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: <20200821075452.1543610-1-enric.balletbo@collabora.com>
 <99821f85-9a38-7591-f982-872603b6ce8a@collabora.com> <CAGp9Lzq-15xjvSVaEJ=2o18o-8ESGWsHf-LNAXXwKfxu4V-0NQ@mail.gmail.com>
 <CACRpkdYPoZX1+rfJb925_+H6YXiwO26cKLpZae=_j=RQKGA0Wg@mail.gmail.com> <CAGp9LzqGwvxGF_bhuyGXu1R2516x=twv1j5e0Wx0EQ7GZ4b-yQ@mail.gmail.com>
In-Reply-To: <CAGp9LzqGwvxGF_bhuyGXu1R2516x=twv1j5e0Wx0EQ7GZ4b-yQ@mail.gmail.com>
From:   Linus Walleij <linus.walleij@linaro.org>
Date:   Thu, 1 Oct 2020 09:58:14 +0200
Message-ID: <CACRpkdYga6XttmMnQ6r--_7wtfebirY5Z58D6xhBkRRDrf98ZQ@mail.gmail.com>
Subject: Re: [PATCH v2] pinctrl: mediatek: Free eint data on failure
To:     Sean Wang <sean.wang@kernel.org>
Cc:     Enric Balletbo i Serra <enric.balletbo@collabora.com>,
        lkml <linux-kernel@vger.kernel.org>,
        Collabora Kernel ML <kernel@collabora.com>,
        Matthias Brugger <matthias.bgg@gmail.com>,
        Nicolas Boichat <drinkcat@chromium.org>, hsinyi@chromium.org,
        linux-arm Mailing List <linux-arm-kernel@lists.infradead.org>,
        "open list:GPIO SUBSYSTEM" <linux-gpio@vger.kernel.org>,
        "moderated list:ARM/Mediatek SoC support" 
        <linux-mediatek@lists.infradead.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Wed, Sep 30, 2020 at 6:33 PM Sean Wang <sean.wang@kernel.org> wrote:
> On Wed, Sep 30, 2020 at 1:47 AM Linus Walleij <linus.walleij@linaro.org> wrote:
> > On Sun, Sep 27, 2020 at 7:57 PM Sean Wang <sean.wang@kernel.org> wrote:
> >
> > > v2 seems the same with v1 or I was missing something.
> > >
> > > I just thought we call devm_ioremap_release to explicitly to free
> > > resource when a certain failure occurs after
> > > devm_ioremap_resource?
> >
> > What is the semantics around mtk_build_eint()?
> >
>
> mtk_build_eint is to add external interrupt function to the
> corresponding bound pins.
> mtk pinctrl driver still can work (than means probe() successfully) to
> keep pinctrl functional even with there is an error in mtk_build_eint.
> So the patch is used to explicitly free those data on failure in
> mtk_build_eint to let unused data is being free:ed immediately.

OK then we need a v3 of this that will call *release
explicitly, indeed.

Thanks Sean!
Linus Walleij