From mboxrd@z Thu Jan  1 00:00:00 1970
From: Jisheng Zhang <Jisheng.Zhang@synaptics.com>
Subject: Re: [PATCH v1 6/7] net: mvneta: Don't use GRO on Armada 3720
Date: Thu, 9 Aug 2018 12:40:41 +0800
Message-ID: <20180809124022.790f1784@xhacker.debian>
References: <20180808152706.21727-1-marek.behun@nic.cz>
        <20180808152706.21727-7-marek.behun@nic.cz>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Cc: linux-arm-kernel@lists.infradead.org, netdev@vger.kernel.org,
        Gregory CLEMENT <gregory.clement@bootlin.com>,
        Tomas Hlavacek <tomas.hlavacek@nic.cz>,
        Russell King - ARM Linux <linux@armlinux.org.uk>,
        "David S. Miller" <davem@davemloft.net>
To: Marek =?UTF-8?B?QmVow7pu?= <marek.behun@nic.cz>,
        Thomas Petazzoni <thomas.petazzoni@bootlin.com>,
        Andrew Lunn <andrew@lunn.ch>
Return-path: <netdev-owner@vger.kernel.org>
Received: from mail-dm3nam03on0089.outbound.protection.outlook.com ([104.47.41.89]:62937
        "EHLO NAM03-DM3-obe.outbound.protection.outlook.com"
        rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP
        id S1727211AbeHIHGe (ORCPT <rfc822;netdev@vger.kernel.org>);
        Thu, 9 Aug 2018 03:06:34 -0400
In-Reply-To: <20180808152706.21727-7-marek.behun@nic.cz>
Sender: netdev-owner@vger.kernel.org
List-ID: <netdev.vger.kernel.org>

+ more people

On Wed,  8 Aug 2018 17:27:05 +0200 Marek Beh=C3=BAn wrote:

> For some reason on Armada 3720 boards (EspressoBin and Turris Mox) the
> networking driver behaves weirdly when using napi_gro_receive.
>=20
> For example downloading a big file from a local network (low ping) is
> fast, but when downloading from a remote server (higher ping), the
> download speed is at first high but drops rapidly to almost nothing or
> absolutely nothing.

We also met this issue on some berlin platforms. I tried to fix the bug,
but no clue so far.

>=20
> This is fixed when using netif_receive_skb instead of napi_gro_receive.

This is a workaround. The good news is this workaround also fixes the issue
we saw on berlin.

>=20
> Signed-off-by: Marek Behun <marek.behun@nic.cz>
> Cc: Russell King - ARM Linux <linux@armlinux.org.uk>
> Cc: netdev@vger.kernel.org
>=20
> diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet=
/marvell/mvneta.c
> index 0ad2f3f7da85..27f3017d94c5 100644
> --- a/drivers/net/ethernet/marvell/mvneta.c
> +++ b/drivers/net/ethernet/marvell/mvneta.c
> @@ -1959,7 +1959,10 @@ static int mvneta_rx_swbm(struct mvneta_port *pp, =
int rx_todo,
> =20
>  			skb->protocol =3D eth_type_trans(skb, dev);
>  			mvneta_rx_csum(pp, rx_status, skb);
> -			napi_gro_receive(&port->napi, skb);
> +			if (pp->neta_armada3700)
> +				netif_receive_skb(skb);
> +			else
> +				napi_gro_receive(&port->napi, skb);
> =20
>  			rcvd_pkts++;
>  			rcvd_bytes +=3D rx_bytes;
> @@ -2001,7 +2004,10 @@ static int mvneta_rx_swbm(struct mvneta_port *pp, =
int rx_todo,
> =20
>  		mvneta_rx_csum(pp, rx_status, skb);
> =20
> -		napi_gro_receive(&port->napi, skb);
> +		if (pp->neta_armada3700)
> +			netif_receive_skb(skb);
> +		else
> +			napi_gro_receive(&port->napi, skb);
>  	}
> =20
>  	if (rcvd_pkts) {