Darian [Tue, 5 Dec 2023 06:31:11 +0000 (14:31 +0800)]
Update xTaskGetIdleTaskHandle() For SMP (#868)
* Update xTaskGetIdleTaskHandle() in SMP
This commit updates xTaskGetIdleTaskHandle() for SMP in the following ways:
- xTaskGetIdleTaskHandle() no longer accepts xCoreID argument in SMP so that
there is not change in API between single-core and SMP
- xTaskGetIdleTaskHandle() now returns the Active idle task handle in SMP,
which matches the behavior in single-core.
- Added xTaskGetIdleTaskHandleForCore() in SMP which accepts an xCoreID
argument. This function can be used to obtain the Passive idle task handles.
* Update xTaskGetIdleTaskHandle
---------
Co-authored-by: Ching-Hsin Lee <chinglee@amazon.com> Co-authored-by: chinglee-iot <61685396+chinglee-iot@users.noreply.github.com> Co-authored-by: Rahul Kar <118818625+kar-rahul-aws@users.noreply.github.com> Co-authored-by: Gaurav-Aggarwal-AWS <33462878+aggarg@users.noreply.github.com>
Rahul Kar [Tue, 28 Nov 2023 13:46:15 +0000 (19:16 +0530)]
Coverity Report Directory Fix (#909)
* Revert "The curl command to send the report expects the tar file to be in its current directory. The step either needed to have the working-directory: set to the build directory, or the tar file needs to be created in the parent directory. (#903)"
Soren Ptak [Tue, 28 Nov 2023 05:08:57 +0000 (00:08 -0500)]
The curl command to send the report expects the tar file to be in its current directory. The step either needed to have the working-directory: set to the build directory, or the tar file needs to be created in the parent directory. (#903)
Soren Ptak [Thu, 23 Nov 2023 16:47:31 +0000 (08:47 -0800)]
Coverity Scan Workflow Fix (#891)
Currently the Coverity Scan attempts to run on every fork that pulls
the file. This leads to anybody who pulls this file getting emails that
their workflow failed to run when the cron job attempts to run. This
PR sets the scan to only run if the repo is FreeRTOS/FreeRTOS-Kernel.
Also, change the scan from a cron job to a job that runs on a commit
to mainline, or if triggered manually.
Gaurav-Aggarwal-AWS [Thu, 23 Nov 2023 05:17:47 +0000 (10:47 +0530)]
Update system call entry mechanism (#898)
Earlier the System Call entry from an unprivileged task
looked like:
1. SVC for entering system call.
2. System call implementation.
3. SVC for exiting system call.
Now, the system call entry needs to make only one SVC
call and everything else is handled internally.
This PR also makes the following changes:
1. Update the Access Control List (ACL) mechanism to
grant access to all the kernel objects before the
scheduler is started.
2. Add one struct param for system calls with 5 parameters.
This removes the need for special handling for system
calls with 5 parameters.
3. Remove raise privilege SVC when MPU wrapper v2 is used.
4. Add additional run time parameter checks to MPU wrappers
for xTaskGenericNotify and xQueueTakeMutexRecursive APIs.
chinglee-iot [Thu, 16 Nov 2023 01:04:05 +0000 (09:04 +0800)]
Add vApplicationGetPassiveIdleTaskMemory for SMP (#890)
* Update vApplicationGetIdleTaskMemory prototype for SMP. Now SMP and
single core use the same prototype for compatibility.
* Add vApplicationGetPassiveIdleTaskMemory for SMP to get passive idle
task memory.
Moral-Hao [Mon, 6 Nov 2023 09:50:32 +0000 (17:50 +0800)]
Fix vTaskSwitchContext for smp. (#879)
The function vTaskSwitchContext in smp has an parameter of core id,
which means this function is not only used for the core who call it.
Thus we should use the task running on the specific core id,
instead of use the task running on the core who call this function.
Devaraj Ranganna [Tue, 31 Oct 2023 04:36:39 +0000 (04:36 +0000)]
aarch64: Rename ARM_CA53_64_BIT/_SRE to Arm_AARCH64/_SRE (#822)
The Cortex-A53 ports are generic and can be used as a starting point
for other Armv8-A application processors. Therefore, rename
`ARM_CA53_64_BIT` to `Arm_AARCH64` and `ARM_CA53_64_BIT_SRE` to
`Arm_AARCH64_SRE`.
With this renaming, existing projects that use old port, should
migrate to renamed port as follows:
Joe Benczarski [Fri, 27 Oct 2023 18:57:52 +0000 (14:57 -0400)]
Support configurable RISC-V chip extension (#773)
* Support configurable RISC-V chip extension
Added the FREERTOS_RISCV_EXTENSION option to allow the user
to select which chip extension they want included. Removed the
port for pulpino to instead use the new option.
