump_stack.c:118
> >  print_address_description.constprop.0+0x1a/0x210 mm/kasan/report.c:383
> >  __kasan_report mm/kasan/report.c:513 [inline]
> >  kasan_report.cold+0x37/0x7c mm/kasan/report.c:530
> >  line6_submit_audio_in_all_urbs+0x10b/0x120 sound/usb/line6/capture.c:72
> >  line6_stream_start+0x207/0x230 sound/usb/line6/pcm.c:197
> >  line6_pcm_acquire+0x161/0x210 sound/usb/line6/pcm.c:318
> >  line6_startup_work+0x42/0x50 sound/usb/line6/driver.c:734
> >  process_one_work+0x94c/0x15f0 kernel/workqueue.c:2269
> >  worker_thread+0x64c/0x1120 kernel/workqueue.c:2415
> >  kthread+0x392/0x470 kernel/kthread.c:291
> >  ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:293
> > 
> > Allocated by task 76:
> >  save_stack+0x1b/0x40 mm/kasan/common.c:48
> >  set_track mm/kasan/common.c:56 [inline]
> >  __kasan_kmalloc.constprop.0+0xc2/0xd0 mm/kasan/common.c:494
> >  kmalloc include/linux/slab.h:555 [inline]
> >  kzalloc include/linux/slab.h:669 [inline]
> >  line6_init_pcm+0x2a7/0x9e0 sound/usb/line6/pcm.c:533
> >  toneport_init+0xdd/0x6b0 sound/usb/line6/toneport.c:419
> >  line6_probe+0xaa0/0x1330 sound/usb/line6/driver.c:809
> 
> It look like the cancel_delayed_work() in line6_disconnect() needs to be 
> cancel_delayed_work_sync().  Unfortunately we can't test this until syzbot 
> is able to reproduce the bug.

Right, but fixing it now wouldn't be bad.
The fix patch below.


thanks,

Takashi

-- 8< --
From: Takashi Iwai <tiwai@suse.de>
Subject: [PATCH] ALSA: line6: Sync the pending work cancel at disconnection

Recently syzkaller reported a UAF in LINE6 driver, and it's likely
because we call cancel_delayed_work() at the disconnect callback
instead of cancel_delayed_work_sync().  Let's use the correct one
instead.

Reported-by: syzbot+145012a46658ac00fc9e@syzkaller.appspotmail.com
Suggested-by: Alan Stern <stern@rowland.harvard.edu>
Cc: <stable@vger.kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/usb/line6/driver.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/sound/usb/line6/driver.c b/sound/usb/line6/driver.c
index 7629116f570e..2746d9698180 100644
--- a/sound/usb/line6/driver.c
+++ b/sound/usb/line6/driver.c
@@ -840,7 +840,7 @@ void line6_disconnect(struct usb_interface *interface)
 	if (WARN_ON(usbdev != line6->usbdev))
 		return;
 
-	cancel_delayed_work(&line6->startup_work);
+	cancel_delayed_work_sync(&line6->startup_work);
 
 	if (line6->urb_listen != NULL)
 		line6_stop_listen(line6);
-- 
2.16.4