While cpu_self->thread_dispatch_disable_level shouldn't ever be zero, it would be better to check it before doing the decrement.
Changes in v2: * Modified the asserts as requested in https://lists.rtems.org/pipermail/devel/2015-February/009821.html. --- cpukit/score/src/threaddispatchdisablelevel.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/cpukit/score/src/threaddispatchdisablelevel.c b/cpukit/score/src/threaddispatchdisablelevel.c index 158fc39..6043fe7 100644 --- a/cpukit/score/src/threaddispatchdisablelevel.c +++ b/cpukit/score/src/threaddispatchdisablelevel.c @@ -107,11 +107,15 @@ uint32_t _Thread_Dispatch_decrement_disable_level( void ) cpu_self = _Per_CPU_Get(); disable_level = cpu_self->thread_dispatch_disable_level; + _Assert( disable_level > 0); --disable_level; cpu_self->thread_dispatch_disable_level = disable_level; _Giant_Do_release( cpu_self ); - _Assert( disable_level != 0 || _Giant.owner_cpu != cpu_self ); + + _Assert( ( disable_level == 0 && _Giant.owner_cpu != cpu_self ) + || ( disable_level != 0 && _Giant.owner_cpu == cpu_self ) + ); _Profiling_Thread_dispatch_enable( cpu_self, disable_level ); _ISR_Enable_without_giant( isr_level ); -- 2.3.4 _______________________________________________ devel mailing list devel@rtems.org http://lists.rtems.org/mailman/listinfo/devel
