From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1762753AbXFASEN (ORCPT ); Fri, 1 Jun 2007 14:04:13 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1760530AbXFASD7 (ORCPT ); Fri, 1 Jun 2007 14:03:59 -0400 Received: from bc.sympatico.ca ([209.226.175.184]:61052 "EHLO tomts22-srv.bellnexxia.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1760879AbXFASD6 (ORCPT ); Fri, 1 Jun 2007 14:03:58 -0400 Date: Fri, 1 Jun 2007 14:03:56 -0400 From: Mathieu Desnoyers To: Matt Mackall Cc: Andi Kleen , akpm@linux-foundation.org, linux-kernel@vger.kernel.org Subject: Re: [patch 2/9] Conditional Calls - Hash Table Message-ID: <20070601180356.GA7641@Krystal> References: <20070530140025.917261793@polymtl.ca> <20070530140227.398040643@polymtl.ca> <20070601160802.GL11166@waste.org> <20070601164623.GB4112@Krystal> <20070601170739.GA11115@waste.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline In-Reply-To: <20070601170739.GA11115@waste.org> X-Editor: vi X-Info: http://krystal.dyndns.org:8080 X-Operating-System: Linux/2.6.21.3-grsec (i686) X-Uptime: 14:00:51 up 4 days, 2:39, 2 users, load average: 0.83, 0.41, 0.44 User-Agent: Mutt/1.5.13 (2006-08-11) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org * Matt Mackall (mpm@selenic.com) wrote: > On Fri, Jun 01, 2007 at 12:46:23PM -0400, Mathieu Desnoyers wrote: > > * Matt Mackall (mpm@selenic.com) wrote: > > > On Thu, May 31, 2007 at 03:42:50PM +0200, Andi Kleen wrote: > > > > Mathieu Desnoyers writes: > > > > > > > > > Reimplementation of the cond calls which uses a hash table to hold the active > > > > > cond_calls. It permits to first arm a cond_call and then load supplementary > > > > > modules that contain this cond_call. > > > > > > > > Hash table is probably overkill. This is a very very slow path operation. > > > > Can you simplify the code? Just a linked list of all the condcall segments > > > > should be enough and then walk it. > > > > > > I think it could be greatly simplified by using symbols instead of > > > strings. > > > > > > That is, doing cond_call(foo, func()) rather than cond_call("foo", > > > func()). Here foo is a structure or type holding the relevant info to > > > deal with the cond_call infrastructure. For unoptimized architectures, > > > it can simply be a bool, which will be faster. > > > > > > This has the added advantage that the compiler will automatically pick > > > up any misspellings of these things. And it saves the space we'd use > > > on the hash table too. > > > > > > > The idea is interesting, but does not fit the problem: AFAIK, it will > > not be possible to do multiple declarations of the same symbol, which is > > needed whenever we want to declare a cond_call() more than once or to > > embed it in an inline function. > > It's not clear to me why either of those things are necessary. An > example please? > Case where we want to declare the same cond_call multiple times : function_a(int var) { ... cond_call(profile_on, profile_hit(...)); ... } function_b(int var, int var2) { ... cond_call(profile_on, profile_hit(...)); ... } Case in inline function : static inline myinlinefct() { ... cond_call(profile_on, profile_hit(...)); ... } somefct() { ... myinlinefct(); ... myinlinefct(); ... } Those will result in multiple declarations of the cond_call. -- Mathieu Desnoyers Computer Engineering Ph.D. Student, Ecole Polytechnique de Montreal OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F BA06 3F25 A8FE 3BAE 9A68