On Mon, Feb 07, 2005 at 05:39:03PM +0000, Steve Hay wrote:
What is it with tests called "ithreads.t"?
Here's what I get @rev151728:
This started failing in my Linux builds too as of 2005-02-05 on:
Perl 5.8.5 on ppc, Perl 5.8.5 on i386, Perl 5.8.3 on i386
but is passing on:
Perl 5.8.0 on i386, Perl 5.8.0 on x86_64
threads execution context switching.
# expected: thread 1 # thread 2 # thread 3 # thread 4 # parent # received: thread 1 # thread 2 # thread 3 # parentthread 4 not ok 1
Should be fixed in svn. The order is not important for this test.
The better fix would be to join the threads (blocking) before printing in the parent, but that's not the point of that test. As XXX states I get leaks when adding join(), so I've left those out until someone else or I get the time to figure it out.
But this raises an interesting question. What happens if a spawned thread didn't have a chance to print before the parent handler has returned. Obviously the correct way to program that kind of distributed handler is to block in the parent until all threads are done. But still...
-- __________________________________________________________________ Stas Bekman JAm_pH ------> Just Another mod_perl Hacker http://stason.org/ mod_perl Guide ---> http://perl.apache.org mailto:[EMAIL PROTECTED] http://use.perl.org http://apacheweek.com http://modperlbook.org http://apache.org http://ticketmaster.com
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
