MIME-Version: 1.0
References: <20230316180940.1601515-1-zyytlz.wz@163.com> <a83b2f5f-732c-6d71-3e0c-1dce76076817@linuxfoundation.org>
 <CAJedcCyb_XP=X=2XzjKJ-ay=mzPdEJEdRsAcU2KbeaQPjLaW3g@mail.gmail.com> <ZBNpf5X+y26+eKYf@kroah.com>
In-Reply-To: <ZBNpf5X+y26+eKYf@kroah.com>
From:   Zheng Hacker <hackerzheng666@gmail.com>
Date:   Fri, 17 Mar 2023 11:59:16 +0800
Message-ID: <CAJedcCwAU+piERJcthGTOY4+MKqjrGNYiNVAvTN5JQ4ugrjPjg@mail.gmail.com>
Subject: Re: [PATCH v2] usbip: vudc: Fix use after free bug in vudc_remove due
 to race condition
To:     "gregkh@linuxfoundation.org" <gregkh@linuxfoundation.org>
Cc:     Shuah Khan <skhan@linuxfoundation.org>,
        Zheng Wang <zyytlz.wz@163.com>,
        "valentina.manea.m@gmail.com" <valentina.manea.m@gmail.com>,
        "linux-usb@vger.kernel.org" <linux-usb@vger.kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "1395428693sheep@gmail.com" <1395428693sheep@gmail.com>,
        "alex000young@gmail.com" <alex000young@gmail.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

gregkh@linuxfoundation.org <gregkh@linuxfoundation.org> =E4=BA=8E2023=E5=B9=
=B43=E6=9C=8817=E6=97=A5=E5=91=A8=E4=BA=94 03:09=E5=86=99=E9=81=93=EF=BC=9A
>
> On Fri, Mar 17, 2023 at 02:27:05AM +0800, Zheng Hacker wrote:
> > =E5=9C=A8 2023=E5=B9=B43=E6=9C=8817=E6=97=A5=E6=98=9F=E6=9C=9F=E4=BA=94=
=EF=BC=8CShuah Khan <skhan@linuxfoundation.org> =E5=86=99=E9=81=93=EF=BC=9A
> >
> > > On 3/16/23 12:09, Zheng Wang wrote:
> > >
> > >> In vudc_probe, it calls init_vudc_hw, which bound &udc->timer with
> > >> v_timer.
> > >>
> > >> When it calls usbip_sockfd_store, it will call v_start_timer to star=
t the
> > >> timer work.
> > >>
> > >> When we call vudc_remove to remove the driver, theremay be a sequenc=
e as
> > >> follows:
> > >>
> > >> Fix it by shutdown the timer work before cleanup in vudc_remove.
> > >>
> > >> Note that removing a driver is a root-only operation, and should nev=
er
> > >> happen.
> > >>
> > >> CPU0                  CPU1
> > >>
> > >>                       |v_timer
> > >> vudc_remove          |
> > >> kfree(udc);          |
> > >> //free shost         |
> > >>                       |udc->gadget
> > >>                       |//use
> > >>
> > >> This bug was found by static analysis.
> > >>
> > >
> > > Tell me which static analysis tool did you use to find this and
> > > the output from the tool.
> >
> >
> >      This is found by codeql=EF=BC=8Cthe rule is complicated. It     fi=
nally found
> > there is no timer stop behavior in remove function.
>
> When using tools like this, you are required to follow the rules in
> Documentation/process/researcher-guidelines.rst
>
> Please do so here.
>

Thanks for your assistance. I'll read it carefully and keep that in mind.
And I'll append more information in the next version of patch.

Best regards,
Zheng

> thanks,
>
> greg k-h