Feature/smp granular locks v4 #1154

sudeep-mohanty · 2024-10-09T13:20:58Z

This PR adds support for granular locking to the FreeRTOS kernel.

Description

Granular locking introduces the concept of having localized locks per kernel data group for SMP configuration. This method is an optional replacement of the existing kernel locks and is controlled by a new port layer configuration, viz., portUSING_GRANULAR_LOCKS. More details about the approach can be found here.

Test Steps

The implementation has been tested on Espressif SoC targets, viz., the ESP32 using the ESP-IDF framework.

1. Testing on an esp32 target

To test the implementation of the granular locking scheme on esp32, setup the ESP-IDF environment on your local machine. The steps to follow are listed in the Getting Started Guide.
Instead of the main ESP-IDF repository, the granular locks implementation resides in this froked repository - https://github.com/sudeep-mohanty/esp-idf.
Once you have cloned the forked repository and setup the ESP-IDF environment, change your branch to feat/granular_locks_tests.
To run the FreeRTOS unit tests, change the directory to components/freertos/test_apps/freertos where all the test cases are located.
Performance tests are located in the performance subfolder at the same location.
Setup the target device with the command idf.py seet-target esp32.
Select the Amazon FreeRTOS SMP kernel using the menuconfig options. To do this you must enter the command -idf.py menuconfig -> Component config -> FreeRTOS -> Kernel -> Run the Amazon SMP FreeRTOS kernel instead (FEATURE UNDER DEVELOPMENT). Save the configuration and exit the menuconfig.
Now, build and flash the unit test application with the command idf.py build flash monitor.
Once the app is up, you can enter the number of the test case and run the unity test case.

TODO

Test setup for other targets (Raspberry Pi Pico)
Generic target tests to be uploaded to the FreeRTOS repository.

Checklist:

I have tested my changes. No regression in existing tests.
I have modified and/or added unit-tests to cover the code changes in this Pull Request.

Related Issue

Closes Granular Lock for SMP #905

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.

sudeep-mohanty · 2024-10-09T13:23:42Z

@chinglee-iot @aggarg This PR introduces the granular locks changes to the FreeRTOS kernel. Please have a look and we could have discussions/changes in this PR context. Thank you.

cc: @ESP-Marius @Dazza0

rawalexe · 2024-10-09T16:55:58Z

Thank you for your contribution, I'll forward this request to the team. There are few error in the PR can you please try fixing them

rawalexe · 2024-10-14T22:28:24Z

Hello @sudeep-mohanty, I am just following up if you had time to fix the build issues

sudeep-mohanty · 2024-10-15T08:26:38Z

@rawalexe Yes! I shall work on the failures and would also do some refactoring for an easier review process. For now, I've put this PR in draft.

ActoryOu · 2024-10-31T05:02:09Z

Hi @sudeep-mohanty,
Could you help check CI failing?

sudeep-mohanty · 2024-10-31T09:27:52Z

Hi @sudeep-mohanty, Could you help check CI failing?

Hi @ActoryOu, I've made some updates which should fix the CI failures however I could not understand why the link-verifier action fails. Seems more like a script failure to me. So this action would still fail. If you have more information on what is causing it, could you let me know and I shall fix it. Thanks.

sudeep-mohanty · 2024-11-01T10:57:39Z

Created a PR to fix the CI failure - FreeRTOS/FreeRTOS#1292

sonarqubecloud · 2024-11-04T15:38:43Z

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
6.3% Duplication on New Code

See analysis details on SonarCloud

timers.c

chinglee-iot

Update the partial review suggestion of tasks.c file and also update previous review suggestion:

Adding core ID to portENTER/EXIT_CRITICAL_DATA_GROUP parameters in these macros doesn't make sense, since the core ID is only valid when the task's executing core remains unchanged.

tasks.c

chinglee-iot · 2025-03-03T07:15:27Z

tasks.c

+                #if ( configUSE_TASK_PREEMPTION_DISABLE == 1 )
+                    || ( ( taskTASK_IS_RUNNING( pxCurrentTCBs[ xCoreID ] ) ) && ( pxCurrentTCBs[ xCoreID ]->xPreemptionDisable > 0U ) )
+                #endif


Please help me to clarify that the following sequence.

Task A is running with data group lock

Task B suspend Task A. prvYieldCore( xTaskACore ) is called

Task A is requested to yield.

Task A's xTaskRunState is set to taskTASK_SCHEDULED_TO_YIELD

Task A runs xTaskSwitchContext(). In this function, the running

state is taskTASK_SCHEDULED_TO_YIELD, which is result in taskTASK_IS_RUNNING is false.