* Add port GCC_RISC_V_GENERIC and IAR_RISC_V_GENERIC
* Add two rics-v generic ports to support FREERTOS_RISCV_EXTENSION
config
---------
Co-authored-by: Joe Benczarski <jbenczarski@trijicon.com> Co-authored-by: chinglee-iot <61685396+chinglee-iot@users.noreply.github.com> Co-authored-by: Ching-Hsin Lee <chinglee@amazon.com> Co-authored-by: kar-rahul-aws <118818625+kar-rahul-aws@users.noreply.github.com> Co-authored-by: Soren Ptak <ptaksoren@gmail.com>
Boris van der Meer [Mon, 23 Oct 2023 06:57:54 +0000 (08:57 +0200)]
Remove default behaviour of FREERTOS_HEAP. (#807)
To build a complete static application (configSUPPORT_DYNAMIC_ALLOCATION
set to 0) an ugly workaround is necessary, because when FREERTOS_HEAP is
not set, heap 4 is automatically selected in the current CMake.
Soren Ptak [Thu, 19 Oct 2023 17:08:24 +0000 (10:08 -0700)]
Kernel Checker CI Workflow File Updates (#804)
* Perform sparse checkout of just the .github folder for the header check instead of all the files. Update python checkout version being used. Update the version of the get changed files action being used.
This PR fixes the bug described in the following issue:
https://github.com/FreeRTOS/FreeRTOS-Kernel/issues/612.
This was originally contributed in the following PR:
https://github.com/FreeRTOS/FreeRTOS-Kernel/pull/625.
The implementation suspends the scheduler before exiting the
critical section (i.e. before enabling interrupts). If we do not do
so, a notification sent from an ISR, which happens after exiting
the critical section and before suspending the scheduler, will
get lost. The sequence of events will be:
1. Exit critical section.
2. Interrupt - ISR calls xTaskNotifyFromISR which adds the task to
the Ready list.
3. Suspend scheduler.
4. prvAddCurrentTaskToDelayedList moves the task to the delayed
or suspended list.
5. Resume scheduler does not touch the task (because it is not on
the pendingReady list), effectively losing the notification from
the ISR.
The same does not happen when we suspend the scheduler before
exiting the critical section. The sequence of events in this case will
be:
1. Suspend scheduler.
2. Exit critical section.
3. Interrupt - ISR calls xTaskNotifyFromISR which adds the task to
the pendingReady list as the scheduler is suspended.
4. prvAddCurrentTaskToDelayedList adds the task to delayed or
suspended list. Note that this operation does not nullify the add
to pendingReady list done in the above step because a different
list item, namely xEventListItem, is used for adding the task to the
pendingReady list. In other words, the task still remains on the
pendingReady list.
5. Resume scheduler moves the task from pendingReady list to the Ready list.
Sudeep Mohanty [Mon, 16 Oct 2023 07:08:55 +0000 (09:08 +0200)]
Add option to set the core affinity for the Timer Svc Task on SMP systems (#805)
This PR introduces the configTIMER_SERVICE_TASK_CORE_AFFINITY option
which allows the system to configure the core affinity of the Timer
Service Task on an SMP system. The default affinity of the Timer Service
Task is set to tskNO_AFFINITY which is the current behavior on SMP
systems.
chinglee-iot [Thu, 12 Oct 2023 02:41:16 +0000 (10:41 +0800)]
Update SMP get idle task memory for static allocation (#784)
* Add static idle task memory support for SMP
* Rename vApplicationMinimalIdleTask to vApplicationPassiveIdleTask
* Rename the macro configUSE_MINIMAL_IDLE_HOOK to configUSE_PASSIVE_IDLE_HOOK
* Update xTaskGetIdleTaskHandle for SMP
* Add more check in xTaskGetIdleTaskHandle
* Support configKERNEL_PROVIDED_STATIC_MEMORY for SMP
Michael Fairman [Tue, 3 Oct 2023 06:00:33 +0000 (23:00 -0700)]
bugfix: correct computation of stack size on Mac Posix port (#816)
Aligns the stack end to a page boundary before computing its
size, since the size depends on both the start and end.
The original change which introduced stack alignment (#674)
only worked for cases where the round + trunc operation would
wind up within the same area, but would lead to segfaults in
other cases.
Also adds a typecast to the `mach_vm_round_page()` call, as
it is actually a macro which casts to `mach_vm_offset_t` and
the result here is used as a `StackType_t` pointer.
Tested on ARM64 and Intel MacOS, as well as ARM64 and Intel
Linux. The test code included a single-task case, as well
as a case with two tasks passing queue messages.
