From: Giuseppe MONTALTO <giuseppe.montalto@st.com>
To: Tom Tromey <tromey@redhat.com>
Cc: "Abid, Hafiz" <Hafiz_Abid@mentor.com>,
"gdb-patches@sourceware.org" <gdb-patches@sourceware.org>
Subject: RE: [PATCH] enhancement of mi_cmd_data_write_memory_bytes for filling memory regions (was [PATCH] new MI command for pattern filling of memory regions)
Date: Tue, 18 Sep 2012 15:19:00 -0000 [thread overview]
Message-ID: <76FE3225DF13124EA2D05B290B624C95E66B003EA8@SAFEX1MAIL1.st.com> (raw)
In-Reply-To: <878vcc1s9f.fsf@fleche.redhat.com>
[-- Attachment #1: Type: text/plain, Size: 2296 bytes --]
Hi Tom,
Thanks for your comments: I've reworked the patch taking them into account.
I have also updated it, so that it should now fit the current HEAD
Please, find it attached.
Regards,
Giuseppe.
> -----Original Message-----
> From: Tom Tromey [mailto:tromey@redhat.com]
> Sent: Friday, September 14, 2012 6:31 PM
> To: Giuseppe MONTALTO
> Cc: Abid, Hafiz; gdb-patches@sourceware.org
> Subject: Re: [PATCH] enhancement of mi_cmd_data_write_memory_bytes
> for filling memory regions (was [PATCH] new MI command for pattern filling
> of memory regions)
>
> >>>>> "Giuseppe" == Giuseppe MONTALTO <giuseppe.montalto@st.com>
> writes:
>
> Giuseppe> New version of the patch.
>
> I'm sorry about the delay on this.
>
> I know it feels uncomfortable, but it really is good to ping patches -- say every
> week or so. Otherwise they tend to slide off the radar.
>
> Giuseppe> + int len, r, i, steps, reminder; long int count, j;
>
> I think now size_t is preferred. (This changed since your patch
> submission.)
>
> Typo: should be "remainder", not "reminder".
>
> Giuseppe> + count = strtol (argv[2], NULL, 10);
>
> I think strtoul instead.
>
> Giuseppe> + if (len < count)
> Giuseppe> + {
> Giuseppe> + /* pattern is made of less bytes than count:
> Giuseppe> + repeat pattern to fill memory. */
>
> Comments should start with an uppercase letter.
>
> Giuseppe> + else
> Giuseppe> + {
> Giuseppe> + /* pattern is longer than (or equal to) count:
> Giuseppe> + just copy len bytes. */
>
> Likewise.
>
> Giuseppe> + xfree (databuf);
>
> It seems like you could do less work for the usual case where len==count.
> That is, avoid the extra malloc and memcpy.
>
> Giuseppe> +set testfile "mi-read-memory"
> Giuseppe> +set srcfile ${testfile}.c
> Giuseppe> +set binfile ${objdir}/${subdir}/${testfile}
>
> Use standard_testfile now.
>
> Giuseppe> +if { [gdb_compile "${srcdir}/${subdir}/${srcfile}" "${binfile}"
> executable {debug additional_flags=-DFAKEARGV}] != "" } {
> Giuseppe> + untested mi-read-memory.exp
>
> Use build_executable instead and don't call untested.
>
> Giuseppe> +return 0
>
> No need for a return at the end of a .exp file.
>
> Tom
[-- Attachment #2: mi_cmd_data_write_memory_bytes.patch --]
[-- Type: application/octet-stream, Size: 6806 bytes --]
gdb/ChangeLog | 5 ++
gdb/mi/mi-main.c | 61 ++++++++++++++++++++++-----
gdb/testsuite/ChangeLog | 4 ++
gdb/testsuite/gdb.mi/mi-fill-memory.exp | 68 +++++++++++++++++++++++++++++++
4 files changed, 127 insertions(+), 11 deletions(-)
diff --git a/gdb/ChangeLog b/gdb/ChangeLog
index 4ab15c4..3a0268e 100644
--- a/gdb/ChangeLog
+++ b/gdb/ChangeLog
@@ -1,3 +1,8 @@
+2012-09-18 Giuseppe Montalto <giuseppe.montalto@st.com>
+
+ * mi/mi-main.c (mi_cmd_data_write_memory): additional
+ parameter for pattern filling of memory regions
+
2012-09-17 Mike Wrighton <wrighton@codesourcery.com>
* MAINTAINERS (Write After Approval): Add "Mike Wrighton".
