From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH 1/7] [NET]: uninline skb_put, de-bloats a lot Date: Thu, 27 Mar 2008 17:04:54 -0700 (PDT) Message-ID: <20080327.170454.124353224.davem@davemloft.net> References: <1206645050.4849.77.camel@localhost> <20080327.150456.39560267.davem@davemloft.net> <1206662557.4849.112.camel@localhost> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: ilpo.jarvinen@helsinki.fi, akpm@linux-foundation.org, netdev@vger.kernel.org, linux-kernel@vger.kernel.org, acme@redhat.com, mpm@selenic.com To: joe@perches.com Return-path: Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:36728 "EHLO sunset.davemloft.net" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1753625AbYC1AEy (ORCPT ); Thu, 27 Mar 2008 20:04:54 -0400 In-Reply-To: <1206662557.4849.112.camel@localhost> Sender: netdev-owner@vger.kernel.org List-ID: From: Joe Perches Date: Thu, 27 Mar 2008 17:02:37 -0700 > I think this is the sort of change you might have argued > against accepting without performance testing not long ago. Look at how complicated the assembler of this function is. Any inlining performance gains will be trumped by the instruction cache usage bloat.