[Bug 1846] posix timing tests: psxtmsem01

[bugzilla-daemon at rtems.org]

https://www.rtems.org/bugzilla/show_bug.cgi?id=1846

Joel Sherrill <joel.sherrill at oarcorp.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |joel.sherrill at oarcorp.com
         Resolution|                            |FIXED

--- Comment #1 from Joel Sherrill <joel.sherrill at oarcorp.com> 2011-07-22 12:06:51 CDT ---
There was some confusion on both our parts over how to get a nested open/close
situation where the close does not actually delete the semaphore.  I added a
couple of sem_unlink() benchmark cases along using a different variable for the
second open.  

The times for close which didn't delete and a close which did delete should
have been different.

Also added cases for sem_unlink() which deletes the underlying semaphore.

*** POSIX TIME TEST PSXTMSEM01 ***
sem_init 308
sem_destroy 392
sem_open (first open 'O_CREAT') 483
sem_open (second open 'O_EXCL') 389
sem_close (named first/nested close) 219
sem_unlink (does not delete) 387
sem_close (named second close) 392
sem_unlink (deletes semaphore) 550
*** END OF POSIX TIME TEST PSXTMSEM01 ***

As a matter of understanding named/unnamed semaphores and how the unlink works
with that, you should look at the code.

FWIW anytime two cases which are different have similar times, you should ask
yourself if the test is doing what you expect. :)

-- 
Configure bugmail: https://www.rtems.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.