MIME-Version: 1.0
References: <20211222094558.2098791-1-jsd@semihalf.com> <20211222094558.2098791-2-jsd@semihalf.com>
 <YcSaqmlI7l4H1lsN@smile.fi.intel.com>
In-Reply-To: <YcSaqmlI7l4H1lsN@smile.fi.intel.com>
From:   =?UTF-8?B?SmFuIETEhWJyb8Wb?= <jsd@semihalf.com>
Date:   Mon, 27 Dec 2021 08:17:34 +0100
Message-ID: <CAOtMz3NTLXYiFg6xYbpVU2rd=YWGCUhG2ekFqxHwFF0XFA6orQ@mail.gmail.com>
Subject: Re: [RFC 1/2] i2c: designware: Add missing locks
To:     Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc:     Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        linux-i2c <linux-i2c@vger.kernel.org>,
        Jarkko Nikula <jarkko.nikula@linux.intel.com>,
        Mika Westerberg <mika.westerberg@linux.intel.com>,
        Wolfram Sang <wsa@kernel.org>,
        Raul E Rangel <rrangel@chromium.org>,
        Marcin Wojtas <mw@semihalf.com>,
        Grzegorz Jaszczyk <jaz@semihalf.com>, upstream@semihalf.com
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

czw., 23 gru 2021 o 16:51 Andy Shevchenko
<andriy.shevchenko@linux.intel.com> napisa=C5=82(a):
>
> On Wed, Dec 22, 2021 at 10:45:57AM +0100, Jan Dabros wrote:
> > All accesses to controller's registers should be protected on
> > probe, disable and xfer paths. This is needed for i2c bus controllers
> > that are shared with but not controlled by kernel.
>
> ...
>
> > +     ret =3D i2c_dw_acquire_lock(dev);
> > +     if (ret)
> > +             return ret;
> > +
> >       ret =3D regmap_read(dev->map, DW_IC_COMP_PARAM_1, &param);
> >       if (ret)
> >               return ret;
> > +     i2c_dw_release_lock(dev);
>
> Not sure this part is fully correct. Please, fix the leakage.

Correct, I will move release() right below regmap_read() call, before
checking return code.

>
> ...
>
> > +     ret =3D i2c_dw_acquire_lock(dev);
> > +     if (ret)
> > +             return;
> >
> >       /* Disable controller */
> >       __i2c_dw_disable(dev);
> > @@ -614,6 +624,8 @@ void i2c_dw_disable(struct dw_i2c_dev *dev)
> >       /* Disable all interrupts */
> >       regmap_write(dev->map, DW_IC_INTR_MASK, 0);
> >       regmap_read(dev->map, DW_IC_CLR_INTR, &dummy);
>
> > +     i2c_dw_release_lock(dev);
>
> Not enough context here, bu I believe there is the same issue(s).

Since we are ignoring values returned by regmap_write()/read() and
also __i2c_dw_disable() returns void I don't see a leakage here. Don't
have more function calls here.