Hi @chinglee-iot,
After reviewing the design, I believe there’s a potential issue, as you’ve rightly pointed out: when a task disables preemption, it may still be susceptible to state changes initiated from the other core. The current design does not appear to account for this scenario.

To address this, we should consider reverting to the earlier approach where the scheduler is suspended entirely, rather than just disabling task preemption. This would offer stronger guarantees against concurrent state modifications across cores.

That said, I’ll also explore whether the current design can be adapted to address this limitation. If you have any other ideas or suggestions, please feel free to share them—happy to discuss further.

I would like to discuss the following approach

check the preemption disable status in the prvYieldCore(). If the core is running a task with preemption disabled, the yield request will be delayed until the preemption enabled.

Add assertion in vTaskSwitchContext(). We should not select task for a core which is running a task with preemption disabled.

Following changes should fix the mentioned problems -

Added task preemption disable/enable during data group critical sections. This helps to avoid problems such as a task "escaping" deletion or being deleted while it is running on another core.

Added a deferred state change processing when a task's preemption is disabled. This helps to avoid the state nullification problem. Currently, it seems that we only need to defer task deletion and task suspension, with deletion give

prvYieldCore() now checks for current task's preemption disabled status.

vTaskSwitchContext() now expects that the current's task must not have preemption disabled when it is invoked by adding an assertion.

tasks.c

chinglee-iot · 2025-03-17T05:43:21Z

tasks.c

+                #if ( configUSE_TASK_PREEMPTION_DISABLE == 1 )
+                    || ( ( taskTASK_IS_RUNNING( pxCurrentTCBs[ xCoreID ] ) ) && ( pxCurrentTCBs[ xCoreID ]->xPreemptionDisable > 0U ) )
+                #endif


I would like to discuss the following approach

check the preemption disable status in the prvYieldCore(). If the core is running a task with preemption disabled, the yield request will be delayed until the preemption enabled.

Add assertion in vTaskSwitchContext(). We should not select task for a core which is running a task with preemption disabled.

chinglee-iot · 2025-03-17T05:52:02Z

tasks.c

+        /* Lock the kernel data group as we are about to access its members */
+        UBaseType_t uxSavedInterruptStatus;
+
+        if( portCHECK_IF_IN_ISR() == pdTRUE )


Let's consider to split this function into two functions.

xTaskRemoveFromEventListSafe()

xTaskRemoveFromEventListFromISRSafe()

Both of the function above calls xTaskRemoveFromEventList. Then we don't have to depend on portCHECK_IF_IN_ISR() function.

@chinglee-iot I revisited this change and I need some advice about how to handle the scenario where xTaskRemoveFromEventList() is called from prvNotifyQueueSetContainer(). The later can be called from both non-ISR functions and from-ISR functions like xQueueGenericSend() and xQueueGenericSendFromISR(). If we add Safe versions of xTaskRemoveFromEventList() we will need to move the portCHECK_IN_ISR() check to queue.c.

tasks.c

chinglee-iot

We still have nested data group critical problem needs to be updated in the implementation.

queue.c

event_groups.c

stream_buffer.c

tasks.c

include/FreeRTOS.h

tasks.c

chinglee-iot

Other review suggestions regarding preemption disabled and entering kernel critical section with nested critical section:

Consider to set xYieldPendings flag if the core is running a task with preemption disabled in prvYieldCore.
In vTaskPreemptionEnable, the task should yield when xYieldPendings is set to 1.
In vTaskExitCritical, xYieldCurrentTask should consider preemption disable. It should be set when preemption is not disabled.

queue.c

sonarqubecloud · 2025-04-02T13:46:10Z

Quality Gate passed

Issues
25 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.3% Duplication on New Code

See analysis details on SonarQube Cloud

timers.c

…herit() xTaskPriorityInherit() is called inside a critical section from queue.c. This commit moves the critical section into xTaskPriorityInherit(). Co-authored-by: Sudeep Mohanty <[email protected]>

Changed xPreemptionDisable to be a count rather than a pdTRUE/pdFALSE. This allows nested calls to vTaskPreemptionEnable(), where a yield only occurs when xPreemptionDisable is 0. Co-authored-by: Sudeep Mohanty <[email protected]>

Adds the required checks for granular locking port macros. Port Config: - portUSING_GRANULAR_LOCKS to enable granular locks - portCRITICAL_NESTING_IN_TCB should be disabled Granular Locking Port Macros: - Spinlocks - portSPINLOCK_TYPE - portINIT_SPINLOCK( pxSpinlock ) - portINIT_SPINLOCK_STATIC - Locking - portGET_SPINLOCK() - portRELEASE_SPINLOCK() Co-authored-by: Sudeep Mohanty <[email protected]>

