Fix array-bounds compiler warning on gcc11+ in list.h #580

archigup · 2022-10-21T00:28:38Z

Fix array-bounds compiler warning on gcc11+ in list.h

Description

listGET_OWNER_OF_NEXT_ENTRY computes ( pxConstList )->pxIndex->pxNext after verifying that ( pxConstList )->pxIndex points to xListEnd, which due to being a MiniListItem_t, can be shorter than a ListItem_t. Thus, ( pxConstList )->pxIndex is a ListItem_t * that extends past the end of the List_t whose xListEnd it points to.

On GCC 11+, this triggers an array-bounds warning. This is fixed by accessing pxNext through a MiniListItem_t instead.

listGET_OWNER_OF_NEXT_ENTRY uses uninitialized memory if the passed list is empty (only has the list end marker). While all uses of this macro are nested within a list empty check or are after an assert, this trips up GCC post GCC 11 on O3. Worse yet, with minilist enabled, this macro will dereference a struct offset past the end of the struct when passed an empty list.

Adding this check makes the behavior of the macro on empty list explicit when the only element is xListEnd, and prevents accessing xListEnd->pvOwner (Which is currently NULL where the List_t is zero-initialized; I have not checked if there are cases where it is not).

GCC gives a warning as it catches that the macro does not work with an empty list, but it does not figure that that code path is unreachable. Adding this check or adding more list non-empty checks/asserts directly outside of the macro fixes the issue. With asserts, the issue is only solved when asserts are enabled, so just adding the check is the better solution.

Test Steps

Compile tasks.h with GCC 12 (or 11) and with 03 as follows:

gcc -c tasks.c -I include -I ${KERNEL_CONFIG_PATH}/ -I portable/ThirdParty/GCC/Posix/ -Wall -Wextra -Werror -O3

Related Issue

FreeRTOS/FreeRTOS#858

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

codecov · 2022-10-21T00:35:57Z

Codecov Report

Base: 94.30% // Head: 94.30% // No change to project coverage 👍

Coverage data is based on head (eb4819d) compared to base (91927ab).
Patch has no changes to coverable lines.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #580   +/-   ##
=======================================
  Coverage   94.30%   94.30%           
=======================================
  Files           6        6           
  Lines        2370     2370           
  Branches      579      579           
=======================================
  Hits         2235     2235           
  Misses         85       85           
  Partials       50       50

Flag	Coverage Δ
unittests	`94.30% <ø> (ø)`

Flags with carried forward coverage won't be shown. Click here to find out more.

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

RichardBarry · 2022-10-21T01:01:32Z

The update adds an additional test into listGET_OWNER_OF_NEXT_ENTRY(), which is called from within taskSELECT_HIGHEST_PRIORITY_TASK(), so gets called at high frequency, including potentially in the tick interrupt. Anything that extends the runtime of high frequency functions needs additional scrutiny, hence I would like to understand this better before merging.

I see the code will assert if the subject list is empty – and I don’t think I’ve seen that assert get hit in cases other than where portUSE_PORT_OPTIMISED_TASK_SELECTION is 0 and the lists are already corrupt. So, if my understanding is correct, the update duplicates code. On first look, it seems the update is only required to work around a compiler bug (the hardest bugs to find, so somebody did a good job here!), in which case we need to report the bug if it is not already being tracked? Or is the compiler’s behaviour legitimate? The update impacts all compilers, so if it is a compiler bug work around, will we take it out when the compiler is fixed?

cobusve · 2022-10-21T04:24:10Z

Have we confirmed if this is actually a bug in the compiler? Because if it is not that changes things a lot.

RichardBarry · 2022-10-21T17:45:45Z

As archigup says, without knowing the execution path, the compiler warning is probably legitimate at as a static check (at any optimisation level). I would not expect there to be an actual error condition at run-time though.

archigup · 2022-10-21T17:57:43Z

I am currently minimizing the testcase to evaluate whether it is a compiler bug.

archigup · 2022-11-30T23:11:25Z

I've minimized the test case to this:

typedef struct xLIST_ITEM
{
    struct xLIST_ITEM * pxNext;
    struct xLIST * pvContainer;
} ListItem_t;

typedef struct xMINI_LIST_ITEM
{
    struct xLIST_ITEM * pxNext;
} MiniListItem_t;

typedef struct xLIST
{
    int uxNumberOfItems;
    ListItem_t * pxIndex;
    MiniListItem_t xListEnd;
} List_t;

static List_t xSuspendedTaskList;

static void prvSearchForNameWithinSingleList( List_t * pxList )
{
    if( ( ( pxList )->uxNumberOfItems ) > 0 )
    {
        pxList->pxIndex = pxList->pxIndex->pxNext;
        if( ( void * ) pxList->pxIndex == ( void * ) &( pxList->xListEnd ) )
        {
            pxList->pxIndex = pxList->pxIndex->pxNext;
        }
    }
}

void xTaskGetHandle( void )
{
    prvSearchForNameWithinSingleList( &xSuspendedTaskList );
}

This makes it clear that the fix in this PR is incidental; the warning can be reproduced without that last assignment in listGET_OWNER_OF_NEXT_ENTRY.

