freertos(IDF): Use common macros for SMP specific critical sections

In IDF FreeRTOS, when building for SMP, there are numerous functions
which require different critical sections when compared to single-core. This
commit encapsulates those difference into a common set of macros whose
behavior depends on "configNUM_CORES > 1". As such...

- Vanilla behavior has been restored for some functions when building for
  single core (i.e., used to call taskENTER_CRITICAL, now disables interrupts
  mactching vanilla behavior).
- Reduces number of "#ifdef (configNUM_CORES > 1)" in functions
- Any SMP only critical sections are now wrapped by
  "#ifdef (configNUM_CORES > 1)" and properly documented via comments.
This commit is contained in:
Darian Leung
2022-11-14 15:27:00 +08:00
parent 4c1ff6016a
commit 087e4318a6
7 changed files with 372 additions and 367 deletions

View File

@@ -403,12 +403,9 @@ PRIVILEGED_DATA static List_t * volatile pxDelayedTaskList; /*< Poi
PRIVILEGED_DATA static List_t * volatile pxOverflowDelayedTaskList; /*< Points to the delayed task list currently being used to hold tasks that have overflowed the current tick count. */
PRIVILEGED_DATA static List_t xPendingReadyList[ configNUM_CORES ]; /*< Tasks that have been readied while the scheduler was suspended. They will be moved to the ready list when the scheduler is resumed. */
#ifdef ESP_PLATFORM
/* Spinlock required for SMP critical sections. This lock protects all of the
* kernel's data structures such as various tasks lists, flags, and tick counts. */
PRIVILEGED_DATA static portMUX_TYPE xKernelLock = portMUX_INITIALIZER_UNLOCKED;
#endif // ESP_PLATFORM
PRIVILEGED_DATA static portMUX_TYPE xKernelLock = portMUX_INITIALIZER_UNLOCKED;
#if ( INCLUDE_vTaskDelete == 1 )
@@ -1537,11 +1534,7 @@ static void prvAddNewTaskToReadyList( TCB_t * pxNewTCB )
configASSERT( ( xTimeIncrement > 0U ) );
configASSERT( xTaskGetSchedulerState() != taskSCHEDULER_SUSPENDED );
#ifdef ESP_PLATFORM /* IDF-3755 */
taskENTER_CRITICAL( &xKernelLock );
#else
vTaskSuspendAll();
#endif // ESP_PLATFORM
prvENTER_CRITICAL_OR_SUSPEND_ALL( &xKernelLock );
{
/* Minor optimisation. The tick count cannot change in this
* block. */
@@ -1597,12 +1590,7 @@ static void prvAddNewTaskToReadyList( TCB_t * pxNewTCB )
mtCOVERAGE_TEST_MARKER();
}
}
#ifdef ESP_PLATFORM /* IDF-3755 */
taskEXIT_CRITICAL( &xKernelLock );
xAlreadyYielded = pdFALSE;
#else
xAlreadyYielded = xTaskResumeAll();
#endif // ESP_PLATFORM
xAlreadyYielded = prvEXIT_CRITICAL_OR_RESUME_ALL( &xKernelLock );
/* Force a reschedule if xTaskResumeAll has not already done so, we may
* have put ourselves to sleep. */
@@ -1631,11 +1619,7 @@ static void prvAddNewTaskToReadyList( TCB_t * pxNewTCB )
if( xTicksToDelay > ( TickType_t ) 0U )
{
configASSERT( xTaskGetSchedulerState() != taskSCHEDULER_SUSPENDED );
#ifdef ESP_PLATFORM /* IDF-3755 */
taskENTER_CRITICAL( &xKernelLock );
#else
vTaskSuspendAll();
#endif // ESP_PLATFORM
prvENTER_CRITICAL_OR_SUSPEND_ALL( &xKernelLock );
{
traceTASK_DELAY();
@@ -1648,12 +1632,7 @@ static void prvAddNewTaskToReadyList( TCB_t * pxNewTCB )
* executing task. */
prvAddCurrentTaskToDelayedList( xTicksToDelay, pdFALSE );
}
#ifdef ESP_PLATFORM /* IDF-3755 */
taskEXIT_CRITICAL( &xKernelLock );
xAlreadyYielded = pdFALSE;
#else
xAlreadyYielded = xTaskResumeAll();
#endif // ESP_PLATFORM
xAlreadyYielded = prvEXIT_CRITICAL_OR_RESUME_ALL( &xKernelLock );
}
else
{
@@ -2836,11 +2815,7 @@ char * pcTaskGetName( TaskHandle_t xTaskToQuery ) /*lint !e971 Unqualified char
/* Task names will be truncated to configMAX_TASK_NAME_LEN - 1 bytes. */
