From mboxrd@z Thu Jan 1 00:00:00 1970 From: Joe Perches Subject: Re: [PATCH net-next 01/14] etherdevice: Add eth__addr CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS code Date: Tue, 03 Mar 2015 12:04:28 -0800 Message-ID: <1425413068.17273.58.camel@perches.com> References: <1425409671.17273.42.camel@perches.com> <20150303.142758.822602589629075339.davem@davemloft.net> <1425411669.17273.52.camel@perches.com> <20150303.145815.1129211297511474723.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 7bit Cc: eric.dumazet@gmail.com, netdev@vger.kernel.org, linux-kernel@vger.kernel.org To: David Miller Return-path: In-Reply-To: <20150303.145815.1129211297511474723.davem@davemloft.net> Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On Tue, 2015-03-03 at 14:58 -0500, David Miller wrote: > From: Joe Perches > Date: Tue, 03 Mar 2015 11:41:09 -0800 > > > Until such time as the linux crosstools compilers are updated, > > (they seem stuck on 4.6.3 from 3 years ago) > > https://www.kernel.org/pub/tools/crosstool/ > > I think the series is a trivial, small improvement. > > You are even admitting that even for your specific case you are > using 3+ year old compiler technology. That's what I said. How is it an "admission"? Those are the kernel.org "recommended" cross-compilers. > These are slow paths, so correctness is the most important aspect. > > Converting these locations to consistently use eth_*() is fine, > but then "optimizing" a 6 by memset for configuration paths when > the compiler should be more than capable of doing this just fine > is not convincing at all to me. Swell, then perhaps you could simply skip 1/14.