Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Jan Kratochvil <jan.kratochvil@redhat.com>
To: Pedro Alves <pedro@codesourcery.com>
Cc: gdb-patches@sourceware.org
Subject: Re: [patch] Fix internal-error on dead LWPs with no associated 	thread
Date: Mon, 29 Jun 2009 17:51:00 -0000	[thread overview]
Message-ID: <20090629174929.GA9652@host0.dyn.jankratochvil.net> (raw)
In-Reply-To: <200906291136.28812.pedro@codesourcery.com>

On Mon, 29 Jun 2009 12:36:28 +0200, Pedro Alves wrote:
> On Monday 29 June 2009 11:09:22, Jan Kratochvil wrote:
> 
> > original bugreport:
> > https://bugzilla.redhat.com/show_bug.cgi?id=471819
> > 
> > linux-nat.c:1662: internal-error: linux_nat_resume: Assertion `lp != NULL' failed.
[...]
> Please try the below patchlet instead.  Do you have any other case where this
> would be needed?
[...]
> +      if (!ptid_equal (ecs->ptid, inferior_ptid))
> +	context_switch (ecs->ptid);
>        target_resume (RESUME_ALL, 0, TARGET_SIGNAL_0);
>        prepare_to_wait (ecs);
>        return;

Your patch works for my testcase and I failed to find any countercase for it.
It has no regressions on {x86_64,i686}-fedora-linux-gnu.

Could you please check it in?  Also the attached testcase needs an approval.


> Does anyone know why does the new_thread_event bit need to resume the target
> at all?  Removing that resume should fix the issue too.

I find it more as a cleanup and going to send a followup with a patch made now
for it.


Thanks,
Jan


gdb/testsuite/
2009-06-29  Jan Kratochvil  <jan.kratochvil@redhat.com>

	* gdb.threads/current-lwp-dead.exp, gdb.threads/current-lwp-dead.c: New.

--- /dev/null
+++ b/gdb/testsuite/gdb.threads/current-lwp-dead.c
@@ -0,0 +1,75 @@
+/* This testcase is part of GDB, the GNU debugger.
+
+   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 <http://www.gnu.org/licenses/>.
+
+   Do not use threads as we need to exploit a bug in LWP code masked by the
+   threads code otherwise.
+
+   INFERIOR_PTID must point to exited LWP.  Here we use the initial LWP as it
+   is automatically INFERIOR_PTID for GDB.
+
+   Finally we need to call target_resume (RESUME_ALL, ...) which we invoke by
+   NEW_THREAD_EVENT (called from the new LWP as initial LWP is exited now).  */
+
+#define _GNU_SOURCE
+#include <sched.h>
+#include <assert.h>
+#include <unistd.h>
+#include <stdlib.h>
+
+#define STACK_SIZE 0x1000
+
+static int
+fn_return (void *unused)
+{
+  return 0;	/* at-fn_return */
+}
+
+static int
+fn (void *unused)
+{
+  int i;
+  unsigned char *stack;
+  int new_pid;
+
+  i = sleep (1);
+  assert (i == 0);
+
+  stack = malloc (STACK_SIZE);
+  assert (stack != NULL);
+
+  new_pid = clone (fn_return, stack + STACK_SIZE, CLONE_FILES | CLONE_VM, NULL,
+		   NULL, NULL, NULL);
+  assert (new_pid > 0);
+
+  return 0;
+}
+
+int
+main (int argc, char **argv)
+{
+  unsigned char *stack;
+  int new_pid;
+
+  stack = malloc (STACK_SIZE);
+  assert (stack != NULL);
+
+  new_pid = clone (fn, stack + STACK_SIZE, CLONE_FILES | CLONE_VM, NULL, NULL,
+		   NULL, NULL);
+  assert (new_pid > 0);
+
+  return 0;
+}
--- /dev/null
+++ b/gdb/testsuite/gdb.threads/current-lwp-dead.exp
@@ -0,0 +1,31 @@
+# This testcase is part of GDB, the GNU debugger.
+
+# 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 <http://www.gnu.org/licenses/>.
+
+# Please email any bugs, comments, and/or additions to this file to:
+# bug-gdb@gnu.org
+
+if { [prepare_for_testing current-lwp-dead.exp current-lwp-dead] } {
+    return -1
+}
+
+if {[runto_main] <= 0} {
+    untested current-lwp-dead.exp
+    return -1
+}
+
+gdb_breakpoint "fn_return"
+gdb_continue_to_breakpoint "fn_return" ".*at-fn_return.*"


  reply	other threads:[~2009-06-29 17:51 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-06-29 10:09 Jan Kratochvil
2009-06-29 10:35 ` Pedro Alves
2009-06-29 17:51   ` Jan Kratochvil [this message]
2009-06-29 18:30     ` Pedro Alves
2009-06-29 18:59       ` Jan Kratochvil
2009-06-29 18:31     ` Jan Kratochvil
2009-06-29 18:41       ` Pedro Alves
2009-06-29 21:35         ` Jan Kratochvil
2009-06-29 21:42           ` Pedro Alves
2009-06-29 21:48             ` Jan Kratochvil

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=20090629174929.GA9652@host0.dyn.jankratochvil.net \
    --to=jan.kratochvil@redhat.com \
    --cc=gdb-patches@sourceware.org \
    --cc=pedro@codesourcery.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