configASSERT( strlen( pcNameToQuery ) < configMAX_TASK_NAME_LEN );
#ifdef ESP_PLATFORM /* IDF-3755 */
taskENTER_CRITICAL( &xKernelLock );
#else
vTaskSuspendAll();
#endif // ESP_PLATFORM
prvENTER_CRITICAL_OR_SUSPEND_ALL( &xKernelLock );
{
/* Search the ready lists. */
do
@@ -2886,11 +2861,7 @@ char * pcTaskGetName( TaskHandle_t xTaskToQuery ) /*lint !e971 Unqualified char
}
#endif
}
#ifdef ESP_PLATFORM /* IDF-3755 */
taskEXIT_CRITICAL( &xKernelLock );
#else
( void ) xTaskResumeAll();
#endif // ESP_PLATFORM
( void ) prvEXIT_CRITICAL_OR_RESUME_ALL( &xKernelLock );
return pxTCB;
}
@@ -2906,11 +2877,7 @@ char * pcTaskGetName( TaskHandle_t xTaskToQuery ) /*lint !e971 Unqualified char
{
UBaseType_t uxTask = 0, uxQueue = configMAX_PRIORITIES;
#ifdef ESP_PLATFORM /* IDF-3755 */
taskENTER_CRITICAL( &xKernelLock );
#else
vTaskSuspendAll();
#endif // ESP_PLATFORM
prvENTER_CRITICAL_OR_SUSPEND_ALL( &xKernelLock );
{
/* Is there a space in the array for each task in the system? */
if( uxArraySize >= uxCurrentNumberOfTasks )
@@ -2969,11 +2936,7 @@ char * pcTaskGetName( TaskHandle_t xTaskToQuery ) /*lint !e971 Unqualified char
mtCOVERAGE_TEST_MARKER();
}
}
#ifdef ESP_PLATFORM /* IDF-3755 */
taskEXIT_CRITICAL( &xKernelLock );
#else
( void ) xTaskResumeAll();
#endif // ESP_PLATFORM
( void ) prvEXIT_CRITICAL_OR_RESUME_ALL( &xKernelLock );
return uxTask;
}
@@ -3008,10 +2971,12 @@ char * pcTaskGetName( TaskHandle_t xTaskToQuery ) /*lint !e971 Unqualified char
void vTaskStepTick( const TickType_t xTicksToJump )
{
#ifdef ESP_PLATFORM
/* For SMP, we require a critical section to access xTickCount */
#if ( configNUM_CORES > 1 )
/* Although this is called with the scheduler suspended. For SMP, we
* still need to take the kernel lock to access xTickCount. */
taskENTER_CRITICAL( &xKernelLock );
#endif
#endif /* configNUM_CORES > 1 */
/* Correct the tick count value after a period during which the tick
* was suppressed. Note this does *not* call the tick hook function for
@@ -3019,9 +2984,11 @@ char * pcTaskGetName( TaskHandle_t xTaskToQuery ) /*lint !e971 Unqualified char
configASSERT( ( xTickCount + xTicksToJump ) <= xNextTaskUnblockTime );
xTickCount += xTicksToJump;
traceINCREASE_TICK_COUNT( xTicksToJump );
#ifdef ESP_PLATFORM
#if ( configNUM_CORES > 1 )
/* Release the previously taken kernel lock. */
taskEXIT_CRITICAL( &xKernelLock );
#endif
#endif /* configNUM_CORES > 1 */
}
#endif /* configUSE_TICKLESS_IDLE */
@@ -3042,16 +3009,17 @@ BaseType_t xTaskCatchUpTicks( TickType_t xTicksToCatchUp )
/* Use xPendedTicks to mimic xTicksToCatchUp number of ticks occurring when
* the scheduler is suspended so the ticks are executed in xTaskResumeAll(). */
vTaskSuspendAll();
#ifdef ESP_PLATFORM
#if ( configNUM_CORES > 1 )
/* For SMP, we still require a critical section to access xPendedTicks even
* if the scheduler is disabled. */
/* Although the scheduler is suspended. For SMP, we still need to take
* the kernel lock to access xPendedTicks. */
taskENTER_CRITICAL( &xKernelLock );
xPendedTicks += xTicksToCatchUp;
#endif /* configNUM_CORES > 1 */
xPendedTicks += xTicksToCatchUp;
#if ( configNUM_CORES > 1 )
/* Release the previously taken kernel lock. */
taskEXIT_CRITICAL( &xKernelLock );
#else // ESP_PLATFORM
xPendedTicks += xTicksToCatchUp;
#endif // ESP_PLATFORM
#endif /* configNUM_CORES > 1 */
xYieldOccurred = xTaskResumeAll();
return xYieldOccurred;
@@ -3067,11 +3035,7 @@ BaseType_t xTaskCatchUpTicks( TickType_t xTicksToCatchUp )
configASSERT( pxTCB );
#ifdef ESP_PLATFORM /* IDF-3755 */
taskENTER_CRITICAL( &xKernelLock );