According to Armv8-M technical reference manual, if the main extension
is not implemented then PSPLIM_NS is RES0. Update the cortex-M23
port to not use the reserved PSPLIM_NS.
These 2 APIs behave exactly as vTaskList and vTaskGetRunTimeStats
except the fact that they take the length of the pcWriteBuffer as the
second argument to ensure that we do not write past the buffer.
vTaskList and vTaskGetRunTimeStats assume that the length of the
buffer is configSTATS_BUFFER_MAX_LENGTH which defaults to 0xFFFF.
This is done to ensure that the existing applications do not break.
New applications should use the new APIs to avoid memory corruption.
Fix portSWITCH_TO_USER_MODE() on Armv7-M MPU ports (#803)
A task's privilege level is stored in ulTaskFlag member in the TCB. Current
implementation of portSWITCH_TO_USER_MODE() does not update this
flag but just lowers the processor's privilege level. This results in many
APIs incorrectly determining task's privilege level and access permissions -
This PR fixes the portSWITCH_TO_USER_MODE() implementation to correctly
update the ulTaskFlag member in the TCB before lowering the processor's
privilege level.
Boris van der Meer [Wed, 20 Sep 2023 10:49:42 +0000 (12:49 +0200)]
Add trace hook macro for most ports (#794)
Add trace hook macro for most ports
In pull request #659 we introduced better support for tracing
tools like systemview. This patchset adds support for more
ports as requested in the original pull request.
Sebastian Brosch [Wed, 20 Sep 2023 10:17:42 +0000 (12:17 +0200)]
Add Trace Hook Macros to all API calls (#786)
This pull-request adds out-of-the-box support for different tracing
tools. New trace hook macros have been added for all public API
functions, one for each function entry and one for each exit. There are
no functional changes, as the macros are by default empty.
For more information see following forum post:
https://forums.freertos.org/t/add-tracing-functionality-for-all-api-calls/18007.
Mehdi [Wed, 20 Sep 2023 05:39:37 +0000 (09:09 +0330)]
Add default implementations of vApplicationGetIdleTaskMemory and vApplicationGetTimerTaskMemory (#790)
This PR introduces configKERNEL_PROVIDED_STATIC_MEMORY option
which the application can set to 1 to use the default implementations
of vApplicationGetIdleTaskMemory and vApplicationGetTimerTaskMemory
functions.
If the application enables static allocation (i.e. sets
configUSE_STATIC_ALLOCATION to 1) and does not provide the above 2
functions, it will result in linker error. The application has two options:
1. Set configKERNEL_PROVIDED_STATIC_MEMORY to 1 to use the default
implementations of these functions.
2. Provide implementations of these 2 functions.
Note that default definitions are only available for non-MPU ports. The
reason is that the stack alignment requirements vary for different
architectures.
This PR adds Access Control to kernel objects on a per task basis to MPU
ports. The following needs to be defined in the `FreeRTOSConfig.h` to
enable this feature:
An unprivileged task by default has access to itself only and no other
kernel object. The application writer needs to explicitly grant an
unprivileged task access to all the kernel objects it needs. The best
place to do that is before starting the scheduler when all the kernel
objects are created.
For example, let's say an unprivileged tasks needs access to a queue and
an event group, the application writer needs to do the following:
The application writer MUST revoke all the accesses before deleting a
task. Failing to do so will result in undefined behavior. In the above
example, the application writer needs to make the following 2 calls
before deleting the task:
* Add taskYIELD_TASK_CORE_IF_USING_PREEMPTION and taskYIELD_ANY_CORE_IF_USING_PREEMPTION to align task yield behavior for single core and SMP.
---------
* Add macro taskTASK_IS_RUNNING_OR_SCHEDULED_TO_YIELD macro to align single core and SMP
* Update for explicit precedence in vTaskDelete
* Update comment when deleting a running task
* Use new version of CI-CD Actions
* Use cSpell spell check, and use ubuntu-20.04 for formatting check
* Format and spell check all files in the portable directory
* Remove the https:// from #errors and #warnings as uncrustify attempts to change it to /*
* Use checkout@v3 instead of checkout@v2 on all jobs
---------
oliverlavery [Wed, 23 Aug 2023 04:57:54 +0000 (05:57 +0100)]
Heap protect (#747)
Setting configENABLE_HEAP_PROTECTOR to 1 obfuscates heap
block pointers by XORing them with an application supplied
canary value. This obfuscation helps to catch heap corruption
should a heap buffer overflow occur.
This PR also adds heap bounds checking to heap_4 and heap_5.
This PR also adds some additional integer underflow checks.
begriffs.com / projects / freertos / log