I am also able to trigger it on 02 in addition to 03 and 0s with this. The warning also goes away when configUSE_MINI_LIST_ITEM is 0.

archigup · 2022-11-30T23:27:39Z

https://gcc.gnu.org/onlinedocs/gcc/Warning-Options.html#index-Warray-bounds

Looking at above gcc docs explains why it only happens at those optimization levels; the warning depends on -ftree-vrp which is not on all optimization levels. I am able to reproduce the warning at -O1 when -ftree-vrp is enabled, and not able to reproduce with -O3 -fno-tree-vrp.

Still odd since it seems it should fall into -Warray-bounds=2 but its happening with -Warray-bounds=1.

archigup · 2022-12-01T00:02:28Z

Narrowing it down further, it seems the issue really is in:

if( ( void * ) pxList->pxIndex == ( void * ) &( pxList->xListEnd ) )
{
    pxList->pxIndex = pxList->pxIndex->pxNext;
}

In the right side of the assignment pxList->pxIndex is a ListItem_t *, but it is known that it points to pxList->xListEnd due to the surrounding check; accessing xListEnd through a ListItem_t * does extend past the memory region that the specific calls of prvSearchForNameWithinSingleList pass in as the pxList parameter. This explains why its only warning where the pxList param is actually a global var. This can be fixed by not accessing it through a type that would extend outside of the memory region. The following equivalent code works:

if( ( void * ) pxList->pxIndex == ( void * ) &( pxList->xListEnd ) )
{
    pxList->pxIndex = pxList->xListEnd.pxNext;
}

Due to the if condition, those two are the same value, but the latter code instead accesses it as a MiniListItem_t which is in bounds.

The following code also alternatively works:

if( ( void * ) pxList->pxIndex == ( void * ) &( pxList->xListEnd ) )
{
    pxList->pxIndex = ( ( MiniListItem_t * ) pxList->pxIndex )->pxNext;
}

listGET_OWNER_OF_NEXT_ENTRY computes `( pxConstList )->pxIndex->pxNext` after verifying that `( pxConstList )->pxIndex` points to `xListEnd`, which due to being a MiniListItem_t, can be shorter than a ListItem_t. Thus, `( pxConstList )->pxIndex` is a `ListItem_t *` that extends past the end of the `List_t` whose `xListEnd` it points to. This is fixed by accessing `pxNext` through a `MiniListItem_t` instead.