#else
vTaskSuspendAll();
#endif // ESP_PLATFORM
prvENTER_CRITICAL_OR_SUSPEND_ALL( &xKernelLock );
{
/* A task can only be prematurely removed from the Blocked state if
* it is actually in the Blocked state. */
@@ -3134,11 +3098,7 @@ BaseType_t xTaskCatchUpTicks( TickType_t xTicksToCatchUp )
xReturn = pdFAIL;
}
}
#ifdef ESP_PLATFORM /* IDF-3755 */
taskEXIT_CRITICAL( &xKernelLock );
#else
( void ) xTaskResumeAll();
#endif // ESP_PLATFORM
( void ) prvEXIT_CRITICAL_OR_RESUME_ALL( &xKernelLock );
return xReturn;
}
@@ -3508,26 +3468,18 @@ BaseType_t xTaskIncrementTick( void )
{
TCB_t * pxTCB;
TaskHookFunction_t xReturn;
UBaseType_t uxSavedInterruptStatus;
/* If xTask is NULL then set the calling task's hook. */
pxTCB = prvGetTCBFromHandle( xTask );
/* Save the hook function in the TCB. A critical section is required as
* the value can be accessed from an interrupt. */
#if ( configNUM_CORES > 1 )
taskENTER_CRITICAL_ISR( &xKernelLock );
#else
UBaseType_t uxSavedInterruptStatus;
uxSavedInterruptStatus = portSET_INTERRUPT_MASK_FROM_ISR();
#endif
prvENTER_CRITICAL_OR_MASK_ISR( &xKernelLock, uxSavedInterruptStatus );
{
xReturn = pxTCB->pxTaskTag;
}
#if ( configNUM_CORES > 1 )
taskEXIT_CRITICAL_ISR( &xKernelLock );
#else
portCLEAR_INTERRUPT_MASK_FROM_ISR( uxSavedInterruptStatus );
#endif
prvEXIT_CRITICAL_OR_UNMASK_ISR( &xKernelLock, uxSavedInterruptStatus );
return xReturn;
}
@@ -3768,8 +3720,12 @@ void vTaskPlaceOnEventList( List_t * const pxEventList,
{
configASSERT( pxEventList );
/* Take the kernel lock as we are about to access the task lists. */
taskENTER_CRITICAL( &xKernelLock );
#if ( configNUM_CORES > 1 )
/* In SMP, we need to take the kernel lock as we are about to access the
* task lists. */
taskENTER_CRITICAL( &xKernelLock );
#endif /* configNUM_CORES > 1 */
/* THIS FUNCTION MUST BE CALLED WITH EITHER INTERRUPTS DISABLED OR THE
* SCHEDULER SUSPENDED AND THE QUEUE BEING ACCESSED LOCKED. */
@@ -3782,7 +3738,10 @@ void vTaskPlaceOnEventList( List_t * const pxEventList,
prvAddCurrentTaskToDelayedList( xTicksToWait, pdTRUE );
taskEXIT_CRITICAL( &xKernelLock );
#if ( configNUM_CORES > 1 )
/* Release the previously taken kernel lock. */
taskEXIT_CRITICAL( &xKernelLock );
#endif /* configNUM_CORES > 1 */
}
/*-----------------------------------------------------------*/
@@ -3792,14 +3751,18 @@ void vTaskPlaceOnUnorderedEventList( List_t * pxEventList,
{
configASSERT( pxEventList );
/* Take the kernel lock as we are about to access the task lists. */
taskENTER_CRITICAL( &xKernelLock );
#if ( configNUM_CORES > 1 )
/* THIS FUNCTION MUST BE CALLED WITH THE SCHEDULER SUSPENDED. It is used by
* the event groups implementation. */
/* In SMP, the event groups haven't suspended the scheduler at this
* point. We need to take the kernel lock instead as we are about to
* access the task lists. */
taskENTER_CRITICAL( &xKernelLock );
#else /* configNUM_CORES > 1 */
/* Note. We currently don't always suspend the scheduler. Todo: IDF-3755
* configASSERT( uxSchedulerSuspended[ xPortGetCoreID() ] != 0 ); */
/* THIS FUNCTION MUST BE CALLED WITH THE SCHEDULER SUSPENDED. It is used by
* the event groups implementation. */
configASSERT( uxSchedulerSuspended[ 0 ] != 0 );
#endif /* configNUM_CORES > 1 */
/* Store the item value in the event list item. It is safe to access the
* event list item here as interrupts won't access the event list item of a
@@ -3815,7 +3778,10 @@ void vTaskPlaceOnUnorderedEventList( List_t * pxEventList,
prvAddCurrentTaskToDelayedList( xTicksToWait, pdTRUE );
taskEXIT_CRITICAL( &xKernelLock );
