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

[Laurent Pinchart]

Hi Oliver,

On Tue, Feb 11, 2020 at 03:31:30PM +0100, Oliver Neukum wrote:
> Am Montag, den 10.02.2020, 16:18 +0200 schrieb Laurent Pinchart:
> > On Mon, Feb 10, 2020 at 03:13:26PM +0100, Oliver Neukum wrote:
> > > Am Montag, den 13.01.2020, 04:24 -0800 schrieb syzbot:
> > > > Hello,
> > > > 
> > > > syzbot found the following crash on:
> > > > 
> > > > HEAD commit:    ae179410 usb: gadget: add raw-gadget interface
> > > > git tree:       https://github.com/google/kasan.git usb-fuzzer
> > > > console output: https://syzkaller.appspot.com/x/log.txt?x=132223fee00000
> > > > kernel config:  https://syzkaller.appspot.com/x/.config?x=ad1d751a3a72ae57
> > > > dashboard link: https://syzkaller.appspot.com/bug?extid=9a48339b077c5a80b869
> > > > compiler:       gcc (GCC) 9.0.0 20181231 (experimental)
> > > > syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=16857325e00000
> > > > C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=142e069ee00000
> > > > 
> > > > IMPORTANT: if you fix the bug, please add the following tag to the commit:
> > > > Reported-by: syzbot+9a48339b077c5a80b869@xxxxxxxxxxxxxxxxxxxxxxxxx
> > > > 
> > > > usb 1-1: New USB device found, idVendor=0bd3, idProduct=0555,  
> > > > bcdDevice=69.6a
> > > > usb 1-1: New USB device strings: Mfr=0, Product=0, SerialNumber=0
> > > > usb 1-1: config 0 descriptor??
> > > > usb 1-1: string descriptor 0 read error: -71
> > > > uvcvideo: Found UVC 0.00 device <unnamed> (0bd3:0555)
> > > > ==================================================================
> > > > BUG: KASAN: use-after-free in uvc_register_terms  
> > > > drivers/media/usb/uvc/uvc_driver.c:2038 [inline]
> > > > BUG: KASAN: use-after-free in uvc_register_chains  
> > > > drivers/media/usb/uvc/uvc_driver.c:2070 [inline]
> > > > BUG: KASAN: use-after-free in uvc_probe.cold+0x2193/0x29de  
> > > > drivers/media/usb/uvc/uvc_driver.c:2201
> > > > Read of size 2 at addr ffff8881d4f1bc2e by task kworker/1:2/94
> > > 
> > > #syz test: https://github.com/google/kasan.git ae179410
> > > 
> > > From db844641a5e30f3cfc0ce9cde156b3cc356b6c0c Mon Sep 17 00:00:00 2001
> > > From: Oliver Neukum <oneukum@xxxxxxxx>
> > > Date: Mon, 10 Feb 2020 15:10:36 +0100
> > > Subject: [PATCH] UVC: deal with unnamed streams
> > > 
> > > The pointer can be NULL
> > > 
> > > Signed-off-by: Oliver Neukum <oneukum@xxxxxxxx>
> > > ---
> > >  drivers/media/usb/uvc/uvc_driver.c | 3 ++-
> > >  1 file changed, 2 insertions(+), 1 deletion(-)
> > > 
> > > diff --git a/drivers/media/usb/uvc/uvc_driver.c b/drivers/media/usb/uvc/uvc_driver.c
> > > index 99883550375e..26558a89f2fe 100644
> > > --- a/drivers/media/usb/uvc/uvc_driver.c
> > > +++ b/drivers/media/usb/uvc/uvc_driver.c
> > > @@ -2069,7 +2069,8 @@ static int uvc_register_terms(struct uvc_device *dev,
> > >  		stream = uvc_stream_by_id(dev, term->id);
> > >  		if (stream == NULL) {
> > >  			uvc_printk(KERN_INFO, "No streaming interface found "
> > > -				   "for terminal %u.", term->id);
> > > +				   "for terminal %u.",
> > > +				   term->id ? term->id : "(Unnamed)");
> > 
> > Have you tried compiling this ?
> 
> Yes. It does compile. Why?

Because term->id is a u8, "(Unnamed)" is a const char *, and %u requires
an integer. I'm surprised the compiler doesn't complain, but in any
case, it's not right :-)

-- 
Regards,

Laurent Pinchart