From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 32151 invoked by alias); 25 May 2012 14:35:56 -0000 Received: (qmail 31828 invoked by uid 22791); 25 May 2012 14:35:54 -0000 X-SWARE-Spam-Status: No, hits=-2.7 required=5.0 tests=AWL,BAYES_00,KHOP_RCVD_UNTRUST,KHOP_THREADED,RCVD_IN_HOSTKARMA_W,RCVD_IN_HOSTKARMA_WL,TW_CP X-Spam-Check-By: sourceware.org Received: from eu1sys200aog113.obsmtp.com (HELO eu1sys200aog113.obsmtp.com) (207.126.144.135) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Fri, 25 May 2012 14:35:39 +0000 Received: from beta.dmz-eu.st.com ([164.129.1.35]) (using TLSv1) by eu1sys200aob113.postini.com ([207.126.147.11]) with SMTP ID DSNKT7+Yt+zMipgxHpu5HNrf9nbgAijbljr6@postini.com; Fri, 25 May 2012 14:35:39 UTC Received: from zeta.dmz-eu.st.com (zeta.dmz-eu.st.com [164.129.230.9]) by beta.dmz-eu.st.com (STMicroelectronics) with ESMTP id 0327913C; Fri, 25 May 2012 14:35:30 +0000 (GMT) Received: from Webmail-eu.st.com (safex1hubcas6.st.com [10.75.90.73]) by zeta.dmz-eu.st.com (STMicroelectronics) with ESMTP id 32DDA4722; Fri, 25 May 2012 14:35:30 +0000 (GMT) Received: from SAFEX1MAIL1.st.com ([10.75.90.1]) by Safex1hubcas6.st.com ([10.75.90.73]) with mapi; Fri, 25 May 2012 16:35:30 +0200 From: Giuseppe MONTALTO To: Tom Tromey Cc: "gdb-patches@sourceware.org" Date: Fri, 25 May 2012 14:35:00 -0000 Subject: RE: [PATCH] new MI command for pattern filling of memory regions Message-ID: <76FE3225DF13124EA2D05B290B624C95E6621EFB26@SAFEX1MAIL1.st.com> References: <76FE3225DF13124EA2D05B290B624C95E65F2D466E@SAFEX1MAIL1.st.com> <87sjf9rzr0.fsf@fleche.redhat.com> <76FE3225DF13124EA2D05B290B624C95E65FECE356@SAFEX1MAIL1.st.com> <871umsqgu8.fsf@fleche.redhat.com> <76FE3225DF13124EA2D05B290B624C95E65FECE48F@SAFEX1MAIL1.st.com> <87397624b9.fsf@fleche.redhat.com> Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable 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: 2012-05/txt/msg00969.txt.bz2 I reworked the patch, including the test script; Accordingly with your feedback, the sscanf cycle is now performed only once. Here's the new patch, please let me know if there is anything more to chang= e. Regards, Giuseppe gdb/ChangeLog | 5 ++ gdb/mi/mi-main.c | 50 +++++++++++++++++---- gdb/testsuite/ChangeLog | 4 ++ gdb/testsuite/gdb.mi/mi-fill-memory.exp | 72 +++++++++++++++++++++++++++= ++++ 4 files changed, 122 insertions(+), 9 deletions(-) diff --git a/gdb/ChangeLog b/gdb/ChangeLog index 309473e..9ec8882 100644 --- a/gdb/ChangeLog +++ b/gdb/ChangeLog @@ -1,3 +1,8 @@ +2012-05-09 Giuseppe Montalto + + * mi/mi-main.c (mi_cmd_data_write_memory): additional + parameter for pattern filling of memory regions + 2012-05-09 Pedro Alves =20 * target.c (set_maintenance_target_async_permitted): Rename to ... diff --git a/gdb/mi/mi-main.c b/gdb/mi/mi-main.c index 90fb624..7315848 100644 --- a/gdb/mi/mi-main.c +++ b/gdb/mi/mi-main.c @@ -1658,7 +1658,8 @@ mi_cmd_data_write_memory (char *command, char **argv,= int argc) /* Implementation of the -data-write-memory-bytes command. =20 ADDR: start address - DATA: string of bytes to write at that address. */ + DATA: string of bytes to write at that address + COUNT: number of bytes to be filled (decimal integer). */ =20 void mi_cmd_data_write_memory_bytes (char *command, char **argv, int argc) @@ -1666,27 +1667,58 @@ mi_cmd_data_write_memory_bytes (char *command, char= **argv, int argc) CORE_ADDR addr; char *cdata; gdb_byte *data; - int len, r, i; + gdb_byte *databuf; + int len, r, i, steps, reminder; + long int count, j; struct cleanup *back_to; =20 - if (argc !=3D 2) - error (_("Usage: ADDR DATA.")); + if (argc !=3D 2 && argc !=3D 3) + error (_("Usage: ADDR DATA [COUNT].")); =20 addr =3D parse_and_eval_address (argv[0]); cdata =3D argv[1]; - len =3D strlen (cdata)/2; + len =3D strlen (cdata) / 2; + if (argc =3D=3D 3) + count =3D strtol(argv[2], NULL, 10); + else + count =3D len; =20 - data =3D xmalloc (len); - back_to =3D make_cleanup (xfree, data); + databuf =3D xmalloc (len * sizeof(gdb_byte)); =20 for (i =3D 0; i < len; ++i) { int x; sscanf (cdata + i * 2, "%02x", &x); - data[i] =3D (gdb_byte) x; + databuf[i] =3D (gdb_byte)x; } =20 - r =3D target_write_memory (addr, data, len); + if (len < count) + { + /* pattern is made of less bytes than count:=20 + repeat pattern to fill memory. */ + data =3D xmalloc (count); + back_to =3D make_cleanup (xfree, data); +=20=20=20=20 + steps =3D count / len; + reminder =3D count % len; /* there may be some spare bytes. */ + for (j =3D 0; j < steps; j++) + memcpy (data + j * len, databuf, (size_t) len); + + if (reminder > 0) /* copy spare bytes too. */ + memcpy (data + steps * len, databuf, (size_t) reminder); + } + else + { + /* pattern is longer than (or equal to) count:=20 + just copy len bytes. */ + data =3D xmalloc (len); + back_to =3D make_cleanup (xfree, data); + memcpy (data, databuf, (size_t) len); + } + + xfree(databuf); + + r =3D target_write_memory (addr, data, count); if (r !