Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1030285AbWISSLF (ORCPT ); Tue, 19 Sep 2006 14:11:05 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1030292AbWISSLF (ORCPT ); Tue, 19 Sep 2006 14:11:05 -0400 Received: from tomts16.bellnexxia.net ([209.226.175.4]:22190 "EHLO tomts16-srv.bellnexxia.net") by vger.kernel.org with ESMTP id S1030285AbWISSLC (ORCPT ); Tue, 19 Sep 2006 14:11:02 -0400 Date: Tue, 19 Sep 2006 14:11:01 -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: <20060919181101.GE26339@Krystal> References: <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> <20060919175405.GC26339@Krystal> <4510308A.1070401@google.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline In-Reply-To: <4510308A.1070401@google.com> X-Editor: vi X-Info: http://krystal.dyndns.org:8080 X-Operating-System: Linux/2.4.32-grsec (i686) X-Uptime: 14:07:23 up 27 days, 15:16, 4 users, load average: 0.26, 0.27, 0.23 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: 1746 Lines: 43 * Martin Bligh (mbligh@google.com) wrote: > Mathieu Desnoyers wrote: > >* Martin Bligh (mbligh@google.com) wrote: > > > >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 not in 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. > > Ugh, yes that's somewhat problematic. It does seem rather unlikely that > there's a function call in the function prologue when we're busy > offloading stuff onto the stack, but still ... > A function call is not the cause of the problem : an interrupt/trap is. > For the cases where we're prepared to overwrite the call instruction in > the caller, rather than insert an extra jump in the callee, can we not > do that atomically by overwriting the address we're jumping to (the > call is obviously there already)? Doesn't fix function pointers, etc, > but might work well for the simple case at least. > I don't think we have any guarantee that the function pointer in the call is aligned, so I guess it would not be an atomic replacement. 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/