Received-SPF: softfail (google.com: domain of transitioning linux-kernel-owner@vger.kernel.org does not designate 23.128.96.19 as permitted sender) client-ip=23.128.96.19;
MIME-Version: 1.0
References: <20220506053740.1113415-1-dzm91@hust.edu.cn> <nycvar.YFH.7.76.2205060852300.28985@cbobk.fhfr.pm>
In-Reply-To: <nycvar.YFH.7.76.2205060852300.28985@cbobk.fhfr.pm>
From:   Dongliang Mu <mudongliangabcd@gmail.com>
Date:   Fri, 6 May 2022 15:15:16 +0800
Message-ID: <CAD-N9QUW_FHFAkDq8AYAEXdOL_6wvo0sNGt0Db4dfDRiKaaNCg@mail.gmail.com>
Subject: Re: [PATCH] HID: bigben: fix slab-out-of-bounds Write in bigben_probe
To:     Jiri Kosina <jikos@kernel.org>
Cc:     Dongliang Mu <dzm91@hust.edu.cn>,
        Benjamin Tissoires <benjamin.tissoires@redhat.com>,
        syzkaller <syzkaller@googlegroups.com>,
        linux-input@vger.kernel.org,
        linux-kernel <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Fri, May 6, 2022 at 2:53 PM Jiri Kosina <jikos@kernel.org> wrote:
>
> On Fri, 6 May 2022, Dongliang Mu wrote:
>
> > From: Dongliang Mu <mudongliangabcd@gmail.com>
> >
> > There is a slab-out-of-bounds Write bug in hid-bigbenff driver.
> > The problem is the driver assumes the device must have an input but
> > some malicious devices violate this assumption.
> >
> > Fix this by checking hid_device's input is non-empty before its usage.
> >
> > Reported-by: syzkaller <syzkaller@googlegroups.com>
> > Signed-off-by: Dongliang Mu <mudongliangabcd@gmail.com>
> > ---
> >  drivers/hid/hid-bigbenff.c | 5 +++++
> >  1 file changed, 5 insertions(+)
> >
> > diff --git a/drivers/hid/hid-bigbenff.c b/drivers/hid/hid-bigbenff.c
> > index 74ad8bf98bfd..c14d1774101d 100644
> > --- a/drivers/hid/hid-bigbenff.c
> > +++ b/drivers/hid/hid-bigbenff.c
> > @@ -347,6 +347,11 @@ static int bigben_probe(struct hid_device *hid,
> >       bigben->report = list_entry(report_list->next,
> >               struct hid_report, list);
> >
> > +     if (list_empty(&hid->inputs)) {
> > +             hid_err(hid, "no inputs found\n");
> > +             return -ENODEV;
> > +     }
> > +
>
> Thanks for the fix. It doesn't seemt o be fully correct though -- as you'd
> be returning -ENODEV here in the situation when hid_hw_start() has already
> happened. So I believe better thing to do here is to do error = -ENODEV;
> goto error_hw_stop;
>
> Could you please fix that up and resend? Thanks,

Oh, yes. I will send a v2 version ASAP.
>
> --
> Jiri Kosina
> SUSE Labs
>