- Updated prvCheckForRunStateChange() for granular locks - Updated vTaskSuspendAll() and xTaskResumeAll() - Now holds the xTaskSpinlock during kernel suspension - Increments/decrements xPreemptionDisable. Only yields when 0, thus allowing for nested suspensions across different data groups Co-authored-by: Sudeep Mohanty <[email protected]>

Updated critical section macros with granular locks. Some tasks.c API relied on their callers to enter critical sections. This assumption no longer works under granular locking. Critical sections added to the following functions: - `vTaskInternalSetTimeOutState()` - `xTaskIncrementTick()` - `vTaskSwitchContext()` - `xTaskRemoveFromEventList()` - `vTaskInternalSetTimeOutState()` - `eTaskConfirmSleepModeStatus()` - `xTaskPriorityDisinherit()` - `pvTaskIncrementMutexHeldCount()` Added missing suspensions to the following functions: - `vTaskPlaceOnEventList()` - `vTaskPlaceOnUnorderedEventList()` - `vTaskPlaceOnEventListRestricted()` Fixed the locking in vTaskSwitchContext() vTaskSwitchContext() must aquire both kernel locks, viz., task lock and ISR lock. This is because, vTaskSwitchContext() can be called from either task context or ISR context. Also, vTaskSwitchContext() must not alter the interrupt state prematurely. Co-authored-by: Sudeep Mohanty <[email protected]>

Updated queue.c to use granular locking - Added xTaskSpinlock and xISRSpinlock - Replaced critical section macros with data group critical section macros such as taskENTER/EXIT_CRITICAL/_FROM_ISR() with queueENTER/EXIT_CRITICAL_FROM_ISR(). - Added vQueueEnterCritical/FromISR() and vQueueExitCritical/FromISR() which map to the data group critical section macros. - Added prvLockQueueForTasks() and prvUnlockQueueForTasks() as the granular locking equivalents to prvLockQueue() and prvUnlockQueue() respectively Co-authored-by: Sudeep Mohanty <[email protected]>

Updated event_groups.c to use granular locking - Added xTaskSpinlock and xISRSpinlock - Replaced critical section macros with data group critical section macros such as taskENTER/EXIT_CRITICAL/_FROM_ISR() with event_groupsENTER/EXIT_CRITICAL/_FROM_ISR(). - Added vEventGroupsEnterCritical/FromISR() and vEventGroupsExitCriti/FromISR() functions that map to the data group critical section macros. - Added prvLockEventGroupForTasks() and prvUnlockEventGroupForTasks() to suspend the event group when executing non-deterministic code. - xEventGroupSetBits() and vEventGroupDelete() accesses the kernel data group directly. Thus, added vTaskSuspendAll()/xTaskResumeAll() to these fucntions. Co-authored-by: Sudeep Mohanty <[email protected]>

Updated stream_buffer.c to use granular locking - Added xTaskSpinlock and xISRSpinlock - Replaced critical section macros with data group critical section macros such as taskENTER/EXIT_CRITICAL/_FROM_ISR() with sbENTER/EXIT_CRITICAL_FROM_ISR(). - Added vStreambuffersEnterCritical/FromISR() and vStreambuffersExitCritical/FromISR() to map to the data group critical section macros. - Added prvLockStreamBufferForTasks() and prvUnlockStreamBufferForTasks() to suspend the stream buffer when executing non-deterministic code. Co-authored-by: Sudeep Mohanty <[email protected]>

Updated timers.c to use granular locking - Added xTaskSpinlock and xISRSpinlock - Replaced critical section macros with data group critical section macros such as taskENTER/EXIT_CRITICAL() with tmrENTER/EXIT_CRITICAL(). - Added vTimerEnterCritical() and vTimerExitCritical() to map to the data group critical section macros. Co-authored-by: Sudeep Mohanty <[email protected]>

Co-authored-by: Sudeep Mohanty <[email protected]>

sonarqubecloud · 2025-05-23T08:09:22Z

Quality Gate passed

Issues
25 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.4% Duplication on New Code

See analysis details on SonarQube Cloud

sudeep-mohanty marked this pull request as draft October 15, 2024 08:25

sudeep-mohanty force-pushed the feature/smp_granular_locks_v4 branch 9 times, most recently from 0159a4a to 5bf8c33 Compare October 30, 2024 10:26

sudeep-mohanty marked this pull request as ready for review October 30, 2024 10:34

sudeep-mohanty requested a review from a team as a code owner October 30, 2024 10:34

sudeep-mohanty force-pushed the feature/smp_granular_locks_v4 branch from 5bf8c33 to 9f8acc7 Compare October 31, 2024 09:25

sudeep-mohanty force-pushed the feature/smp_granular_locks_v4 branch 2 times, most recently from c79962d to f50d476 Compare October 31, 2024 12:43

sudeep-mohanty mentioned this pull request Nov 1, 2024

change(freertos/smp/granular_locks): Updated prvYieldCore() coverage tests FreeRTOS/FreeRTOS#1292

