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: <20210124073955.728797-1-christophe.jaillet@wanadoo.fr> <fe72d1d2-8222-5cda-74ff-9327725f196a@web.de>
In-Reply-To: <fe72d1d2-8222-5cda-74ff-9327725f196a@web.de>
From:   Richard Weinberger <richard.weinberger@gmail.com>
Date:   Sun, 24 Jan 2021 21:14:32 +0100
Message-ID: <CAFLxGvzFDPBGY1uoB7vqJreOSfs9N2PGMDTbQ24PYBWBRVR4oQ@mail.gmail.com>
Subject: Re: [PATCH] mtd: rawnand: Fix an error handling path in 'ebu_dma_start()'
To:     Christophe Jaillet <christophe.jaillet@wanadoo.fr>
Cc:     linux-mtd@lists.infradead.org,
        Vignesh Raghavendra <vigneshr@ti.com>,
        Martin Blumenstingl <martin.blumenstingl@googlemail.com>,
        Richard Weinberger <richard@nod.at>,
        kernel-janitors@vger.kernel.org,
        LKML <linux-kernel@vger.kernel.org>,
        Ramuthevar Vadivel Murugan 
        <vadivel.muruganx.ramuthevar@linux.intel.com>,
        Miquel Raynal <miquel.raynal@bootlin.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Sun, Jan 24, 2021 at 9:13 PM Markus Elfring <Markus.Elfring@web.de> wrot=
e:
>
> > If 'dmaengine_prep_slave_single()' fails, we must undo a previous
> > 'dma_map_single()' call, as already done in all the other error handlin=
g
> > paths of this function.
>
> Would you ever like to use an imperative wording for the change descripti=
on?
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/D=
ocumentation/process/submitting-patches.rst?id=3Dfdbc80bdc4365078a0f7d65631=
171cb80e3ffd6e#n89
>
>
> =E2=80=A6
> > +++ b/drivers/mtd/nand/raw/intel-nand-controller.c
> > @@ -318,8 +318,10 @@ static int ebu_dma_start(struct ebu_nand_controlle=
r *ebu_host, u32 dir,
> >       }
> >
> >       tx =3D dmaengine_prep_slave_single(chan, buf_dma, len, dir, flags=
);
> > -     if (!tx)
> > -             return -ENXIO;
> > +     if (!tx) {
> > +             ret =3D -ENXIO;
> > +             goto err_unmap;
> > +     }
> >
> >       tx->callback =3D callback;
> =E2=80=A6
>
> By the way:
> Can it be nicer to achieve the statement =E2=80=9Cret =3D -EIO;=E2=80=9D =
by a jump for
> a target like =E2=80=9Ce_io=E2=80=9D so that less exception handling code=
 would be duplicated
> for this function implementation?

Please feel free to ignore Markus.

https://lore.kernel.org/lkml/X+x3pIanr18Ep4ga@kroah.com/

--=20
Thanks,
//richard