diff --git a/gdb/mi/mi-main.c b/gdb/mi/mi-main.c
index f1d21bc..5f670c2 100644
--- a/gdb/mi/mi-main.c
+++ b/gdb/mi/mi-main.c
@@ -1656,7 +1656,8 @@ mi_cmd_data_write_memory (char *command, char **argv, int argc)
/* Implementation of the -data-write-memory-bytes command.
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). */
void
mi_cmd_data_write_memory_bytes (char *command, char **argv, int argc)
@@ -1664,30 +1665,68 @@ 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;
+ size_t len, r, i, steps, remainder;
+ long int count, j;
struct cleanup *back_to;
- if (argc != 2)
- error (_("Usage: ADDR DATA."));
+ if (argc != 2 && argc != 3)
+ error (_("Usage: ADDR DATA [COUNT]."));
addr = parse_and_eval_address (argv[0]);
cdata = argv[1];
len = strlen (cdata)/2;
+ if (argc == 3)
+ count = strtoul (argv[2], NULL, 10);
+ else
+ count = len;
- data = xmalloc (len);
- back_to = make_cleanup (xfree, data);
+ databuf = xmalloc (len * sizeof (gdb_byte));
for (i = 0; i < len; ++i)
{
int x;
sscanf (cdata + i * 2, "%02x", &x);
- data[i] = (gdb_byte) x;
+ databuf[i] = (gdb_byte) x;
}
- r = target_write_memory (addr, data, len);
+ if (len < count)
+ {
+ /* Pattern is made of less bytes than count:
+ repeat pattern to fill memory. */
+ data = xmalloc (count);
+ back_to = make_cleanup (xfree, data);
+
+ steps = count / len;
+ remainder = count % len; /* there may be some spare bytes. */
+ for (j = 0; j < steps; j++)
+ memcpy (data + j * len, databuf, len);
+
+ if (remainder > 0) /* copy spare bytes too. */
+ memcpy (data + steps * len, databuf, remainder);
+ }
+ else if (len > count)
+ {
+ /* Pattern is longer than count:
+ just copy len bytes. */
+ data = xmalloc (len);
+ back_to = make_cleanup (xfree, data);
+ memcpy (data, databuf, len);
+ }
+ else
+ {
+ /* Pattern is equal to count:
+ we can just use databuf. */
+ back_to = make_cleanup (xfree, data);
+ data = databuf;
+ }
+
+ r = target_write_memory (addr, data, count);
if (r != 0)
error (_("Could not write memory"));
+ xfree (databuf);
+
do_cleanups (back_to);
}
@@ -2097,10 +2136,10 @@ mi_cmd_execute (struct mi_parse *parse)
current_context = parse;
- if (parse->cmd->suppress_notification != NULL)
+ if (strncmp (parse->command, "break-", sizeof ("break-") - 1 ) == 0)
{
- make_cleanup_restore_integer (parse->cmd->suppress_notification);
- *parse->cmd->suppress_notification = 1;
+ make_cleanup_restore_integer (&mi_suppress_breakpoint_notifications);
+ mi_suppress_breakpoint_notifications = 1;
}
if (parse->cmd->argv_func != NULL)
diff --git a/gdb/testsuite/ChangeLog b/gdb/testsuite/ChangeLog
index b458aba..42f66be 100644
--- a/gdb/testsuite/ChangeLog
+++ b/gdb/testsuite/ChangeLog
@@ -1,3 +1,7 @@
+2012-09-18 Giuseppe Montalto <giuseppe.montalto@st.com>
+
+ * gdb.mi/mi-fill-memory.exp: New test.
