From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff Garzik Subject: Re: [PATCH 1/9] Add TSO support Date: Tue, 13 May 2008 01:17:46 -0400 Message-ID: <4829247A.5020407@pobox.com> References: <20080507114121.GL2217@solarflare.com> <20080507115112.GM2217@solarflare.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org, linux-net-drivers@solarflare.com To: Ben Hutchings Return-path: Received: from srv5.dvmed.net ([207.36.208.214]:49365 "EHLO mail.dvmed.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752020AbYEMFRs (ORCPT ); Tue, 13 May 2008 01:17:48 -0400 In-Reply-To: <20080507115112.GM2217@solarflare.com> Sender: netdev-owner@vger.kernel.org List-ID: Ben Hutchings wrote: > The SFC4000 controller does not have hardware support for TSO, and the > core GSO code incurs a high cost in allocating and freeing skbs. This > TSO implementation uses lightweight packet header structures and is > substantially faster. > > Signed-off-by: Ben Hutchings > --- > drivers/net/sfc/efx.c | 4 +- > drivers/net/sfc/ethtool.c | 27 ++ > drivers/net/sfc/net_driver.h | 14 + > drivers/net/sfc/tx.c | 664 ++++++++++++++++++++++++++++++++++++++++++ > 4 files changed, 708 insertions(+), 1 deletions(-) applied 1-9, with two comments: 1) your subject line is missing a useful prefix. see http://linux.yyz.us/patch-format.html 2) normally TSO addition would not be permitting during the bug-fixes-only phase of the kernel development (2.6.26-rc is current), but since the driver is new, it is reasonable because there can be no regressions