=3D 0) error (_("Could not write memory")); =20 diff --git a/gdb/testsuite/ChangeLog b/gdb/testsuite/ChangeLog index 43549bf..a49274f 100644 --- a/gdb/testsuite/ChangeLog +++ b/gdb/testsuite/ChangeLog @@ -1,3 +1,7 @@ +2012-05-09 Giuseppe Montalto + + * gdb.mi/mi-fill-memory.exp: New test. + 2012-05-08 Maciej W. Rozycki =20 * gdb.mi/mi-var-display.exp: Check for the existence of $fp diff --git a/gdb/testsuite/gdb.mi/mi-fill-memory.exp b/gdb/testsuite/gdb.mi= /mi-fill-memory.exp new file mode 100644 index 0000000..b6961b1 --- /dev/null +++ b/gdb/testsuite/gdb.mi/mi-fill-memory.exp @@ -0,0 +1,72 @@ +# Copyright (C) 2012 Free Software Foundation, Inc. +# Copyright (C) 2012 STMicroelectronics + +# 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 . + +# +# test basic Machine interface (MI) operations +# +# Verify that, using the MI, we can load a program and do +# other basic things that are used by all test files through mi_gdb_exit, +# mi_gdb_start, mi_delete_breakpoints, mi_gdb_reinitialize_dir and +# mi_gdb_load, so we can safely use those. +# +# The goal is not to test gdb functionality, which is done by other tests, +# but the command syntax and correct output response to MI operations. +#=20 +# added for testing the -data-write-memory-bytes MI command enhancements +# + +load_lib mi-support.exp +set MIFLAGS "-i=3Dmi" + +gdb_exit +if [mi_gdb_start] { + continue +} + +set testfile "mi-read-memory" +set srcfile ${testfile}.c +set binfile ${objdir}/${subdir}/${testfile} +if { [gdb_compile "${srcdir}/${subdir}/${srcfile}" "${binfile}" executabl= e {debug additional_flags=3D-DFAKEARGV}] !=3D "" } { + untested mi-read-memory.exp + return -1 +} + + +mi_run_to_main +mi_next_to "main" "" "mi-read-memory.c" "20" "next at main" + +mi_gdb_test "1-data-write-memory-bytes"\ + "1\\\^error,msg=3D\"Usage: ADDR DATA \\\[COUNT\\\]\.\""\ + "no arguments" + +mi_gdb_test "2-data-write-memory-bytes 8"\ + "2\\\^error,msg=3D\"Usage: ADDR DATA \\\[COUNT\\\]\.\""\ + "one argument missing" + +mi_gdb_test "3-data-write-memory-bytes \$pc ab"\ + "3\\\^done" \ + "memory successfully written" + +mi_gdb_test "4-data-write-memory-bytes \$pc ab 8"\ + "4\\\^done" \ + "memory successfully filled (8 bytes)" + +mi_gdb_test "5-interpreter-exec console \"x \$pc\"" \ + ".*0xabababab.*" \ + "pattern correctly read from memory" + +mi_gdb_exit +return 0 > -----Original Message----- > From: Giuseppe MONTALTO > Sent: Friday, May 11, 2012 6:06 PM > To: 'Tom Tromey' > Cc: gdb-patches@sourceware.org > Subject: RE: [PATCH] new MI command for pattern filling of memory > regions >=20 > Thanks, >=20 > my answers are interleaved below. >=20 > > -----Original Message----- > > From: Tom Tromey [mailto:tromey@redhat.com] > > Sent: Friday, May 11, 2012 4:14 PM > > To: Giuseppe MONTALTO > > Cc: gdb-patches@sourceware.org > > Subject: Re: [PATCH] new MI command for pattern filling of memory > > regions > > > > The indentation is wrong here and elsewhere in the patch. > > There are missing spaces as well. > > See the GNU Coding Standards for details. >=20 > I'll try to better format the code. >=20 > > > > I don't think the cast here, or other casts in the patch, are > > necessary. >=20 > Consider them gone. >=20 > > > > Giuseppe> + /* pattern is made of less bytes than count: > > Giuseppe> + repeat pattern to fill memory. */ > > Giuseppe> + data =3D xmalloc (count/len); > > > > This allocates count/len bytes... > > > > Giuseppe> + steps =3D count/len; > > Giuseppe> + for (j =3D 0; j < steps; j++) > > Giuseppe> + { > > Giuseppe> + for (i =3D 0; i < len; ++i) > > Giuseppe> + { > > Giuseppe> + int x; > > Giuseppe> + sscanf (cdata + i * 2, "%02x", &x); > > Giuseppe> + data[i + j * len] =3D (gdb_byte)x; > > > > ... but writes 'count' bytes. >=20 > you're right; I was supposed to allocate 'count' bytes. >=20 > Incidentally, there's a worse error there: > is that I'm counting the cycles basing on the result of an integer > division, hence not taking into account the spare bytes! >=20 > I'll fix that too... >=20 > Giuseppe