#if ( configNUM_CORES > 1 )
/* Release the previously taken kernel lock. */
taskEXIT_CRITICAL( &xKernelLock );
#endif /* configNUM_CORES > 1 */
}
/*-----------------------------------------------------------*/
@@ -3827,8 +3793,12 @@ void vTaskPlaceOnUnorderedEventList( List_t * pxEventList,
{
configASSERT( pxEventList );
/* Take the kernel lock as we are about to access the task lists. */
taskENTER_CRITICAL( &xKernelLock );
#if ( configNUM_CORES > 1 )
/* In SMP, we need to take the kernel lock as we are about to access
* the task lists. */
taskENTER_CRITICAL( &xKernelLock );
#endif /* configNUM_CORES > 1 */
/* This function should not be called by application code hence the
* 'Restricted' in its name. It is not part of the public API. It is
@@ -3853,7 +3823,10 @@ void vTaskPlaceOnUnorderedEventList( List_t * pxEventList,
traceTASK_DELAY_UNTIL( ( xTickCount + xTicksToWait ) );
prvAddCurrentTaskToDelayedList( xTicksToWait, xWaitIndefinitely );
taskEXIT_CRITICAL( &xKernelLock );
#if ( configNUM_CORES > 1 )
/* Release the previously taken kernel lock. */
taskEXIT_CRITICAL( &xKernelLock );
#endif /* configNUM_CORES > 1 */
}
#endif /* configUSE_TIMERS */
@@ -3865,12 +3838,24 @@ BaseType_t xTaskRemoveFromEventList( const List_t * const pxEventList )
BaseType_t xReturn;
/* THIS FUNCTION MUST BE CALLED FROM A CRITICAL SECTION. It can also be
* called from a critical section within an ISR.
*
* However, we still need to take the kernel lock as we are about to access
* kernel data structures. Note that we use the ISR version of the macro as
* this function could be called from an ISR critical section. */
taskENTER_CRITICAL_ISR( &xKernelLock );
* called from a critical section within an ISR. */
#if ( configNUM_CORES > 1 )
/* In SMP, we need to take the kernel lock (even if the caller is
* already in a critical section by taking a different lock) as we are
* about to access the task lists, which are protected by the kernel
* lock. This function can also be called from an ISR context, so we
* need to check whether we are in an ISR.*/
if( portCHECK_IF_IN_ISR() == pdFALSE )
{
taskENTER_CRITICAL( &xKernelLock );
}
else
{
taskENTER_CRITICAL_ISR( &xKernelLock );
}
#endif /* configNUM_CORES > 1 */
{
/* Before taking the kernel lock, another task/ISR could have already
* emptied the pxEventList. So we insert a check here to see if
@@ -3965,13 +3950,23 @@ BaseType_t xTaskRemoveFromEventList( const List_t * const pxEventList )
xReturn = pdFALSE;
}
}
taskEXIT_CRITICAL_ISR( &xKernelLock );
#if ( configNUM_CORES > 1 )
/* Release the previously taken kernel lock. */
if( portCHECK_IF_IN_ISR() == pdFALSE )
{
taskEXIT_CRITICAL( &xKernelLock );
}
else
{
taskEXIT_CRITICAL_ISR( &xKernelLock );
}
#endif /* configNUM_CORES > 1 */
return xReturn;
}
/*-----------------------------------------------------------*/
#ifdef ESP_PLATFORM
#if ( configNUM_CORES > 1 )
void vTaskTakeKernelLock( void )
{
/* We call the tasks.c critical section macro to take xKernelLock */
@@ -3983,7 +3978,7 @@ BaseType_t xTaskRemoveFromEventList( const List_t * const pxEventList )
/* We call the tasks.c critical section macro to release xKernelLock */
taskEXIT_CRITICAL( &xKernelLock );
}
#endif // ESP_PLATFORM
#endif /* configNUM_CORES > 1 */
void vTaskRemoveFromUnorderedEventList( ListItem_t * pxEventListItem,
const TickType_t xItemValue )
@@ -3991,14 +3986,17 @@ void vTaskRemoveFromUnorderedEventList( ListItem_t * pxEventListItem,
TCB_t * pxUnblockedTCB;
BaseType_t xCurCoreID = xPortGetCoreID();
/* THIS FUNCTION MUST BE CALLED WITH THE KERNEL LOCK ALREADY TAKEN.
