[rtems commit] score: Fix _Thread_Start_multitasking() on SMP

[Sebastian Huber]

Module:    rtems
Branch:    master
Commit:    152ad7d1cc9d26eed0186cbb4a40b3b84bf1fee8
Changeset: http://git.rtems.org/rtems/commit/?id=152ad7d1cc9d26eed0186cbb4a40b3b84bf1fee8

Author:    Sebastian Huber <sebastian.huber at embedded-brains.de>
Date:      Tue Jun 24 09:31:14 2014 +0200

score: Fix _Thread_Start_multitasking() on SMP

The _CPU_Context_Restart_self() implementations usually assume that self
context is executing.

FIXME: We have a race condition in _Thread_Start_multitasking() in case
another thread already performed scheduler operations and moved the heir
thread to another processor.  The time frame for this is likely too
small to be practically relevant.

---

 cpukit/score/src/threadstartmultitasking.c |   13 +++++++++++++
 1 files changed, 13 insertions(+), 0 deletions(-)

diff --git a/cpukit/score/src/threadstartmultitasking.c b/cpukit/score/src/threadstartmultitasking.c
index c1c8725..4467937 100644
--- a/cpukit/score/src/threadstartmultitasking.c
+++ b/cpukit/score/src/threadstartmultitasking.c
@@ -61,6 +61,19 @@ void _Thread_Start_multitasking( void )
 
   _Profiling_Thread_dispatch_disable( cpu_self, 0 );
 
+#if defined(RTEMS_SMP)
+  /*
+   * The _CPU_Context_Restart_self() implementations usually assume that self
+   * context is executing.
+   *
+   * FIXME: We have a race condition here in case another thread already
+   * performed scheduler operations and moved our heir thread to another
+   * processor.  The time frame for this is likely too small to be practically
+   * relevant.
+   */
+  _CPU_Context_Set_is_executing( &heir->Registers, true );
+#endif
+
 #if defined(_CPU_Start_multitasking)
   _CPU_Start_multitasking( &heir->Registers );
 #else