[Bug 296] Unlimited test does not deal with system tasks

[bugzilla-daemon at rtems.org]

http://www.rtems.org/bugzilla/show_bug.cgi?id=296


Chris Johns <chrisj at rtems.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |ASSIGNED

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |CLOSED
             Status|CLOSED                      |WAITING
             Status|ASSIGNED                    |RESOLVED
           Platform|                            |All
         Resolution|                            |FIXED
   Target Milestone|---                         |4.9


The auto-extend disable test (test1) assumes only the test tasks are running.
This may not be the case as BSP created tasks can be running. For example the
monitor, network tasks, and IO driver tasks.

Test1 tries to make the configuration table number of tasks. If the BSP has
consumed a few of these the test fails. The failure is a result of test1
actually passing.

Also, the test (infact all tests) assumes it can control the configuration
table. This could/should be part of the BSP rather than the application.

Release:
RTEMS-4.5.0 + CVS

Environment:
Cybertec SFP-101 (Coldfire), bare-5200, tests built external to RTEMS.


--- Comment #1 from Chris Johns <chrisj at rtems.org>  2002-10-29 22:49:00 ---
Fix:
Make test1 create only the remaining number of tasks needed to get the RTEMS_TOO_MANY error.

--- Comment #2 from Chris Johns <chrisj at rtems.org>  2002-10-30 23:17:08 ---
State-Changed-From-To: open->analyzed
State-Changed-Why: Patch attached. The patch is not compiled or tested.


--- Comment #3 from Joel Sherrill <joel.sherrill at oarcorp.com>  2003-01-03 14:51:49 ---
State-Changed-From-To: analyzed->closed
State-Changed-Why: Patch applied.


--- Comment #4 from Joel Sherrill <joel.sherrill at oarcorp.com>  2003-01-03 16:57:11 ---
State-Changed-From-To: closed->feedback
State-Changed-Why: Sorry to reopen this but the test does not compile with the
    attached patch.  This is the error:
    
    In file included from ../../../../../../../current/c/src/tests/samples/unlimited/system.h:56,
                     from ../../../../../../../current/c/src/tests/samples/unlimited/init.c:25:
    ../../../../../dmv177/lib/include/confdefs.h:970: initializer element is not constant
    ../../../../../dmv177/lib/include/confdefs.h:970: (near initialization for `Configuration_RTEMS_API.maximum_tasks')
    ../../../../../dmv177/lib/include/confdefs.h:1017: initializer element is not constant
    ../../../../../dmv177/lib/include/confdefs.h:1017: (near initialization for `Configuration.work_space_size')
    In file included from ../../../../../../../current/c/src/tests/samples/unlimited/system.h:56,
                     from ../../../../../../../current/c/src/tests/samples/unlimited/init.c:25:
    ../../../../../dmv177/lib/include/confdefs.h:1057:6: "->" is not valid in #if expressions


--- Comment #5 from Joel Sherrill <joel.sherrill at oarcorp.com>  2008-08-05 12:18:51 ---
unlimited reliably compiles and runs now.  This PR has long been resolved


-- 
Configure bugmail: http://www.rtems.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.