* It is used by the event flags implementation, thus those functions
* should call vTaskTakeKernelLock() before calling this function. */
#if ( configNUM_CORES > 1 )
/*
* Todo: IDF-5785
* configASSERT( uxSchedulerSuspended[ xCurCoreID ] != pdFALSE );
*/
/* THIS FUNCTION MUST BE CALLED WITH THE KERNEL LOCK ALREADY TAKEN.
* It is used by the event flags implementation, thus those functions
* should call vTaskTakeKernelLock() before calling this function. */
#else /* configNUM_CORES > 1 */
/* THIS FUNCTION MUST BE CALLED WITH THE SCHEDULER SUSPENDED. It is used by
* the event flags implementation. */
configASSERT( uxSchedulerSuspended != pdFALSE );
#endif /* configNUM_CORES > 1 */
/* Store the new item value in the event list. */
listSET_LIST_ITEM_VALUE( pxEventListItem, xItemValue | taskEVENT_LIST_ITEM_VALUE_IN_USE );
@@ -4066,18 +4064,19 @@ void vTaskInternalSetTimeOutState( TimeOut_t * const pxTimeOut )
* On a single core configuration, this problem doesn't appear as this function is meant to be called from
* a critical section, disabling the (tick) interrupts.
*/
#if ( ( ESP_PLATFORM == 1 ) && ( configNUM_CORES > 1 ) )
#if ( configNUM_CORES > 1 )
configASSERT( pxTimeOut );
taskENTER_CRITICAL( &xKernelLock );
#endif // ( ( ESP_PLATFORM == 1 ) && ( configNUM_CORES > 1 ) )
#endif /* configNUM_CORES > 1 */
/* For internal use only as it does not use a critical section. */
pxTimeOut->xOverflowCount = xNumOfOverflows;
pxTimeOut->xTimeOnEntering = xTickCount;
#if ( ( ESP_PLATFORM == 1 ) && ( configNUM_CORES > 1 ) )
#if ( configNUM_CORES > 1 )
/* Release the previously taken kernel lock. */
taskEXIT_CRITICAL( &xKernelLock );
#endif // ( ( ESP_PLATFORM == 1 ) && ( configNUM_CORES > 1 ) )
#endif /* configNUM_CORES > 1 */
}
/*-----------------------------------------------------------*/
@@ -4288,11 +4287,7 @@ static portTASK_FUNCTION( prvIdleTask, pvParameters )
if( xExpectedIdleTime >= configEXPECTED_IDLE_TIME_BEFORE_SLEEP )
{
#ifdef ESP_PLATFORM /* IDF-3755 */
taskENTER_CRITICAL( &xKernelLock );
#else
vTaskSuspendAll();
#endif // ESP_PLATFORM
prvENTER_CRITICAL_OR_SUSPEND_ALL( &xKernelLock );
{
/* Now the scheduler is suspended, the expected idle
* time can be sampled again, and this time its value can
@@ -4316,11 +4311,7 @@ static portTASK_FUNCTION( prvIdleTask, pvParameters )
mtCOVERAGE_TEST_MARKER();
}
}
#ifdef ESP_PLATFORM /* IDF-3755 */
taskEXIT_CRITICAL( &xKernelLock );
#else
( void ) xTaskResumeAll();
#endif // ESP_PLATFORM
( void ) prvEXIT_CRITICAL_OR_RESUME_ALL( &xKernelLock );
}
else
{
@@ -4389,11 +4380,22 @@ static portTASK_FUNCTION( prvIdleTask, pvParameters )
if( xIndex < configNUM_THREAD_LOCAL_STORAGE_POINTERS )
{
taskENTER_CRITICAL( &xKernelLock );
#if ( configNUM_CORES > 1 )
/* For SMP, we need to take the kernel lock here as we
* another core could also update this task's TLSP at the
* same time. */
taskENTER_CRITICAL( &xKernelLock );
#endif /* ( configNUM_CORES > 1 ) */
pxTCB = prvGetTCBFromHandle( xTaskToSet );
