From: Paul Hilfinger <hilfingr@gnat.com>
To: cagney@gnu.org
Cc: gdb-patches@sources.redhat.com
Subject: Re: [RFA] breakpoint.c: Avoid double freeing in breakpoint_re_set_one
Date: Wed, 28 Jan 2004 12:30:00 -0000 [thread overview]
Message-ID: <20040128123010.A0ED8F28D8@nile.gnat.com> (raw)
In-Reply-To: <40171322.7080704@gnu.org> (message from Andrew Cagney on Tue, 27 Jan 2004 20:40:50 -0500)
> > Anyway, I've committed the attached (slightly tweaked).
>
> Doh! Try ...
I was going to say that was quite an editing job!
> You wouldn't have a test case - at least something that causes this code
> to go through the motions? That way on systems with pedantic memory
> managers there'd bee some sort of error.
As you suggest, this is a little hard to test. I am not too familiar with
the testsuite innards yet. Here is a draft of a possible test. Comments?
Paul
---------------------------- gdb.base/badfree.exp --------------------
# Copyright 2004 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 2 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, write to the Free Software
# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
# Please email any bugs, comments, and/or additions to this file to:
# bug-gdb@prep.ai.mit.edu
# Author: Paul N. Hilfinger (Hilfinger@gnat.com)
# Test that GDB cleans up properly after errors that result when a
# breakpoint is reset.
if $tracelevel then {
strace $tracelevel
}
# IDT/SIM apparently doesn't have enough file descriptors to allow the
# problem checked by this test to occur.
if [istarget "mips-idt-*"] {
return 0;
}
set testfile "badfree"
set srcfile ${testfile}.c
set binfile ${objdir}/${subdir}/${testfile}
if { [gdb_compile "${srcdir}/${subdir}/${srcfile}" "${binfile}" executable {debug additional_flags=-DVARIABLE=var1}] != "" } {
gdb_suppress_entire_file "Testcase compile failed, so all tests in this file will automatically fail."
}
set oldtimeout $timeout
set timeout 10
verbose "Timeout is now 10 seconds" 2
proc expect_to_stop_here { ident } {
global gdb_prompt
global decimal
# the "at foo.c:36" output we get with -g.
# the "in func" output we get without -g.
gdb_expect 30 {
-re "Breakpoint \[0-9\]*, stop_here .*$gdb_prompt $" {
return 1
}
-re "$gdb_prompt $" {
fail "running to stop_here $ident"
return 0
}
timeout {
fail "running to stop_here $ident (timeout)"
return 0
}
}
return 1
}
gdb_exit
gdb_start
gdb_reinitialize_dir $srcdir/$subdir
gdb_load ${binfile}
gdb_test "break stop_here if (var1 == 42)\n" \
"Breakpoint.*at.* file .*$srcfile, line.*" \
"setting conditional breakpoint on function"
gdb_run_cmd
expect_to_stop_here "first time"
gdb_continue_to_end "breakpoint first time through"
# Now we recompile the executable, but without a variable named "var1", first
# waiting to insure that even on fast machines, the file modification times
# are distinct. This will force GDB to reload the file on the
# next "run" command causing an error when GDB tries to tries to reset
# the breakpoint.
sleep 2
if { [gdb_compile "${srcdir}/${subdir}/${srcfile}" "${binfile}" executable {debug additional_flags=-DVARIABLE=var2}] != "" } {
# Complication: Since GDB generally holds an open file descriptor on the
# executable at this point, there are some systems in which the
# re-compilation will fail. In such cases, we'll consider the test
# (vacuously) passed providing that re-running it succeeds as before.
gdb_run_cmd
expect_to_stop_here "after re-compile fails"
gdb_continue_to_end "after re-compile fails"
} else {
send_gdb "run\n"
gdb_expect {
-re ".* has changed; .*Error in re-setting .*No symbol .var1..*Program exited normally.*"\
{ pass "running with invalidated bpt condition after executable changes" }
timeout {
fail "(timeout) running with invalidated bpt condition after executable changes" }
}
}
---------------------------- gdb.base/badfree.c --------------------
/*
* Test that GDB cleans up properly after errors that result when a
* breakpoint is reset.
*/
/* VARIABLE is a macro defined on the compiler command line. */
#include <stdlib.h>
int VARIABLE = 42;
void stop_here ()
{
VARIABLE *= 2;
}
main ()
{
stop_here ();
exit (0);
}
next prev parent reply other threads:[~2004-01-28 12:30 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-01-13 10:06 Paul Hilfinger
2004-01-28 1:36 ` Andrew Cagney
2004-01-28 1:44 ` Andrew Cagney
2004-01-28 12:30 ` Paul Hilfinger [this message]
2004-01-28 20:33 ` Andrew Cagney
2004-01-28 21:42 ` Paul Hilfinger
2004-01-29 11:16 ` [PATCH] Test for " Paul Hilfinger
2004-01-29 14:06 ` Andrew Cagney
-- strict thread matches above, loose matches on Subject: below --
2004-01-09 8:22 [RFA] breakpoint.c: Avoid " Paul Hilfinger
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=20040128123010.A0ED8F28D8@nile.gnat.com \
--to=hilfingr@gnat.com \
--cc=cagney@gnu.org \
--cc=gdb-patches@sources.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