From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ezequiel Garcia Subject: Re: [PATCH v3 0/8] net: ethernet: marvell: Assorted fixes Date: Fri, 23 May 2014 17:01:27 -0300 Message-ID: <20140523200127.GA14290@arch.cereza> References: <1400800023-1562-1-git-send-email-ezequiel.garcia@free-electrons.com> <20140523.153616.2078296737930596480.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: netdev@vger.kernel.org, thomas.petazzoni@free-electrons.com, sebastian.hesselbarth@gmail.com, gregory.clement@free-electrons.com, alior@marvell.com, tawfik@marvell.com, andrew@lunn.ch To: David Miller Return-path: Received: from top.free-electrons.com ([176.31.233.9]:56865 "EHLO mail.free-electrons.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751113AbaEWUCI (ORCPT ); Fri, 23 May 2014 16:02:08 -0400 Content-Disposition: inline In-Reply-To: <20140523.153616.2078296737930596480.davem@davemloft.net> Sender: netdev-owner@vger.kernel.org List-ID: On 23 May 03:36 PM, David Miller wrote: > From: Ezequiel Garcia > Date: Thu, 22 May 2014 20:06:55 -0300 >=20 > > New round for this assorted fixes and clean-up series. There is mor= e room for > > clean-ups, and I'll start preparing more patches once these are acc= epted. > >=20 > > This series consists of cleanups and minor improvements on mvneta, = mv643xx_eth > > and mvmdio drivers. None of the patches imply any functionality cha= nge, except > > for the patch six "Change the number of default rx queues to one". > >=20 > > This patch reduces the driver's allocated resources and makes the m= ultiqueue > > path in the poll function not get taken. The previous patchset cont= ains more > > details: > >=20 > > http://permalink.gmane.org/gmane.linux.network/315015 > >=20 > > As usual, any feedback on this will be well received! >=20 > Series applied to net-next, but some comments: >=20 > 1) Never use web sites to reference informative material, instead put= it > all explicitly into the commit message in your posting. >=20 > Web sites and archives come and go, the commit message is forever = and > readily available to someone studying your changes. >=20 OK, I'll keep this in mind. > 2) Just get rid of that MVNETA_RX_POLL_WEIGHT and simply use the > generic default NAPI_POLL_WEIGHT which has the same value. >=20 > Any local NAPI weight selection must be done for a good reason and > be fully documented with performance metrics et al. >=20 OK. > 3) I know you inherited this, but the MTU change failure behavior is > not so great. If the system is low on memory it could fail and lo= sing > an interface completely during an MTU change is extremely poor beh= avior. >=20 > I would suggest holding onto the old allocated TXQ resources, and > restoring them if the allocation of the new ones fail. >=20 > That way the MTU change itself would fail, but the interface would > remain up with the original MTU. >=20 > Please make sure that you properly revert the MTU adjustment in > this failure path. >=20 Oh, I see. Thanks for the feedback, I'll take a look at this next week. --=20 Ezequiel Garc=EDa, Free Electrons Embedded Linux, Kernel and Android Engineering http://free-electrons.com