Bug 150911 - exiting pthreads from an xterm process kills the xterm
Summary: exiting pthreads from an xterm process kills the xterm
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Fedora
Classification: Fedora
Component: kernel
Version: 3
Hardware: i686
OS: Linux
medium
high
Target Milestone: ---
Assignee: Dave Jones
QA Contact: Brian Brock
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2005-03-11 22:18 UTC by Adam Spanberger
Modified: 2015-01-04 22:17 UTC (History)
2 users (show)

Fixed In Version:
Clone Of:
Environment:
Last Closed: 2005-03-14 16:59:58 UTC
Type: ---
Embargoed:


Attachments (Terms of Use)
Sample code to generate the problem (475 bytes, text/plain)
2005-03-11 22:18 UTC, Adam Spanberger
no flags Details

Description Adam Spanberger 2005-03-11 22:18:01 UTC
Description of problem:
I have a simple application that creates pthreads.  The behavior is the same 
whether the thread is detached or whether I try to join to it.  When I run the 
application from a normal terminal, it works as expected.  If I try to run it 
in its own xterm (ie. xterm -e appName), the xterm dies as soon as the first 
thread exits.

Version-Release number of selected component (if applicable):
2.6.9-1.667

How reproducible:
Very.  I was also able to reproduce it by running the application using a 
dejagnu script.

Steps to Reproduce:
1.gcc threadTest.c -lpthread 
2.xterm -e ./a.out
3.
  
Actual results:
xterm exits after first thread exits.

Expected results:
xterm exits when all five threads of the parent process exit

Additional info:
LD_ASSUME_KERNEL <= 2.4.19 works
LD_ASSUME_KERNEL >= 2.4.20 does not work

I'm providing a very simple test case that demonstrates the problem.

Comment 1 Adam Spanberger 2005-03-11 22:18:02 UTC
Created attachment 111904 [details]
Sample code to generate the problem

Comment 2 Dave Jones 2005-03-11 22:19:51 UTC
is this repeatable with the errata kernel ?
2.6.9-1.667 is ancient now.

Comment 3 Adam Spanberger 2005-03-14 14:44:55 UTC
Updating to 2.6.10-1.770 fixed the problem.  Thanks for your help.


Note You need to log in before you can comment on or make changes to this bug.