pxTCB->pvThreadLocalStoragePointers[ xIndex ] = pvValue;
pxTCB->pvThreadLocalStoragePointersDelCallback[ xIndex ] = xDelCallback;
taskEXIT_CRITICAL( &xKernelLock );
#if ( configNUM_CORES > 1 )
/* Release the previously taken kernel lock. */
taskEXIT_CRITICAL( &xKernelLock );
#endif /* configNUM_CORES > 1 */
}
}
@@ -4414,10 +4416,22 @@ static portTASK_FUNCTION( prvIdleTask, pvParameters )
if( xIndex < configNUM_THREAD_LOCAL_STORAGE_POINTERS )
{
taskENTER_CRITICAL( &xKernelLock );
#if ( configNUM_CORES > 1 )
/* For SMP, we need to take the kernel lock here as we
* another core could also update this task's TLSP at the
* same time. */
taskENTER_CRITICAL( &xKernelLock );
#endif /* ( configNUM_CORES > 1 ) */
pxTCB = prvGetTCBFromHandle( xTaskToSet );
configASSERT( pxTCB != NULL );
pxTCB->pvThreadLocalStoragePointers[ xIndex ] = pvValue;
taskEXIT_CRITICAL( &xKernelLock );
#if ( configNUM_CORES > 1 )
/* Release the previously taken kernel lock. */
taskEXIT_CRITICAL( &xKernelLock );
#endif /* configNUM_CORES > 1 */
}
}
#endif /* configTHREAD_LOCAL_STORAGE_DELETE_CALLBACKS == 1 */
@@ -4634,22 +4648,14 @@ static void prvCheckTasksWaitingTermination( void )
* it should be reported as being in the Blocked state. */
if( eState == eSuspended )
{
#ifdef ESP_PLATFORM /* IDF-3755 */
taskENTER_CRITICAL( &xKernelLock );
#else
vTaskSuspendAll();
#endif // ESP_PLATFORM
prvENTER_CRITICAL_OR_SUSPEND_ALL( &xKernelLock );
{
if( listLIST_ITEM_CONTAINER( &( pxTCB->xEventListItem ) ) != NULL )
{
pxTaskStatus->eCurrentState = eBlocked;
}
}
#ifdef ESP_PLATFORM /* IDF-3755 */
taskEXIT_CRITICAL( &xKernelLock );
#else
( void ) xTaskResumeAll();
#endif // ESP_PLATFORM
( void ) prvEXIT_CRITICAL_OR_RESUME_ALL( &xKernelLock );
}
}
#endif /* INCLUDE_vTaskSuspend */
@@ -5006,7 +5012,12 @@ static void prvResetNextTaskUnblockTime( void )
TCB_t * const pxMutexHolderTCB = pxMutexHolder;
BaseType_t xReturn = pdFALSE;
taskENTER_CRITICAL( &xKernelLock );
#if ( configNUM_CORES > 1 )
/* For SMP, we need to take the kernel lock here as we are about to
* access kernel data structures. */
taskENTER_CRITICAL( &xKernelLock );
#endif /* ( configNUM_CORES > 1 ) */
/* If the mutex was given back by an interrupt while the queue was
* locked then the mutex holder might now be NULL. _RB_ Is this still
@@ -5085,7 +5096,10 @@ static void prvResetNextTaskUnblockTime( void )
mtCOVERAGE_TEST_MARKER();
}
taskEXIT_CRITICAL( &xKernelLock );
#if ( configNUM_CORES > 1 )
/* Release the previously taken kernel lock. */
taskEXIT_CRITICAL_ISR( &xKernelLock );
#endif /* ( configNUM_CORES > 1 ) */
return xReturn;
}
@@ -5100,7 +5114,12 @@ static void prvResetNextTaskUnblockTime( void )
TCB_t * const pxTCB = pxMutexHolder;
BaseType_t xReturn = pdFALSE;
taskENTER_CRITICAL( &xKernelLock );
#if ( configNUM_CORES > 1 )
/* For SMP, we need to take the kernel lock here as we are about to
* access kernel data structures. */
taskENTER_CRITICAL( &xKernelLock );
#endif /* ( configNUM_CORES > 1 ) */
if( pxMutexHolder != NULL )
{
