From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andi Kleen Subject: Re: [PATCH 3/4] [XFRM]: Kill some bloat Date: Tue, 8 Jan 2008 03:05:29 +0100 Message-ID: <20080108020529.GC16156@one.firstfloor.org> References: <20080105.231658.168081302.davem@davemloft.net> <20080107.175458.127194310.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: andi@firstfloor.org, herbert@gondor.apana.org.au, ilpo.jarvinen@helsinki.fi, netdev@vger.kernel.org, acme@redhat.com, paul.moore@hp.com, latten@us.ibm.com To: David Miller Return-path: Received: from one.firstfloor.org ([213.235.205.2]:50704 "EHLO one.firstfloor.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752407AbYAHCDJ (ORCPT ); Mon, 7 Jan 2008 21:03:09 -0500 Content-Disposition: inline In-Reply-To: <20080107.175458.127194310.davem@davemloft.net> Sender: netdev-owner@vger.kernel.org List-ID: On Mon, Jan 07, 2008 at 05:54:58PM -0800, David Miller wrote: > From: Andi Kleen > Date: Tue, 08 Jan 2008 01:23:11 +0100 > > > David Miller writes: > > > > > Similarly I question just about any inline usage at all in *.c files > > > > Don't forget the .h files. Especially a lot of stuff in tcp.h should > > be probably in some .c file and not be inline. > > I explicitly left them out. > > Most of them are abstractions of common 2 or 3 instruction > calculations, and thus should stay inline. Definitely not in tcp.h. It has quite a lot of very long functions, of which very few really need to be inline: (AFAIK the only one where it makes really sense is tcp_set_state due to constant evaluation; although I never quite understood why the callers just didn't call explicit functions to do these actions) % awk ' { line++ } ; /^{/ { start = line } ; /^}/ { n++; r += line-start-2; } ; END { print r/n }' < include/net/tcp.h 9.48889 The average function length is 9 lines. -Andi