+
2012-09-17 Yao Qi <yao@codesourcery.com>
* gdb.base/list.exp (set_listsize): Don't set arg to "unlimited"
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..3df6a96
--- /dev/null
+++ b/gdb/testsuite/gdb.mi/mi-fill-memory.exp
@@ -0,0 +1,68 @@
+# 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 <http://www.gnu.org/licenses/>.
+
+#
+# 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.
+#
+# added for testing the -data-write-memory-bytes MI command enhancements
+#
+
+load_lib mi-support.exp
+set MIFLAGS "-i=mi"
+
+gdb_exit
+if [mi_gdb_start] {
+ continue
+}
+
+standard_testfile "mi-read-memory"
+
+if {[build_executable ${testfile}.exp ${binfile} ${srcfile}.c {debug additional_flags=-DFAKEARGV}] == -1} {
+ 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=\"Usage: ADDR DATA \\\[COUNT\\\]\.\""\
+ "no arguments"
+
+mi_gdb_test "2-data-write-memory-bytes 8"\
+ "2\\\^error,msg=\"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
next prev parent reply other threads:[~2012-09-18 15:19 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-05-09 16:18 [PATCH] new MI command for pattern filling of memory regions Giuseppe MONTALTO
2012-05-09 16:30 ` Tom Tromey
2012-05-09 18:11 ` Tom Tromey
2012-05-10 11:41 ` Giuseppe MONTALTO
2012-05-10 13:57 ` Tom Tromey
2012-05-11 8:53 ` Giuseppe MONTALTO
2012-05-11 14:36 ` Tom Tromey
2012-05-11 16:06 ` Giuseppe MONTALTO
2012-05-25 14:35 ` Giuseppe MONTALTO
2012-06-12 9:07 ` Giuseppe MONTALTO
2012-06-12 10:01 ` Abid, Hafiz
2012-06-12 14:13 ` Giuseppe MONTALTO
2012-06-12 14:22 ` [PATCH] enhancement of mi_cmd_data_write_memory_bytes for filling memory regions (was [PATCH] new MI command for pattern filling of memory regions) Giuseppe MONTALTO
2012-09-14 16:30 ` Tom Tromey
2012-09-18 15:19 ` Giuseppe MONTALTO [this message]
2012-09-26 20:44 ` Tom Tromey
2012-09-27 15:27 ` Giuseppe MONTALTO
2012-09-28 20:07 ` Tom Tromey
2012-10-01 9:29 ` Giuseppe MONTALTO
2012-10-17 20:59 ` Tom Tromey
2012-10-18 13:40 ` Giuseppe MONTALTO
2012-10-18 15:41 ` Pedro Alves
2012-10-18 16:16 ` Giuseppe MONTALTO
2012-10-18 16:28 ` [+docs] " Pedro Alves
2012-10-18 17:31 ` Eli Zaretskii
2012-10-18 17:30 ` Eli Zaretskii
2012-10-19 14:12 ` Giuseppe MONTALTO
2012-10-19 16:34 ` Eli Zaretskii
2012-10-19 16:56 ` Pedro Alves
2012-10-22 7:02 ` Giuseppe MONTALTO
2012-11-08 20:37 ` Tom Tromey
2012-11-09 8:59 ` Giuseppe MONTALTO
2012-11-12 16:52 ` Tom Tromey
2012-11-13 10:50 ` Giuseppe MONTALTO
2012-11-13 21:19 ` Tom Tromey
2012-10-18 16:38 ` Giuseppe MONTALTO
2012-10-18 19:45 ` Tom Tromey
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=76FE3225DF13124EA2D05B290B624C95E66B003EA8@SAFEX1MAIL1.st.com \
--to=giuseppe.montalto@st.com \
--cc=Hafiz_Abid@mentor.com \
--cc=gdb-patches@sourceware.org \
--cc=tromey@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox