Bug #60460 for threads:

Wed Aug 18 05:18:26 2010 ppisar [...] redhat.com - Ticket created

When sending signal to thread that has undefined handler, the Perl interpreter segfaults in threads.so. The reproduced is attached. However if handler is defined for other signal than you send later, then it will not segfault. I thought it's bug in threads module, but I found it depends on perl include path: petr@dhcp-0-146:/tmp/threads-1.77-build/usr/lib64/perl5 $ find . ./threads.pm ./auto ./auto/threads ./auto/threads/.packlist ./auto/threads/threads.bs ./auto/threads/threads.so ./perllocal.pod petr@dhcp-0-146:/tmp/threads-1.77-build/usr/lib64/perl5 $ perl -I . /tmp/perl_thread_kill Perl exited with active threads: 1 running and unjoined 0 finished and unjoined 0 running and detached petr@dhcp-0-146:/tmp/threads-1.77-build/usr/lib64/perl5 $ perl -I $PWD /tmp/perl_thread_kill Segmentation fault I verified perl opened the same files using strace tool. I can reproduce it on x86_64 Fedora Linux distribution (perl-5.10.0, 5.10.1, 5.12.1 with threads 1.72, 1.75, 1.77), and on i586 in Debian (perl-5.10.1, threads-1.72). I think problem is in perl itself rather than threads module. Do you have any idea what's wrong? Can you reproduce it? Some debugging can be foun on <https://bugzilla.redhat.com/show_bug.cgi?id=622190>.

Subject:

perl_thread_kill

Download perl_thread_kill
application/octet-stream 277b

Message body not shown because it is not plain text.

Wed Sep 22 22:05:51 2010 jdhedden [...] cpan.org - Correspondence added

Subject:

Thread segfault with no signal handler

On Wed Aug 18 05:18:26 2010, ppisar wrote: Show quoted text

> When sending signal to thread that has undefined handler, the Perl > interpreter segfaults in threads.so. The reproduced is attached.

Fixed with v1.78 posted to CPAN.

Wed Sep 22 22:05:51 2010 The RT System itself - Status changed from 'new' to 'open'

Wed Sep 22 22:05:52 2010 jdhedden [...] cpan.org - Status changed from 'open' to 'resolved'

Thu Sep 23 07:15:30 2010 jdhedden [...] cpan.org - Fixed in 1.78 added

Mon Sep 27 11:40:04 2010 ppisar [...] redhat.com - Correspondence added

Subject:	Re: [rt.cpan.org #60460] Thread segfault with no signal handler
Date:	Mon, 27 Sep 2010 17:39:53 +0200
To:	"Jerry D. Hedden via RT" <bug-threads [...] rt.cpan.org>
From:	Petr Pisar <ppisar [...] redhat.com>

On Wed, Sep 22, 2010 at 10:05:51PM -0400, Jerry D. Hedden via RT wrote: Show quoted text

> Fixed with v1.78 posted to CPAN.

Unfortunatelly it does not seem so: t/kill2.t .......... 1/3 # PROG: # use threads; # my $test = sub { # $SIG{'STOP'} = sub { threads->exit() }; # while(1) { sleep(1) } # }; # my $thr = threads->create($test); # threads->yield(); # $thr->detach(); # eval { # $thr->kill('STOP'); # }; # print((! $@) ? 'ok' : 'not ok'); # EXPECTED: # ok # GOT: # not ok # STATUS: 0 # Failed at t/kill2.t line 51 t/kill2.t .......... Failed 1/3 subtests Tested with perl v5.10.1 and v5.12.2 on x86_64 (Fedora 13, Fedora 14). What perl version do you use or which data do you need from me? -- Petr

Mon Sep 27 11:40:06 2010 The RT System itself - Status changed from 'resolved' to 'open'

Mon Sep 27 11:52:35 2010 jdhedden [...] cpan.org - Correspondence added

Subject:	Re: [rt.cpan.org #60460] Thread segfault with no signal handler
Date:	Mon, 27 Sep 2010 11:51:56 -0400
To:	bug-threads [...] rt.cpan.org
From:	"Jerry D. Hedden" <jdhedden [...] cpan.org>

This is an issue with the tests and not the code. I fixed them with v1.79. Try that. On Mon, Sep 27, 2010 at 11:40, Petr Pisar via RT <bug-threads@rt.cpan.org> wrote: Show quoted text

> Queue: threads > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=60460 > > > On Wed, Sep 22, 2010 at 10:05:51PM -0400, Jerry D. Hedden via RT wrote:

>> Fixed with v1.78 posted to CPAN.

> > Unfortunatelly it does not seem so: > > t/kill2.t .......... 1/3 # PROG: > # use threads; > # my $test = sub { > # $SIG{'STOP'} = sub { threads->exit() }; > # while(1) { sleep(1) } > # }; > # my $thr = threads->create($test); > # threads->yield(); > # $thr->detach(); > # eval { > # $thr->kill('STOP'); > # }; > # print((! $@) ? 'ok' : 'not ok'); > # EXPECTED: > # ok > # GOT: > # not ok > # STATUS: 0 > # Failed at t/kill2.t line 51 > t/kill2.t .......... Failed 1/3 subtests > > Tested with perl v5.10.1 and v5.12.2 on x86_64 (Fedora 13, Fedora 14). > > What perl version do you use or which data do you need from me? > > -- Petr > >

Fri Oct 08 07:55:58 2010 ppisar [...] redhat.com - Correspondence added

Subject:	Re: [rt.cpan.org #60460] Thread segfault with no signal handler
Date:	Fri, 8 Oct 2010 13:55:46 +0200
To:	"Jerry D. Hedden via RT" <bug-threads [...] rt.cpan.org>
From:	Petr Pisar <ppisar [...] redhat.com>

On Mon, Sep 27, 2010 at 11:52:36AM -0400, Jerry D. Hedden via RT wrote: Show quoted text

> <URL: https://rt.cpan.org/Ticket/Display.html?id=60460 > > > This is an issue with the tests and not the code. I fixed them with > v1.79. Try that. >

Thanks. The test in 1.79 works for me. Also I'm not able to reproduce the `-I .' versus `-I $PWD' schism. I do not know why. Feel free to close this bug are resolved as the unhandled signal does not segfault perl anymore. -- Petr

Fri Oct 08 09:22:27 2010 jdhedden [...] cpan.org - Status changed from 'open' to 'resolved'

Fri Oct 08 09:22:28 2010 jdhedden [...] cpan.org - Fixed in 1.79 added

Fri Oct 08 09:22:28 2010 jdhedden [...] cpan.org - Fixed in 1.78 deleted