Closed

1 task

sudeep-mohanty force-pushed the feature/smp_granular_locks_v4 branch 4 times, most recently from 70eb466 to edc1c98 Compare November 4, 2024 15:38

sudeep-mohanty marked this pull request as draft November 15, 2024 14:25

chinglee-iot reviewed Dec 9, 2024

View reviewed changes

timers.c Outdated Show resolved Hide resolved

chinglee-iot reviewed Feb 4, 2025

View reviewed changes

tasks.c Outdated Show resolved Hide resolved

tasks.c Outdated Show resolved Hide resolved

tasks.c Show resolved Hide resolved

sudeep-mohanty force-pushed the feature/smp_granular_locks_v4 branch from bf1d964 to d3784c3 Compare February 4, 2025 09:50

sudeep-mohanty force-pushed the feature/smp_granular_locks_v4 branch from d3784c3 to 0847cd0 Compare February 26, 2025 11:22

chinglee-iot reviewed Mar 3, 2025

View reviewed changes

tasks.c Outdated Show resolved Hide resolved

chinglee-iot reviewed Mar 17, 2025

View reviewed changes

chinglee-iot reviewed Mar 24, 2025

View reviewed changes

sudeep-mohanty marked this pull request as draft March 24, 2025 14:32

chinglee-iot reviewed Mar 26, 2025

View reviewed changes

queue.c Show resolved Hide resolved

chinglee-iot reviewed Mar 27, 2025

View reviewed changes

queue.c Outdated Show resolved Hide resolved

queue.c Outdated Show resolved Hide resolved

sudeep-mohanty force-pushed the feature/smp_granular_locks_v4 branch 4 times, most recently from d65b8e6 to f0609db Compare April 2, 2025 13:45

chinglee-iot reviewed May 20, 2025

View reviewed changes

timers.c Outdated Show resolved Hide resolved

sudeep-mohanty force-pushed the feature/smp_granular_locks_v4 branch from f0609db to 8eb906d Compare May 22, 2025 13:27

Dazza0 and others added 3 commits May 23, 2025 08:49

refactor(freertos/smp): Move critical sections inside xTaskPriorityIn…

f836da4

…herit() xTaskPriorityInherit() is called inside a critical section from queue.c. This commit moves the critical section into xTaskPriorityInherit(). Co-authored-by: Sudeep Mohanty <[email protected]>

sudeep-mohanty force-pushed the feature/smp_granular_locks_v4 branch from 8eb906d to c32b603 Compare May 23, 2025 06:50

Dazza0 and others added 7 commits May 23, 2025 10:08

feat(freertos/smp): Add Granular Locking V4 proposal documents

da5dccf

Co-authored-by: Sudeep Mohanty <[email protected]>

sudeep-mohanty force-pushed the feature/smp_granular_locks_v4 branch from c32b603 to da5dccf Compare May 23, 2025 08:08

Feature/smp granular locks v4 #1154

Are you sure you want to change the base?

Feature/smp granular locks v4 #1154

Uh oh!

Conversation

sudeep-mohanty commented Oct 9, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Test Steps

TODO

Checklist:

Related Issue

Uh oh!

sudeep-mohanty commented Oct 9, 2024

Uh oh!

rawalexe commented Oct 9, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rawalexe commented Oct 14, 2024

Uh oh!

sudeep-mohanty commented Oct 15, 2024

Uh oh!

ActoryOu commented Oct 31, 2024

Uh oh!

sudeep-mohanty commented Oct 31, 2024

Uh oh!

sudeep-mohanty commented Nov 1, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

sonarqubecloud bot commented Nov 4, 2024

Quality Gate passed

Uh oh!

Uh oh!

chinglee-iot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

chinglee-iot Mar 3, 2025

Choose a reason for hiding this comment

Uh oh!

sudeep-mohanty Mar 10, 2025

Choose a reason for hiding this comment

Uh oh!

chinglee-iot Mar 17, 2025

Choose a reason for hiding this comment

Uh oh!

sudeep-mohanty May 23, 2025

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

chinglee-iot Mar 17, 2025

Choose a reason for hiding this comment

Uh oh!

chinglee-iot Mar 17, 2025

Choose a reason for hiding this comment

Uh oh!

sudeep-mohanty May 23, 2025

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

chinglee-iot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

chinglee-iot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

sudeep-mohanty commented Oct 9, 2024 •

edited

Loading

rawalexe commented Oct 9, 2024 •

edited

Loading

sudeep-mohanty commented Nov 1, 2024 •

edited

Loading