From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 4315 invoked by alias); 2 May 2012 21:44:18 -0000 Received: (qmail 4257 invoked by uid 22791); 2 May 2012 21:44:14 -0000 X-SWARE-Spam-Status: No, hits=-2.8 required=5.0 tests=AWL,BAYES_00,KHOP_THREADED,T_RP_MATCHES_RCVD X-Spam-Check-By: sourceware.org Received: from sibelius.xs4all.nl (HELO glazunov.sibelius.xs4all.nl) (83.163.83.176) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Wed, 02 May 2012 21:43:43 +0000 Received: from glazunov.sibelius.xs4all.nl (kettenis@localhost [127.0.0.1]) by glazunov.sibelius.xs4all.nl (8.14.5/8.14.3) with ESMTP id q42LheGb009522; Wed, 2 May 2012 23:43:40 +0200 (CEST) Received: (from kettenis@localhost) by glazunov.sibelius.xs4all.nl (8.14.5/8.14.3/Submit) id q42Lhd3G018067; Wed, 2 May 2012 23:43:39 +0200 (CEST) Date: Wed, 02 May 2012 21:44:00 -0000 Message-Id: <201205022143.q42Lhd3G018067@glazunov.sibelius.xs4all.nl> From: Mark Kettenis To: hjl.tools@gmail.com CC: gdb-patches@sourceware.org In-reply-to: (hjl.tools@gmail.com) Subject: Re: PATCH: Also check for `movl %esp, %ebp' for x32 References: <20120410202953.GA23862@intel.com> <201204171143.q3HBh9LV021759@glazunov.sibelius.xs4all.nl> Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org X-SW-Source: 2012-05/txt/msg00035.txt.bz2 > Date: Sun, 29 Apr 2012 11:37:49 -0700 > From: "H.J. Lu" > > On Tue, Apr 24, 2012 at 9:27 AM, H.J. Lu wrote: > > On Tue, Apr 17, 2012 at 7:30 AM, H.J. Lu wrote: > >> On Tue, Apr 17, 2012 at 4:43 AM, Mark Kettenis wrote: > >>>> Date: Tue, 10 Apr 2012 13:29:53 -0700 > >>>> From: "H.J. Lu" > >>>> > >>>> Hi, > >>>> > >>>> X32 may use `movl %esp, %ebp' in prologue.  This patch checks it for > >>>> x32.  Tested on Linux/x86-64.  OK for trunk? > >>> > >>> Sorry, but I'm not sure it is a good idea to mix ABIs in the code like > >>> that.  Up until now, I've made a conscious attempt to keep the i386 > >>> and amd64 ABIs seperated out as much as possible.  Can you post a > >>> complete diff of the -tdep.c related changes to support x32 in GDB, > >>> such that I can judge where this is heading? > >> > >> Here is the complete x32 GDB patch: > >> > >> http://sourceware.org/ml/gdb-patches/2012-04/msg00476.html > >> > > > > Hi Mark, > > > > Have you looked at my change? > > > > Thanks. > > Ping. Sorry; been travelling too much lately... I did have a look at it, but still have some questions. > Hi, > > X32 may use `movl %esp, %ebp' in prologue. This patch checks it for > x32. Tested on Linux/x86-64. OK for trunk? But the prologues generated by various compilers are expected to be otherwise the same for both the x32 ABI and the normal 64-bit ABI? I guess x32 has to use "pushq %rbp" as "pushl %ebp" isn't available. And I guess you want to keep the stack 16-byte aligned anyway. I suppose that "movq %rsp, %rbp" is still ok for x32, but "movl %esp, %ebp" can be encoded in less bytes, so it might be a bit more efficient for x32. But what about the stack align code that we check for in amd64_analyze_stack_align()? Wouldn't that be different for x32 as well?