MIME-Version: 1.0
References: <20211123115333.624335-1-mailhol.vincent@wanadoo.fr>
 <20211123115333.624335-2-mailhol.vincent@wanadoo.fr> <aafec053-1054-4797-e1f1-e89586fe326f@hartkopp.net>
In-Reply-To: <aafec053-1054-4797-e1f1-e89586fe326f@hartkopp.net>
From:   Vincent MAILHOL <mailhol.vincent@wanadoo.fr>
Date:   Wed, 24 Nov 2021 08:21:26 +0900
Message-ID: <CAMZ6RqKn4PjYsE+MJHM36KPjcJX_RzW6cLtxyc94DWFz__AuQA@mail.gmail.com>
Subject: Re: [PATCH v1 1/2] can: do not increase rx statistics when receiving
 CAN error frames
To:     Oliver Hartkopp <socketcan@hartkopp.net>
Cc:     Marc Kleine-Budde <mkl@pengutronix.de>, linux-can@vger.kernel.org,
        netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
        linux-arm-kernel@lists.infradead.org, linux-sunxi@lists.linux.dev,
        Jimmy Assarsson <extja@kvaser.com>,
        Nicolas Ferre <nicolas.ferre@microchip.com>,
        Alexandre Belloni <alexandre.belloni@bootlin.com>,
        Ludovic Desroches <ludovic.desroches@microchip.com>,
        Chandrasekar Ramakrishnan <rcsekar@samsung.com>,
        Maxime Ripard <mripard@kernel.org>,
        Chen-Yu Tsai <wens@csie.org>,
        Jernej Skrabec <jernej.skrabec@gmail.com>,
        Appana Durga Kedareswara rao <appana.durga.rao@xilinx.com>,
        Naga Sureshkumar Relli <naga.sureshkumar.relli@xilinx.com>,
        Michal Simek <michal.simek@xilinx.com>,
        Stephane Grosjean <s.grosjean@peak-system.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 8BIT
Precedence: bulk

On Wed. 24 Nov. 2021 à 06:01, Oliver Hartkopp <socketcan@hartkopp.net> wrote:
> On 23.11.21 12:53, Vincent Mailhol wrote:
> > CAN error skb is an interface specific to socket CAN. The CAN error
> > skb does not correspond to any actual CAN frame sent on the wire. Only
> > an error flag and a delimiter are transmitted when an error occurs
> > (c.f. ISO 11898-1 section 10.4.4.2 "Error flag").
> >
> > For this reason, it makes no sense to increment the rx_packets and
> > rx_bytes fields of struct net_device_stats because no actual payload
> > were transmitted on the wire.
> >
>
> (..)
>
> > diff --git a/drivers/net/can/dev/rx-offload.c b/drivers/net/can/dev/rx-offload.c
> > index 37b0cc65237b..bb47e9a49240 100644
> > --- a/drivers/net/can/dev/rx-offload.c
> > +++ b/drivers/net/can/dev/rx-offload.c
> > @@ -54,8 +54,10 @@ static int can_rx_offload_napi_poll(struct napi_struct *napi, int quota)
> >               struct can_frame *cf = (struct can_frame *)skb->data;
> >
> >               work_done++;
> > -             stats->rx_packets++;
> > -             stats->rx_bytes += cf->len;
> > +             if (!(cf->can_id & CAN_ERR_MASK)) {
>
> This looks wrong.

And it is.

> Did you think of CAN_ERR_FLAG ??

Yes, I will fix this in v2.

> > +                     stats->rx_packets++;
> > +                     stats->rx_bytes += cf->len;
> > +             }
> >               netif_receive_skb(skb);
>
> (..)
>
> > diff --git a/drivers/net/can/usb/ucan.c b/drivers/net/can/usb/ucan.c
> > index 1679cbe45ded..d582c39fc8d0 100644
> > --- a/drivers/net/can/usb/ucan.c
> > +++ b/drivers/net/can/usb/ucan.c
> > @@ -621,8 +621,10 @@ static void ucan_rx_can_msg(struct ucan_priv *up, struct ucan_message_in *m)
> >               memcpy(cf->data, m->msg.can_msg.data, cf->len);
> >
> >       /* don't count error frames as real packets */
> > -     stats->rx_packets++;
> > -     stats->rx_bytes += cf->len;
> > +     if (!(cf->can_id & CAN_ERR_FLAG)) {
>
> Ah, here we are :-)

Thank you!