system/5126: threads + fork() = child process dies with SIGPROF

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

system/5126: threads + fork() = child process dies with SIGPROF

Adam Montague
>Number:         5126
>Category:       system
>Synopsis:       threads + fork() = child process dies with SIGPROF
>Confidential:   yes
>Severity:       serious
>Priority:       medium
>Responsible:    bugs
>State:          open
>Quarter:        
>Keywords:      
>Date-Required:
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Fri May 19 20:10:01 GMT 2006
>Closed-Date:
>Last-Modified:
>Originator:     Adam Montague
>Release:        3.9
>Organization:
net
>Environment:
       
        System      : OpenBSD 3.9
        Architecture: OpenBSD.i386
        Machine     : i386
>Description:
        Threaded apps that call fork() don't have their timer disabled
and still get sent SIGPROF, resulting in "child died with signal 27"
style errors.
>How-To-Repeat:
        see the test code here from the freebsd bug report:
http://lists.freebsd.org/pipermail/freebsd-bugs/2003-October/003797.html
>Fix:


>Release-Note:
>Audit-Trail:
>Unformatted:

Reply | Threaded
Open this post in threaded view
|

Re: system/5126: threads + fork() = child process dies with SIGPROF

Adam Montague
It seems that the child died with signal 27 only happens on a bsd.mp
kernel.  The only app I know of that reliably reproduces this is pike's
test suite.

Here's the output of the pike testsuite on a 3.9 bsd.mp machine:
http://pike.ida.liu.se/generated/pikefarm/7.7/1552_253/verifylog.txt

On Fri, 19 May 2006 14:15:01 -0600 (MDT) Gnats <[hidden email]> wrote:

> Thank you very much for your problem report.
> It has the internal identification `system/5126'.
> The individual assigned to look at your
> report is: bugs.
>
> >Category:       system
> >Responsible:    bugs
> >Synopsis:       threads + fork() = child process dies with SIGPROF
> >Arrival-Date:   Fri May 19 20:10:01 GMT 2006

Reply | Threaded
Open this post in threaded view
|

Re: system/5126: threads + fork() = child process dies with SIGPROF

Adam Montague
In reply to this post by Adam Montague
The following reply was made to PR system/5126; it has been noted by GNATS.

From: Adam Montague <[hidden email]>
To: [hidden email], [hidden email]
Cc:  
Subject: Re: system/5126: threads + fork() = child process dies with SIGPROF
Date: Tue, 23 May 2006 15:29:05 -0400

 It seems that the child died with signal 27 only happens on a bsd.mp
 kernel.  The only app I know of that reliably reproduces this is pike's
 test suite.
 
 Here's the output of the pike testsuite on a 3.9 bsd.mp machine:
 http://pike.ida.liu.se/generated/pikefarm/7.7/1552_253/verifylog.txt
 
 On Fri, 19 May 2006 14:15:01 -0600 (MDT) Gnats <[hidden email]> wrote:
 
 > Thank you very much for your problem report.
 > It has the internal identification `system/5126'.
 > The individual assigned to look at your
 > report is: bugs.
 >
 > >Category:       system
 > >Responsible:    bugs
 > >Synopsis:       threads + fork() = child process dies with SIGPROF
 > >Arrival-Date:   Fri May 19 20:10:01 GMT 2006