@@ -5169,7 +5188,10 @@ static void prvResetNextTaskUnblockTime( void )
mtCOVERAGE_TEST_MARKER();
}
taskEXIT_CRITICAL( &xKernelLock );
#if ( configNUM_CORES > 1 )
/* Release the previously taken kernel lock. */
taskEXIT_CRITICAL_ISR( &xKernelLock );
#endif /* ( configNUM_CORES > 1 ) */
return xReturn;
}
@@ -5186,7 +5208,12 @@ static void prvResetNextTaskUnblockTime( void )
UBaseType_t uxPriorityUsedOnEntry, uxPriorityToUse;
const UBaseType_t uxOnlyOneMutexHeld = ( UBaseType_t ) 1;
taskENTER_CRITICAL( &xKernelLock );
#if ( configNUM_CORES > 1 )
/* For SMP, we need to take the kernel lock here as we are about to
* access kernel data structures. */
taskENTER_CRITICAL( &xKernelLock );
#endif /* ( configNUM_CORES > 1 ) */
if( pxMutexHolder != NULL )
{
@@ -5281,7 +5308,10 @@ static void prvResetNextTaskUnblockTime( void )
mtCOVERAGE_TEST_MARKER();
}
taskEXIT_CRITICAL( &xKernelLock );
#if ( configNUM_CORES > 1 )
/* Release the previously taken kernel lock. */
taskEXIT_CRITICAL( &xKernelLock );
#endif /* ( configNUM_CORES > 1 ) */
}
#endif /* configUSE_MUTEXES */
@@ -5615,18 +5645,27 @@ static void prvResetNextTaskUnblockTime( void )
TickType_t uxTaskResetEventItemValue( void )
{
TickType_t uxReturn;
TCB_t * pxCurTCB;
BaseType_t xCoreID;
taskENTER_CRITICAL( &xKernelLock );
pxCurTCB = pxCurrentTCB[ xPortGetCoreID() ];
#if ( configNUM_CORES > 1 )
uxReturn = listGET_LIST_ITEM_VALUE( &( pxCurTCB->xEventListItem ) );
/* For SMP, we need to take the kernel lock here to ensure nothing else
* modifies the task's event item value simultaneously. */
taskENTER_CRITICAL( &xKernelLock );
#endif /* ( configNUM_CORES > 1 ) */
xCoreID = xPortGetCoreID();
uxReturn = listGET_LIST_ITEM_VALUE( &( pxCurrentTCB[ xCoreID ]->xEventListItem ) );
/* Reset the event list item to its normal value - so it can be used with
* queues and semaphores. */
listSET_LIST_ITEM_VALUE( &( pxCurTCB->xEventListItem ), ( ( TickType_t ) configMAX_PRIORITIES - ( TickType_t ) pxCurTCB->uxPriority ) ); /*lint !e961 MISRA exception as the casts are only redundant for some ports. */
listSET_LIST_ITEM_VALUE( &( pxCurrentTCB[ xCoreID ]->xEventListItem ), ( ( TickType_t ) configMAX_PRIORITIES - ( TickType_t ) pxCurrentTCB[ xCoreID ]->uxPriority ) ); /*lint !e961 MISRA exception as the casts are only redundant for some ports. */
taskEXIT_CRITICAL( &xKernelLock );
#if ( configNUM_CORES > 1 )
/* Release the previously taken kernel lock. */
taskEXIT_CRITICAL_ISR( &xKernelLock );
#endif /* ( configNUM_CORES > 1 ) */
return uxReturn;
}
@@ -5636,21 +5675,31 @@ TickType_t uxTaskResetEventItemValue( void )
TaskHandle_t pvTaskIncrementMutexHeldCount( void )
{
TCB_t * curTCB;
TCB_t * pxCurTCB;
BaseType_t xCoreID;
#if ( configNUM_CORES > 1 )
/* For SMP, we need to take the kernel lock here as we are about to
* access kernel data structures. */
taskENTER_CRITICAL( &xKernelLock );
#endif /* ( configNUM_CORES > 1 ) */
xCoreID = xPortGetCoreID();
/* If xSemaphoreCreateMutex() is called before any tasks have been created
