From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 18789 invoked by alias); 13 Nov 2007 21:47:04 -0000 Received: (qmail 18781 invoked by uid 22791); 13 Nov 2007 21:47:04 -0000 X-Spam-Check-By: sourceware.org Received: from hq.tensilica.com (HELO mailapp.tensilica.com) (65.205.227.29) by sourceware.org (qpsmtpd/0.31) with ESMTP; Tue, 13 Nov 2007 21:47:02 +0000 Received: from localhost ([127.0.0.1]) by mailapp.tensilica.com with esmtp (Exim 4.34) id 1Is3bI-0007kW-3C for binutils@sources.redhat.com; Tue, 13 Nov 2007 13:47:16 -0800 Received: from mailapp.tensilica.com ([127.0.0.1]) by localhost (mailapp [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 21005-05 for ; Tue, 13 Nov 2007 13:47:15 -0800 (PST) Received: from heron.hq.tensilica.com ([192.168.11.123]) by mailapp.tensilica.com with esmtp (Exim 4.34) id 1Is3bF-0007WH-As for binutils@sources.redhat.com; Tue, 13 Nov 2007 13:47:13 -0800 Received: from [192.168.11.123] (heron.hq.tensilica.com [192.168.11.123]) by heron.hq.tensilica.com (Postfix) with ESMTP id 103705EC20 for ; Tue, 13 Nov 2007 13:46:55 -0800 (PST) Message-ID: <473A1B4E.8030504@tensilica.com> Date: Tue, 13 Nov 2007 21:47:00 -0000 From: Bob Wilson User-Agent: Thunderbird 1.5.0.14pre (X11/20071023) MIME-Version: 1.0 To: binutils@sources.redhat.com Subject: Re: [PATCH] fix DWARF for ia64 References: <473A01B5.30907@tensilica.com> In-Reply-To: <473A01B5.30907@tensilica.com> Content-Type: multipart/mixed; boundary="------------050508010201020607060702" Mailing-List: contact binutils-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: binutils-owner@sourceware.org X-SW-Source: 2007-11/txt/msg00141.txt.bz2 This is a multi-part message in MIME format. --------------050508010201020607060702 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-length: 700 Here is a follow-on patch. I noticed that dwarf2_emit_label() can also use the new dwarf2_consume_line_info function. Is this OK? I forgot to say how I tested this. I built binutils for xtensa-elf, ia64-unknown-linux-gnu and i686-pc-linux-gnu targets, all hosted on i686-pc-linux-gnu, and ran the testsuite. I've rerun those tests now with this follow-on patch. Everything passed as expected with one exception: the ld-ia64/line.exp test fails but I see the same failure without any of my changes. ld fails to show the source line number in an error message. That does seem suspicious. Is this a known problem? gas/ * dwarf2dbg.c (dwarf2_emit_label): Use dwarf2_consume_line_info. --------------050508010201020607060702 Content-Type: text/x-diff; name="gas-debug-ia64-2.patch" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="gas-debug-ia64-2.patch" Content-length: 540 --- dwarf2dbg.c.2 2007-11-13 12:41:33.000000000 -0800 +++ dwarf2dbg.c 2007-11-13 12:42:42.000000000 -0800 @@ -419,17 +419,11 @@ if (debug_type == DEBUG_DWARF2) dwarf2_where (&loc); else - { - loc = current; - dwarf2_loc_directive_seen = FALSE; - } + loc = current; loc.flags |= DWARF2_FLAG_BASIC_BLOCK; - current.flags &= ~(DWARF2_FLAG_BASIC_BLOCK - | DWARF2_FLAG_PROLOGUE_END - | DWARF2_FLAG_EPILOGUE_BEGIN); - + dwarf2_consume_line_info (); dwarf2_gen_line_info_1 (label, &loc); } --------------050508010201020607060702--