Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751899AbWISRyJ (ORCPT ); Tue, 19 Sep 2006 13:54:09 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751900AbWISRyJ (ORCPT ); Tue, 19 Sep 2006 13:54:09 -0400 Received: from tomts13-srv.bellnexxia.net ([209.226.175.34]:14516 "EHLO tomts13-srv.bellnexxia.net") by vger.kernel.org with ESMTP id S1751899AbWISRyH (ORCPT ); Tue, 19 Sep 2006 13:54:07 -0400 Date: Tue, 19 Sep 2006 13:54:05 -0400 From: Mathieu Desnoyers To: Martin Bligh Cc: prasanna@in.ibm.com, Andrew Morton , "Frank Ch. Eigler" , Ingo Molnar , Paul Mundt , linux-kernel , Jes Sorensen , Tom Zanussi , Richard J Moore , Michel Dagenais , Christoph Hellwig , Greg Kroah-Hartman , Thomas Gleixner , William Cohen , ltt-dev@shafik.org, systemtap@sources.redhat.com, Alan Cox Subject: Re: [PATCH] Linux Kernel Markers Message-ID: <20060919175405.GC26339@Krystal> References: <20060918234502.GA197@Krystal> <20060919081124.GA30394@elte.hu> <451008AC.6030006@google.com> <20060919154612.GU3951@redhat.com> <4510151B.5070304@google.com> <20060919093935.4ddcefc3.akpm@osdl.org> <45101DBA.7000901@google.com> <20060919063821.GB23836@in.ibm.com> <45102641.7000101@google.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline In-Reply-To: <45102641.7000101@google.com> X-Editor: vi X-Info: http://krystal.dyndns.org:8080 X-Operating-System: Linux/2.4.32-grsec (i686) X-Uptime: 13:50:16 up 27 days, 14:58, 4 users, load average: 0.41, 0.35, 0.28 User-Agent: Mutt/1.5.13 (2006-08-11) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1477 Lines: 37 * Martin Bligh (mbligh@google.com) wrote: > How about we combine all three ideas together ... > > 1. Load modified copy of the function in question. > 2. overwrite the first instruction of the routine with an int3 that > does what you say (atomically) > 3. Then overwrite the second instruction with a jump that's faster > 4. Now atomically overwrite the int3 with a nop, and let the jump > take over. > Very good idea.. However, overwriting the second instruction with a jump could be dangerous on preemptible and SMP kernels, because we never know if a thread has an IP in any of its contexts that would return exactly at the middle of the jump. I think it would be doable to overwrite a 5+ bytes instruction with a NOP non-atomically in all cases, but as the instructions nin the prologue seems to be smaller : prologue on x86 0: 55 push %ebp 1: 89 e5 mov %esp,%ebp epilogue on x86 3: 5d pop %ebp 4: c3 ret Then is can be a problem. Ideas are welcome. Mathieu OpenPGP public key: http://krystal.dyndns.org:8080/key/compudj.gpg Key fingerprint: 8CD5 52C3 8E3C 4140 715F BA06 3F25 A8FE 3BAE 9A68 - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/