From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 24348 invoked by alias); 29 Apr 2009 10:27:35 -0000 Received: (qmail 24337 invoked by uid 22791); 29 Apr 2009 10:27:34 -0000 X-SWARE-Spam-Status: No, hits=-2.4 required=5.0 tests=AWL,BAYES_00,SPF_HELO_PASS,SPF_PASS X-Spam-Check-By: sourceware.org Received: from mx2.redhat.com (HELO mx2.redhat.com) (66.187.237.31) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Wed, 29 Apr 2009 10:27:25 +0000 Received: from int-mx2.corp.redhat.com (int-mx2.corp.redhat.com [172.16.27.26]) by mx2.redhat.com (8.13.8/8.13.8) with ESMTP id n3TARNtx009303 for ; Wed, 29 Apr 2009 06:27:23 -0400 Received: from ns3.rdu.redhat.com (ns3.rdu.redhat.com [10.11.255.199]) by int-mx2.corp.redhat.com (8.13.1/8.13.1) with ESMTP id n3TARNWh021255 for ; Wed, 29 Apr 2009 06:27:23 -0400 Received: from host0.dyn.jankratochvil.net (sebastian-int.corp.redhat.com [172.16.52.221]) by ns3.rdu.redhat.com (8.13.8/8.13.8) with ESMTP id n3TARLcq029614 for ; Wed, 29 Apr 2009 06:27:22 -0400 Received: from host0.dyn.jankratochvil.net (localhost [127.0.0.1]) by host0.dyn.jankratochvil.net (8.14.3/8.14.3) with ESMTP id n3TARK5G013140 for ; Wed, 29 Apr 2009 12:27:20 +0200 Received: (from jkratoch@localhost) by host0.dyn.jankratochvil.net (8.14.3/8.14.3/Submit) id n3TARJnp013137 for gdb-patches@sourceware.org; Wed, 29 Apr 2009 12:27:19 +0200 Date: Wed, 29 Apr 2009 10:27:00 -0000 From: Jan Kratochvil To: gdb-patches@sourceware.org Subject: [patch] Fix i386 memory-by-register access on amd64 Message-ID: <20090429102719.GA10117@host0.dyn.jankratochvil.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.5.19 (2009-01-05) 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: 2009-04/txt/msg00786.txt.bz2 Hi, original bugreport: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=181390 (gdb) x/x $esp 0xffffce70: 0x00000001 (gdb) x/x $ebx 0xffffce70: Cannot access memory at address 0xffffce70 (gdb) x/x 0xffffce70 0xffffce70: 0x00000001 One point is there should have been printed this error message instead: 0xffffffffffffce70: Cannot access memory at address 0xffffffffffffce70 but this problem is just a consequence of paddress() truncating the printed address width. This printing issue is unrelated to the patch below. The error happens because $ebx is considered signed while $esp unsigned, as initialized by i386_register_type (or also amd64_register_type). Therefore the address width should be cut to the right size at the right point of processing, I hope I caught (one of) such points. Regression-tested on x86_64-unknown-linux-gnu (PASS), i386 build with unix/-m32 (test skipped) and native build with unix/-m64 (new test FAILs as the test's additionla_flags=-m32 gets overriden by target board's -m64). Thanks, Jan 2006-09-28 Jan Kratochvil Fix signed 32bit inferior registers on 64bit GDB. * gdb/value.c (value_as_address): Make it static, rename it to ... (value_as_address1): ... this function. (value_as_address): New function. 2008-03-02 Jan Kratochvil * gdb.arch/amd64-i386-address.exp, gdb.arch/amd64-i386-address.S: New. --- gdb/value.c 21 Mar 2009 03:03:53 -0000 1.79 +++ gdb/value.c 29 Apr 2009 10:07:20 -0000 @@ -1258,11 +1258,11 @@ value_as_double (struct value *val) return foo; } -/* Extract a value as a C pointer. Does not deallocate the value. - Note that val's type may not actually be a pointer; value_as_long - handles all the cases. */ -CORE_ADDR -value_as_address (struct value *val) +/* Extract a value as a C pointer. Helper for value_as_address still does not + truncate the CORE_ADDR width. */ + +static CORE_ADDR +value_as_address1 (struct value *val) { /* Assume a CORE_ADDR can fit in a LONGEST (for now). Not sure whether we want this to be true eventually. */ @@ -1362,6 +1362,27 @@ value_as_address (struct value *val) return unpack_long (value_type (val), value_contents (val)); #endif } + +/* Extract a value as a C pointer. Does not deallocate the value. + Note that val's type may not actually be a pointer; value_as_long + handles all the cases. + + This wrapper truncates the width to match target address width, */ + +CORE_ADDR +value_as_address (struct value *val) +{ + CORE_ADDR addr; + int addr_bit = gdbarch_addr_bit (current_gdbarch); + + addr = value_as_address1 (val); + + /* Compare ADDR_BIT first to avoid a compiler warning on shift overflow. */ + if (addr_bit < (sizeof (CORE_ADDR) * HOST_CHAR_BIT)) + addr &= ((CORE_ADDR) 1 << addr_bit) - 1; + + return addr; +} /* Unpack raw data (copied from debugee, target byte order) at VALADDR as a long, or as a double, assuming the raw data is described --- /dev/null 1 Jan 1970 00:00:00 -0000 +++ gdb/testsuite/gdb.arch/amd64-i386-address.S 29 Apr 2009 10:07:20 -0000 @@ -0,0 +1,25 @@ +/* Copyright 2009 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 3 of the License, or + (at your option) any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program. If not, see . + + This file is part of the gdb testsuite. */ + +_start: .globl _start + nop + int3 + movl %esp, %ebx + /* Examining memory from $ebx fails, from $esp it succeeds. */ + int3 + nop + nop --- /dev/null 1 Jan 1970 00:00:00 -0000 +++ gdb/testsuite/gdb.arch/amd64-i386-address.exp 29 Apr 2009 10:07:20 -0000 @@ -0,0 +1,44 @@ +# Copyright 2009 Free Software Foundation, Inc. + +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 3 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see . + +# This file is part of the gdb testsuite. + +# Test UNsigned extension of the 32-bit inferior address on a 64-bit host. + +if {![istarget "x86_64-*-*"]} then { + verbose "Skipping amd64->i386 adress test." + return +} + +if [prepare_for_testing amd64-i386-address.exp amd64-i386-address amd64-i386-address.S [list debug "additional_flags=-m32 -nostdlib"]] { + return -1 +} + +gdb_run_cmd + +set test "trap stop" +gdb_test_multiple "" $test { + -re "Program received signal SIGTRAP,.*_start .*$gdb_prompt $" { + pass $test + } +} + +gdb_test "stepi" ".*_start .*int3.*" + +gdb_test "x/x \$esp" "0x\[0-9a-f\]*:\t0x0*1" + +# Failure case would be: +# 0xff8d7f00: Cannot access memory at address 0xff8d7f00 +gdb_test "x/x \$ebx" "0x\[0-9a-f\]*:\t0x0*1"