From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 18067 invoked by alias); 12 Mar 2012 13:14:36 -0000 Received: (qmail 18059 invoked by uid 22791); 12 Mar 2012 13:14:35 -0000 X-SWARE-Spam-Status: No, hits=-1.4 required=5.0 tests=AWL,BAYES_00,T_RP_MATCHES_RCVD X-Spam-Check-By: sourceware.org Received: from mail.allinea.com (HELO mail.allinea.com) (94.125.131.200) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Mon, 12 Mar 2012 13:14:22 +0000 Received: from localhost (localhost [127.0.0.1]) by mail.allinea.com (Postfix) with ESMTP id 5D3C412E204; Mon, 12 Mar 2012 13:14:21 +0000 (GMT) Received: from mail.allinea.com ([127.0.0.1]) by localhost (mail.allinea.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id v1sV9wsdPJIR; Mon, 12 Mar 2012 13:14:21 +0000 (GMT) Received: from [192.168.0.6] (cpc11-sotn9-2-0-cust158.15-1.cable.virginmedia.com [81.101.103.159]) (Authenticated sender: cjanuary) by mail.allinea.com (Postfix) with ESMTPSA id EDE4E12E202; Mon, 12 Mar 2012 13:14:20 +0000 (GMT) Message-ID: <1331558060.2783.20.camel@gumtree> Subject: Re: [PATCH] AIX: Add another way of setting up the frame pointer to skip_prologue From: Chris January To: Joel Brobecker Cc: gdb-patches@sourceware.org Date: Mon, 12 Mar 2012 13:14:00 -0000 In-Reply-To: <20120309194834.GG2853@adacore.com> References: <1331217602.2742.14.camel@gumtree> <20120309194834.GG2853@adacore.com> Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Mime-Version: 1.0 X-IsSubscribed: yes 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: 2012-03/txt/msg00394.txt.bz2 On Fri, 2012-03-09 at 11:48 -0800, Joel Brobecker wrote: > Chris, > Same comments as before: Looks good, but has it been validated against > the testsuite? No, I have not got the test suite working reliably on AIX yet. > I am just going to suggest one tiny change, which is a bit of a nitpick, > but I think makes the code relatively consistent: > I would put the comment outside of the closing parens. I know we do > this for multi-line conditions, but this one fits in a single line, > so we can put it outside, and that way, it doesn't look like a parens > is missing. Make sense to me, modified patch below. Chris --- 2012-03-08 Chris January * rs6000-tdep.c (skip_prologue): Support the oril r29, r1, 0x0 instruction. --- diff --git a/gdb/rs6000-tdep.c b/gdb/rs6000-tdep.c index a8ff777..9d4fd40 100644 --- a/gdb/rs6000-tdep.c +++ b/gdb/rs6000-tdep.c @@ -1822,6 +1822,15 @@ skip_prologue (struct gdbarch *gdbarch, CORE_ADDR pc, CORE_ADDR lim_pc, /* Set up frame pointer */ } + else if (op == 0x603d0000) /* oril r29, r1, 0x0 */ + { + fdata->frameless = 0; + framep = 1; + fdata->alloca_reg = (tdep->ppc_gp0_regnum + 29); + continue; + + /* Another way to set up the frame pointer. */ + } else if (op == 0x603f0000 /* oril r31, r1, 0x0 */ || op == 0x7c3f0b78) { /* mr r31, r1 */