From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jisheng Zhang Subject: Re: [PATCH v1 6/7] net: mvneta: Don't use GRO on Armada 3720 Date: Thu, 9 Aug 2018 20:08:32 +0800 Message-ID: <20180809200832.7de1fa9e@xhacker.debian> References: <20180808152706.21727-1-marek.behun@nic.cz> <20180808152706.21727-7-marek.behun@nic.cz> <20180809124022.790f1784@xhacker.debian> <20180809192755.24e57782@xhacker.debian> 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 , Tomas Hlavacek , Russell King - ARM Linux , "David S. Miller" To: Marek =?UTF-8?B?QmVow7pu?= , Thomas Petazzoni , Andrew Lunn Return-path: Received: from mail-sn1nam02on0070.outbound.protection.outlook.com ([104.47.36.70]:21280 "EHLO NAM02-SN1-obe.outbound.protection.outlook.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1727768AbeHIOgE (ORCPT ); Thu, 9 Aug 2018 10:36:04 -0400 In-Reply-To: <20180809192755.24e57782@xhacker.debian> Sender: netdev-owner@vger.kernel.org List-ID: On Thu, 9 Aug 2018 19:27:55 +0800 Jisheng Zhang wrote: > Hi, >=20 > On Thu, 9 Aug 2018 12:40:41 +0800 Jisheng Zhang wrote: >=20 > > + more people > >=20 > > On Wed, 8 Aug 2018 17:27:05 +0200 Marek Beh=C3=BAn wrote: > > =20 > > > 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. =20 > >=20 > > We also met this issue on some berlin platforms. I tried to fix the bug, > > but no clue so far. > > =20 > > >=20 > > > This is fixed when using netif_receive_skb instead of napi_gro_receiv= e. =20 > >=20 > > This is a workaround. The good news is this workaround also fixes the i= ssue > > we saw on berlin. > > =20 > > >=20 > > > Signed-off-by: Marek Behun > > > Cc: Russell King - ARM Linux > > > Cc: netdev@vger.kernel.org > > >=20 > > > diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethe= rnet/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 >=20 > I think I found the root cause, if neta_armada3700 is true, the port got = from > this_cpu_ptr(pp->ports) is invalid, this is bug... I'll cook a patch for = this correct it as: the port's(port is got from this_cpu_ptr(pp->ports) napi is invalid. Patch is sent out. Could you please try? Per my test, it solves the issue we saw on berlin. >=20 > Thanks >=20 > > > =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) { =20 > > =20 >=20