Multi-threaded program debugging failed with gdb on Sparc

Bug #7028 reported by Debian Bug Importer
4
Affects Status Importance Assigned to Milestone
gdb (Debian)
Fix Released
Unknown
gdb (Ubuntu)
Invalid
Medium
Unassigned

Bug Description

Automatically imported from Debian bug report #259276 http://bugs.debian.org/259276

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Automatically imported from Debian bug report #259276 http://bugs.debian.org/259276

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Message-ID: <email address hidden>
Date: Tue, 13 Jul 2004 15:13:08 -0400
From: "Nizar Ahmouda" <email address hidden>
To: <email address hidden>
Subject: Multi-threaded program debugging failed with gdb on Sparc

Package: gdb
Version: 6.1
Severity: serious

Hi,

I am trying to debug a multi-threaded program with gdb on Sparc. After
the creation of a second thread, while the program is being traced, I
either get an assertion failure in gdb or the gdb hangs (depending on
the gdb version - I tried a few of them). This seems to occur after the
creation of a second thread.

Version 5.2 of gdb hangs at the creation of the second thread while
versions 6.x prints the following error:

/build/buildd/gdb-6.1/gdb/lin-lwp.c:1534: internal-error: lin_lwp_wait:
Assertion `iterate_over_lwps (running_callback, NULL)' failed.

To reproduce the bug you simply need to for example "gdb
/usr/lib/galeon-bin" (which is the first multithreaded app that came to
my mind). That should trigger the same behaviour as when I am running
my app.

I set the severity to serious, as this seems to heavily affect sparc
while debugging multithreaded apps, but I believe on other platforms,
like i386 it might not show up.

Revision history for this message
Matt Zimmerman (mdz) wrote :

sparc-specific

Revision history for this message
In , Andreas Metzler (ametzler-logic) wrote :

severity 259276 important
# not rc
thanks
On Tue, Jul 13, 2004 at 03:13:08PM -0400, Nizar Ahmouda wrote:
> Package: gdb
[...]
> I am trying to debug a multi-threaded program with gdb on Sparc. After
> the creation of a second thread, while the program is being traced, I
> either get an assertion failure in gdb or the gdb hangs (depending on
> the gdb version - I tried a few of them). This seems to occur after the
> creation of a second thread.
[...]
> I set the severity to serious, as this seems to heavily affect sparc
> while debugging multithreaded apps, but I believe on other platforms,
> like i386 it might not show up.
[...]

Hello,
I've asked gdb's maintainer and he agrees that this bug is not
release-critical. It does not match the criteria for serious, there is
neither a policy violation nor "in the package maintainer's opinion,
makes the package unsuitable for release."

I am therefore downgrading the report from serious to important.
              cu andreas

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Message-ID: <email address hidden>
Date: Wed, 1 Sep 2004 16:44:18 +0200
From: Andreas Metzler <email address hidden>
To: <email address hidden>
Subject: Re: Multi-threaded program debugging failed with gdb on Sparc

severity 259276 important
# not rc
thanks
On Tue, Jul 13, 2004 at 03:13:08PM -0400, Nizar Ahmouda wrote:
> Package: gdb
[...]
> I am trying to debug a multi-threaded program with gdb on Sparc. After
> the creation of a second thread, while the program is being traced, I
> either get an assertion failure in gdb or the gdb hangs (depending on
> the gdb version - I tried a few of them). This seems to occur after the
> creation of a second thread.
[...]
> I set the severity to serious, as this seems to heavily affect sparc
> while debugging multithreaded apps, but I believe on other platforms,
> like i386 it might not show up.
[...]

Hello,
I've asked gdb's maintainer and he agrees that this bug is not
release-critical. It does not match the criteria for serious, there is
neither a policy violation nor "in the package maintainer's opinion,
makes the package unsuitable for release."

I am therefore downgrading the report from serious to important.
              cu andreas

Revision history for this message
In , Simon Josefsson (simon-josefsson) wrote : Gdb crash on x86 too (was: Re: Reproducible gtk crash)

I get the same gdb crash as reported in 259276, but on x86, see below.

I was running fully upgraded Debian testing when I wrote the message
below (around 2004-11-20). I think the gdb version I was using at the
time was 6.1-3, but has since then upgraded to 6.3-4.

