Re: WARNING in snd_usbmidi_submit_urb/usb_submit_urb

[Andrey Konovalov]

On Mon, Jun 1, 2020 at 10:43 AM Greg KH <gregkh@xxxxxxxxxxxxxxxxxxx> wrote:
>
> On Mon, Jun 01, 2020 at 12:24:03AM -0700, syzbot wrote:
> > syzbot has bisected this bug to:
> >
> > commit f2c2e717642c66f7fe7e5dd69b2e8ff5849f4d10
> > Author: Andrey Konovalov <andreyknvl@xxxxxxxxxx>
> > Date:   Mon Feb 24 16:13:03 2020 +0000
> >
> >     usb: gadget: add raw-gadget interface
> >
> > bisection log:  https://syzkaller.appspot.com/x/bisect.txt?x=164afcf2100000
> > start commit:   bdc48fa1 checkpatch/coding-style: deprecate 80-column warn..
> > git tree:       upstream
> > final crash:    https://syzkaller.appspot.com/x/report.txt?x=154afcf2100000
> > console output: https://syzkaller.appspot.com/x/log.txt?x=114afcf2100000
> > kernel config:  https://syzkaller.appspot.com/x/.config?x=129ea1e5950835e5
> > dashboard link: https://syzkaller.appspot.com/bug?extid=5f1d24c49c1d2c427497
> > syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=12d70cf2100000
> >
> > Reported-by: syzbot+5f1d24c49c1d2c427497@xxxxxxxxxxxxxxxxxxxxxxxxx
> > Fixes: f2c2e717642c ("usb: gadget: add raw-gadget interface")
> >
> > For information about bisection process see: https://goo.gl/tpsmEJ#bisection
>
> So the tool that was used to create the bug has bisected the problem to
> the patch that adds the tool to the kernel to test for the issue?
>
> This feels wrong...

That's the expected result of bisection with the way it's intended to
work. We'll be getting more bisection results pointing to that commit
for old USB bugs. For new ones (that are introduced after raw-gadget),
the bisection should point to proper commits.