* then pxCurrentTCB will be NULL. */
taskENTER_CRITICAL( &xKernelLock );
if( pxCurrentTCB[ xPortGetCoreID() ] != NULL )
if( pxCurrentTCB[ xCoreID ] != NULL )
{
( pxCurrentTCB[ xPortGetCoreID() ]->uxMutexesHeld )++;
( pxCurrentTCB[ xCoreID ]->uxMutexesHeld )++;
}
curTCB = pxCurrentTCB[ xPortGetCoreID() ];
taskEXIT_CRITICAL( &xKernelLock );
pxCurTCB = pxCurrentTCB[ xCoreID ];
#if ( configNUM_CORES > 1 )
/* Release the previously taken kernel lock. */
taskEXIT_CRITICAL( &xKernelLock );
#endif /* ( configNUM_CORES > 1 ) */
return curTCB;
return pxCurTCB;
}
#endif /* configUSE_MUTEXES */
@@ -5971,6 +6020,7 @@ TickType_t uxTaskResetEventItemValue( void )
TCB_t * pxTCB;
uint8_t ucOriginalNotifyState;
BaseType_t xReturn = pdPASS;
UBaseType_t uxSavedInterruptStatus;
configASSERT( xTaskToNotify );
configASSERT( uxIndexToNotify < configTASK_NOTIFICATION_ARRAY_ENTRIES );
@@ -5995,7 +6045,7 @@ TickType_t uxTaskResetEventItemValue( void )
pxTCB = xTaskToNotify;
taskENTER_CRITICAL_ISR( &xKernelLock );
prvENTER_CRITICAL_OR_MASK_ISR( &xKernelLock, uxSavedInterruptStatus );
{
if( pulPreviousNotificationValue != NULL )
{
@@ -6089,7 +6139,7 @@ TickType_t uxTaskResetEventItemValue( void )
}
}
}
taskEXIT_CRITICAL_ISR( &xKernelLock );
prvEXIT_CRITICAL_OR_UNMASK_ISR( &xKernelLock, uxSavedInterruptStatus );
return xReturn;
}
@@ -6105,7 +6155,7 @@ TickType_t uxTaskResetEventItemValue( void )
{
TCB_t * pxTCB;
uint8_t ucOriginalNotifyState;
UBaseType_t uxSavedInterruptStatus;
configASSERT( xTaskToNotify );
configASSERT( uxIndexToNotify < configTASK_NOTIFICATION_ARRAY_ENTRIES );
@@ -6130,7 +6180,7 @@ TickType_t uxTaskResetEventItemValue( void )
pxTCB = xTaskToNotify;
taskENTER_CRITICAL_ISR( &xKernelLock );
prvENTER_CRITICAL_OR_MASK_ISR( &xKernelLock, uxSavedInterruptStatus );
{
ucOriginalNotifyState = pxTCB->ucNotifyState[ uxIndexToNotify ];
pxTCB->ucNotifyState[ uxIndexToNotify ] = taskNOTIFICATION_RECEIVED;
@@ -6180,7 +6230,7 @@ TickType_t uxTaskResetEventItemValue( void )
}
}
}
taskEXIT_CRITICAL_ISR( &xKernelLock );
prvEXIT_CRITICAL_OR_UNMASK_ISR( &xKernelLock, uxSavedInterruptStatus );
}
#endif /* configUSE_TASK_NOTIFICATIONS */
@@ -6252,11 +6302,23 @@ TickType_t uxTaskResetEventItemValue( void )
uint32_t ulTaskGetIdleRunTimeCounter( void )
{
taskENTER_CRITICAL( &xKernelLock );
tskTCB * pxTCB = ( tskTCB * ) xIdleTaskHandle[ xPortGetCoreID() ];
taskEXIT_CRITICAL( &xKernelLock );
uint32_t ulRunTimeCounter;
return pxTCB->ulRunTimeCounter;
#if ( configNUM_CORES > 1 )
/* For SMP, we need to take the kernel lock here as we are about to
* access kernel data structures. */
taskENTER_CRITICAL( &xKernelLock );
#endif /* ( configNUM_CORES > 1 ) */
ulRunTimeCounter = xIdleTaskHandle[ xPortGetCoreID() ]->ulRunTimeCounter;
#if ( configNUM_CORES > 1 )
/* Release the previously taken kernel lock. */
taskEXIT_CRITICAL( &xKernelLock );
#endif /* ( configNUM_CORES > 1 ) */
return ulRunTimeCounter;
}
#endif /* if ( ( configGENERATE_RUN_TIME_STATS == 1 ) && ( INCLUDE_xTaskGetIdleTaskHandle == 1 ) ) */