The gdb crash is not reproducible, though.

Simon Josefsson <email address hidden> writes:

> Btw, sometimes it crashes gdb as well:
>
> [Thread debugging using libthread_db enabled]
> [New Thread 1083202432 (LWP 441)]
> [New Thread 1112292272 (LWP 480)]
> [New Thread 1112820656 (LWP 481)]
> [New Thread 1113156528 (LWP 482)]
> /nevyn/local/gdb/gdb-6.1/gdb/lin-lwp.c:1503: internal-error: lin_lwp_wait: Assertion `iterate_over_lwps (running_callback, NULL)' failed.
> A problem internal to GDB has been detected,
> further debugging may prove unreliable.
> Quit this debugging session? (y or n) n
>
> /nevyn/local/gdb/gdb-6.1/gdb/lin-lwp.c:1503: internal-error: lin_lwp_wait: Assertion `iterate_over_lwps (running_callback, NULL)' failed.
> A problem internal to GDB has been detected,
> further debugging may prove unreliable.
> Create a core file of GDB? (y or n) y
> (gdb) bt
> Cannot fetch general-purpose registers for thread 1083202432: generic error
> (gdb) q
> The program is running. Exit anyway? (y or n) y
> Quitting: thread_db_get_info: cannot get thread info: generic error

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Message-ID: <email address hidden>
Date: Mon, 22 Nov 2004 17:36:52 +0100
From: Simon Josefsson <email address hidden>
To: <email address hidden>
Subject: Gdb crash on x86 too (was: Re: Reproducible gtk crash)

I get the same gdb crash as reported in 259276, but on x86, see below.

I was running fully upgraded Debian testing when I wrote the message
below (around 2004-11-20). I think the gdb version I was using at the
time was 6.1-3, but has since then upgraded to 6.3-4.

The gdb crash is not reproducible, though.

Simon Josefsson <email address hidden> writes:

> Btw, sometimes it crashes gdb as well:
>
> [Thread debugging using libthread_db enabled]
> [New Thread 1083202432 (LWP 441)]
> [New Thread 1112292272 (LWP 480)]
> [New Thread 1112820656 (LWP 481)]
> [New Thread 1113156528 (LWP 482)]
> /nevyn/local/gdb/gdb-6.1/gdb/lin-lwp.c:1503: internal-error: lin_lwp_wait: Assertion `iterate_over_lwps (running_callback, NULL)' failed.
> A problem internal to GDB has been detected,
> further debugging may prove unreliable.
> Quit this debugging session? (y or n) n
>
> /nevyn/local/gdb/gdb-6.1/gdb/lin-lwp.c:1503: internal-error: lin_lwp_wait: Assertion `iterate_over_lwps (running_callback, NULL)' failed.
> A problem internal to GDB has been detected,
> further debugging may prove unreliable.
> Create a core file of GDB? (y or n) y
> (gdb) bt
> Cannot fetch general-purpose registers for thread 1083202432: generic error
> (gdb) q
> The program is running. Exit anyway? (y or n) y
> Quitting: thread_db_get_info: cannot get thread info: generic error

Revision history for this message
In , Daniel Jacobowitz (drow) wrote : Re: Bug#259276: Multi-threaded program debugging failed with gdb on Sparc

Version: 6.6.dfsg.90.20070912-1

On Tue, Jul 13, 2004 at 03:13:08PM -0400, Nizar Ahmouda wrote:
> I set the severity to serious, as this seems to heavily affect sparc
> while debugging multithreaded apps, but I believe on other platforms,
> like i386 it might not show up.

Thread debugging on SPARC was broken for a while. Dave Miller fixed
it on 2006-04-07 in the GDB CVS and the fix is now in testing and unstable.

--
Daniel Jacobowitz
CodeSourcery

Changed in gdb:
status: New → Fix Released
Revision history for this message
In , Debbugs Internal Request (owner-bugs) wrote : Internal Control

# A New Hope
# A log time ago, in a galaxy far, far away
# something happened.
#
# Magically this resulted in the following
# action being taken, but this fake control
# message doesn't tell you why it happened
#
# The action:
# Bug archived.
thanks
# This fakemail brought to you by your local debbugs
# administrator

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.