From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: [Patch net-next 3/7] inetpeer: use generic union inet_addr Date: Mon, 22 Jul 2013 19:26:11 -0700 Message-ID: <1374546371.4990.113.camel@edumazet-glaptop> References: <1374476713-8838-1-git-send-email-amwang@redhat.com> <1374476713-8838-4-git-send-email-amwang@redhat.com> <1374506315.4990.14.camel@edumazet-glaptop> <1374545131.24933.4.camel@cr0> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org, "David S. Miller" To: Cong Wang Return-path: Received: from mail-ob0-f172.google.com ([209.85.214.172]:58692 "EHLO mail-ob0-f172.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751343Ab3GWC0O (ORCPT ); Mon, 22 Jul 2013 22:26:14 -0400 Received: by mail-ob0-f172.google.com with SMTP id wo10so9029036obc.31 for ; Mon, 22 Jul 2013 19:26:13 -0700 (PDT) In-Reply-To: <1374545131.24933.4.camel@cr0> Sender: netdev-owner@vger.kernel.org List-ID: On Tue, 2013-07-23 at 10:05 +0800, Cong Wang wrote: > On Mon, 2013-07-22 at 08:18 -0700, Eric Dumazet wrote: > > On Mon, 2013-07-22 at 15:05 +0800, Cong Wang wrote: > > > From: Cong Wang > > > > > > struct inetpeer_addr is pretty similar to generic union inet_addr, > > > therefore can be safely converted to it. > > > > Its 'safe' but adds 50% increase for struct tcp_metrics_block > > > > I fail to see this mentioned in the changelog. > > I asked you in RFC, but you don't give me any response, this is the > reason. :) > I mentioned an increase in size of the structures, you replied : "rearrange the fields of struct inet_peer in case of cacheline miss?" which was a complete different matter, I had other more urgent work I guess. > > I guess its no big deal, but why do you think this code used hand coded > > functions instead of generic ? > > > > > > > > I don't understand what you are asking here, seems totally unrelated > with the point you raised above, therefore I am completely confused... You are sending a patch with possible performance impact, and you only says "its safe", without any real study. It might have an impact, who knows. I said "its no big deal", so consider this as an informative message.