From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 10202 invoked by alias); 8 Mar 2012 19:40:36 -0000 Received: (qmail 10187 invoked by uid 22791); 8 Mar 2012 19:40:33 -0000 X-SWARE-Spam-Status: No, hits=-6.7 required=5.0 tests=AWL,BAYES_00,RCVD_IN_DNSWL_HI,SPF_HELO_PASS,T_RP_MATCHES_RCVD X-Spam-Check-By: sourceware.org Received: from mx1.redhat.com (HELO mx1.redhat.com) (209.132.183.28) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Thu, 08 Mar 2012 19:40:14 +0000 Received: from int-mx12.intmail.prod.int.phx2.redhat.com (int-mx12.intmail.prod.int.phx2.redhat.com [10.5.11.25]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id q28JdxqM012336 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Thu, 8 Mar 2012 14:39:59 -0500 Received: from host2.jankratochvil.net (ovpn-116-19.ams2.redhat.com [10.36.116.19]) by int-mx12.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id q28Jdr5C012874 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NO); Thu, 8 Mar 2012 14:39:55 -0500 Date: Thu, 08 Mar 2012 19:40:00 -0000 From: Jan Kratochvil To: Joel Brobecker Cc: gdb-patches@sourceware.org Subject: [commit] [patch] Fix CU relative vs. absolute offsets [Re: RFC: problem with DW_OP_GNU_deref_type and dwarf's get_base_type callback] Message-ID: <20120308193952.GA31141@host2.jankratochvil.net> References: <20120305223429.GM2867@adacore.com> <20120307170940.GA22619@host2.jankratochvil.net> <20120307185730.GQ2853@adacore.com> <20120307194655.GS2853@adacore.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20120307194655.GS2853@adacore.com> User-Agent: Mutt/1.5.21 (2010-09-15) 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/msg00284.txt.bz2 On Wed, 07 Mar 2012 20:46:55 +0100, Joel Brobecker wrote: > FWIW: I didn't see any regression with our testsuite and GCC 4.7. Checked it in, with a new testcase for one part of it. The part - if (type_sig->type_offset != die->offset) + if (type_sig->per_cu.offset + type_sig->type_offset + != die->offset) comes from: Re: [1/4] RFC: skip DIEs which only declare an enum http://sourceware.org/ml/gdb-patches/2011-07/msg00428.html and while it has no new testcase: (1) after changing it the debug output clearly shows it is right: type_sig->type_offset=0x25 != die->offset=0x82e type_sig->type_offset=0x3c != die->offset=0x7ce -> type_sig->type_offset=0x821 != die->offset=0x82e type_sig->type_offset=0x7d4 != die->offset=0x7ce (2) it is also shown in the [patch 2/2] enforcing proper .debug_*-section vs. CU relative offset types. For the dwarf2_fetch_die_location_block part it was even easy to write a testcase showing now FAIL->PASS. Thanks, Jan http://sourceware.org/ml/gdb-cvs/2012-03/msg00134.html --- src/gdb/ChangeLog 2012/03/08 19:08:09 1.13979 +++ src/gdb/ChangeLog 2012/03/08 19:37:04 1.13980 @@ -1,5 +1,15 @@ 2012-03-08 Jan Kratochvil + Fix CU relative vs. absolute DIE offsets. + * dwarf2loc.h (dwarf2_fetch_die_location_block): Rename parameter + offset to offset_in_cu. + * dwarf2read.c (process_enumeration_scope): Add CU offset to + TYPE_OFFSET. + (dwarf2_fetch_die_location_block): Rename parameter offset to + offset_in_cu. New variable offset, add CU offset to OFFSET_IN_CU. + +2012-03-08 Jan Kratochvil + * libunwind-frame.c: Rename to ... * ia64-libunwind-tdep.c: ... here. * libunwind-frame.h: Rename to ... --- src/gdb/testsuite/ChangeLog 2012/03/08 07:42:50 1.3127 +++ src/gdb/testsuite/ChangeLog 2012/03/08 19:37:07 1.3128 @@ -1,5 +1,11 @@ 2012-03-08 Jan Kratochvil + Fix CU relative vs. absolute DIE offsets. + * gdb.dwarf2/dw2-op-call.S: New compilation unit preceding the existing + one. + +2012-03-08 Jan Kratochvil + Fix false FAIL on distros with relro linkage as default. * gdb.reverse/solib-precsave.exp: Try to compile the test using -Wl,-z,norelro first. --- src/gdb/dwarf2loc.h 2012/01/05 21:53:14 1.29 +++ src/gdb/dwarf2loc.h 2012/03/08 19:37:07 1.30 @@ -61,7 +61,7 @@ CORE_ADDR pc); struct dwarf2_locexpr_baton dwarf2_fetch_die_location_block - (unsigned int offset, struct dwarf2_per_cu_data *per_cu, + (unsigned int offset_in_cu, struct dwarf2_per_cu_data *per_cu, CORE_ADDR (*get_frame_pc) (void *baton), void *baton); --- src/gdb/dwarf2read.c 2012/03/06 23:41:50 1.619 +++ src/gdb/dwarf2read.c 2012/03/08 19:37:07 1.620 @@ -8031,7 +8031,8 @@ = lookup_signatured_type_at_offset (dwarf2_per_objfile->objfile, cu->per_cu->debug_types_section, cu->per_cu->offset); - if (type_sig->type_offset != die->offset) + if (type_sig->per_cu.offset + type_sig->type_offset + != die->offset) return; } @@ -14202,11 +14203,12 @@ dwarf2_locexpr_baton->data has lifetime of PER_CU->OBJFILE. */ struct dwarf2_locexpr_baton -dwarf2_fetch_die_location_block (unsigned int offset, +dwarf2_fetch_die_location_block (unsigned int offset_in_cu, struct dwarf2_per_cu_data *per_cu, CORE_ADDR (*get_frame_pc) (void *baton), void *baton) { + unsigned int offset = per_cu->offset + offset_in_cu; struct dwarf2_cu *cu; struct die_info *die; struct attribute *attr; --- src/gdb/testsuite/gdb.dwarf2/dw2-op-call.S 2012/01/04 08:17:51 1.5 +++ src/gdb/testsuite/gdb.dwarf2/dw2-op-call.S 2012/03/08 19:37:08 1.6 @@ -23,6 +23,23 @@ array3: .2byte 3 .section .debug_info +.Lcu0_begin: + /* CU header */ + .4byte .Lcu0_end - .Lcu0_start /* Length of Compilation Unit */ +.Lcu0_start: + .2byte 2 /* DWARF Version */ + .4byte .Labbrev1_begin /* Offset into abbrev section */ + .byte 4 /* Pointer size */ + + /* CU die */ + .uleb128 1 /* Abbrev: DW_TAG_compile_unit */ + .ascii "file0.txt\0" /* DW_AT_name */ + .ascii "GNU C 3.3.3\0" /* DW_AT_producer */ + .byte 2 /* DW_LANG_C (C) */ + + .byte 0 /* End of children of CU */ +.Lcu0_end: + .Lcu1_begin: /* CU header */ .4byte .Lcu1_end - .Lcu1_start /* Length of Compilation Unit */