MIME-Version: 1.0
References: <Y/yxGMvBbMGiehC6@lore-desk> <20230227135241.947052-1-void0red@gmail.com>
 <Y/y5Asxw3T3m4jCw@lore-desk>
In-Reply-To: <Y/y5Asxw3T3m4jCw@lore-desk>
From:   Kang Chen <void0red@gmail.com>
Date:   Mon, 27 Feb 2023 22:28:55 +0800
Message-ID: <CANE+tVq=djTShowm1o86184c34QYEax0_4Ls=y0_qFL=wo+EsQ@mail.gmail.com>
Subject: Re: [PATCH v2] wifi: mt76: add a check of vzalloc in mt7615_coredump_work
To:     Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Cc:     angelogioacchino.delregno@collabora.com, davem@davemloft.net,
        edumazet@google.com, kuba@kernel.org, kvalo@kernel.org,
        linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org,
        linux-mediatek@lists.infradead.org, linux-wireless@vger.kernel.org,
        lorenzo@kernel.org, matthias.bgg@gmail.com, nbd@nbd.name,
        netdev@vger.kernel.org, pabeni@redhat.com, ryder.lee@mediatek.com,
        sean.wang@mediatek.com, shayne.chen@mediatek.com
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Hi, Lorenzo

Thanks for your suggestions.
I totally agree with you.

Best regards,
Kang Chen

On Mon, Feb 27, 2023 at 10:07=E2=80=AFPM Lorenzo Bianconi
<lorenzo.bianconi@redhat.com> wrote:
>
> > From: Kang Chen <void0red@gmail.com>
> >
> > vzalloc may fails, dump might be null and will cause
> > illegal address access later.
> >
> > Fixes: d2bf7959d9c0 ("mt76: mt7663: introduce coredump support")
> > Signed-off-by: Kang Chen <void0red@gmail.com>
> > ---
> > v2 -> v1: add Fixes tag
> >
> >  drivers/net/wireless/mediatek/mt76/mt7615/mac.c | 3 +++
> >  1 file changed, 3 insertions(+)
> >
> > diff --git a/drivers/net/wireless/mediatek/mt76/mt7615/mac.c b/drivers/=
net/wireless/mediatek/mt76/mt7615/mac.c
> > index a95602473..73d84c301 100644
> > --- a/drivers/net/wireless/mediatek/mt76/mt7615/mac.c
> > +++ b/drivers/net/wireless/mediatek/mt76/mt7615/mac.c
> > @@ -2367,6 +2367,9 @@ void mt7615_coredump_work(struct work_struct *wor=
k)
> >       }
> >
> >       dump =3D vzalloc(MT76_CONNAC_COREDUMP_SZ);
> > +     if (!dump)
> > +             return;
> > +
> >       data =3D dump;
> >
> >       while (true) {
> > --
> > 2.34.1
> >
>
> revieweing the code I guess the right approach would be the one used in
> mt7921_coredump_work():
> - free pending skbs
> - not run dev_coredumpv()
>
> What do you think?
>
> Regards,
> Lorenzo