Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <CAEAjamvq+puThrxfo80TOy=xgbQEQNT6xvxy4w6hP2O1By66uw@mail.gmail.com>
 <20200501070538.GB887524@kroah.com>
In-Reply-To: <20200501070538.GB887524@kroah.com>
From:   Kyungtae Kim <kt0755@gmail.com>
Date:   Wed, 6 May 2020 00:08:16 -0400
Message-ID: <CAEAjamsTt5E0NgauzFXYzN=GK2-y0tvgYx9B26kYAbBkDqfagg@mail.gmail.com>
Subject: Re: KASAN: slab-out-of-bounds Read in gadget_dev_desc_UDC_store
To:     Greg KH <gregkh@linuxfoundation.org>
Cc:     balbi@kernel.org, syzkaller <syzkaller@googlegroups.com>,
        USB list <linux-usb@vger.kernel.org>,
        LKML <linux-kernel@vger.kernel.org>,
        Dave Tian <dave.jing.tian@gmail.com>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Fri, May 01, 2020 at 09:05:38AM +0200, Greg KH wrote:
> On Thu, Apr 30, 2020 at 11:03:54PM -0400, Kyungtae Kim wrote:
> > We report a bug (in linux-5.6.8) found by FuzzUSB (a modified version
> > of syzkaller).
> >
> > This happened when the size of "name" buffer is smaller than that of
> > "page" buffer
> > (after function kstrdup executed at line 263).
> > I guess it comes from the "page" buffer containing 0 value in the middle.
> > So accessing the "name" buffer with "len" variable, which is used to
> > indicate the size of "page" buffer,
> > triggered memory access violation.
> > To fix, it may need to check the size of name buffer, and try to use
> > right index variable.
>
> Can you submit a patch for this as you have a reproducer to test the
> issue?
>
> thanks,
>
> greg k-h

I just submitted a patch after testing with the repro.

Regards,
Kyungtae