@ChristosZosi

* Fix array-bounds compiler warning on gcc11+ in list.h (FreeRTOS#580) listGET_OWNER_OF_NEXT_ENTRY computes `( pxConstList )->pxIndex->pxNext` after verifying that `( pxConstList )->pxIndex` points to `xListEnd`, which due to being a MiniListItem_t, can be shorter than a ListItem_t. Thus, `( pxConstList )->pxIndex` is a `ListItem_t *` that extends past the end of the `List_t` whose `xListEnd` it points to. This is fixed by accessing `pxNext` through a `MiniListItem_t` instead. * move the prototype for vApplicationIdleHook to task.h. (FreeRTOS#600) Co-authored-by: pluess <[email protected]> Co-authored-by: Gaurav-Aggarwal-AWS <[email protected]> * Update equal priority task preemption (FreeRTOS#603) * vTaskResume and vTaskPrioritySet don't preempt equal priority task * Update vTaskResumeAll not to preempt task with equal priority * Fix in xTaskResumeFromISR * Update FreeRTOS/FreeRTOS build checks (FreeRTOS#613) This is needed to be compatible with the refactoring done in this PR - FreeRTOS/FreeRTOS#889 Signed-off-by: Gaurav Aggarwal <[email protected]> Signed-off-by: Gaurav Aggarwal <[email protected]> * Add ulTaskGetRunTimeCounter and ulTaskGetRunTimePercent (FreeRTOS#611) Allow ulTaskGetIdleRunTimeCounter and ulTaskGetIdleRunTimePercent to be used whenever configGENERATE_RUN_TIME_STATS is enabled, as this is the only requirement for these functions to work. * Fix some CMake documentation typos (FreeRTOS#616) The quick start instructions for CMake mention the "master" git branch which has been replaced by "main" in the current repo. The main CMakeLists.txt documents how to integrate a custom port. Fix a typo in the suggested CMake code. * Added support of 64bit events. (FreeRTOS#597) * Added support of 64bit even Signed-off-by: Cervenka Dusan <[email protected]> * Added missing brackets Signed-off-by: Cervenka Dusan <[email protected]> * Made proper name for tick macro. Signed-off-by: Cervenka Dusan <[email protected]> * Improved macro evaluation Signed-off-by: Cervenka Dusan <[email protected]> * Fixed missed port files + documentation Signed-off-by: Cervenka Dusan <[email protected]> * Changes made on PR Signed-off-by: Cervenka Dusan <[email protected]> * Fix macro definition. Signed-off-by: Cervenka Dusan <[email protected]> * Formatted code with uncrustify Signed-off-by: Cervenka Dusan <[email protected]> --------- Signed-off-by: Cervenka Dusan <[email protected]> * Introduce portMEMORY_BARRIER for Microblaze port. (FreeRTOS#621) The introduction of `portMEMORY_BARRIER` will ensure the places in the kernel use a barrier will work. For example, `xTaskResumeAll` has a memory barrier to ensure its correctness when compiled with optimization enabled. Without the barrier `xTaskResumeAll` can fail (e.g. start reading and writing to address 0 and/or infinite looping) when `xPendingReadyList` contains more than one task to restore. In `xTaskResumeAll` the compiler chooses to cache the `pxTCB` the first time through the loop for use in every subsequent loop. This is incorrect as the removal of `pxTCB->xEventListItem` will actually change the value of `pxTCB` if it was read again at the top of the loop. The barrier forces the compiler to read `pxTCB` again at the top of the loop. The compiler is operating correctly. The removal `pxTCB->xEventListItem` executes on a `List_t *` and `ListItem_t *`. This means that the compiler can assume that any `MiniListItem_t` values are unchanged by the loop (i.e. "strict-aliasing"). This allows the compiler to cache `pxTCB` as it is obtained via a `MiniListItem_t`. This is incorrect in this case because it is possible for a `ListItem_t *` to actually alias a `MiniListItem_t`. This is technically a "violation of aliasing rules" so we use the the barrier to disable the strict-aliasing optimization in this loop. * Do not call exit() on MSVC Port when calling vPortEndScheduler (FreeRTOS#624) * make port exitable * correctly set xPortRunning to False * add suggestions from Review Co-authored-by: Gaurav-Aggarwal-AWS <[email protected]> * add suggestions from Review Co-authored-by: Gaurav-Aggarwal-AWS <[email protected]> --------- Co-authored-by: Gaurav-Aggarwal-AWS <[email protected]> * Update PR template to include checkbox for Unit Test related changes (FreeRTOS#627) * Fix build failure introduced in PR FreeRTOS#597 (FreeRTOS#629) The PR FreeRTOS#597 introduced a new config option configTICK_TYPE_WIDTH_IN_BITS which can be defined to one of the following: * TICK_TYPE_WIDTH_16_BITS - Tick type is 16 bit wide. * TICK_TYPE_WIDTH_32_BITS - Tick type is 32 bit wide. * TICK_TYPE_WIDTH_64_BITS - Tick type is 64 bit wide. Earlier we supported 16 and 32 bit width for tick type which was controlled using the config option configUSE_16_BIT_TICKS. The PR tried to maintain backward compatibility by honoring configUSE_16_BIT_TICKS. The backward compatibility did not work as expected though, as the macro configTICK_TYPE_WIDTH_IN_BITS was used before it was defined. This PR addresses it by ensuring that the macro configTICK_TYPE_WIDTH_IN_BITS is defined before it is used. Testing 1. configUSE_16_BIT_TICKS is defined to 0. Source (function xTaskIncrementTick in tasks.c): ``` const TickType_t xConstTickCount = xTickCount + ( TickType_t ) 1; ``` Assembly: ``` 109e: 4b50 ldr r3, [pc, FreeRTOS#320] ; (11e0 <xTaskIncrementTick+0x150>) 10a0: f8d3 4134 ldr.w r4, [r3, FreeRTOS#308] ; 0x134 10a4: 3401 adds r4, #1 10a6: f8c3 4134 str.w r4, [r3, FreeRTOS#308] ; 0x134 ``` It is clear from assembly that the tick type is 32 bit. 2. configUSE_16_BIT_TICKS is defined to 1. Source (function xTaskIncrementTick in tasks.c): ``` const TickType_t xConstTickCount = xTickCount + ( TickType_t ) 1; ``` Assembly: ``` 10e2: 4b53 ldr r3, [pc, FreeRTOS#332] ; (1230 <xTaskIncrementTick+0x15c>) 10e4: f8b3 4134 ldrh.w r4, [r3, FreeRTOS#308] ; 0x134 10e8: b2a4 uxth r4, r4 10ea: 3401 adds r4, #1 10ec: b2a4 uxth r4, r4 10ee: f8a3 4134 strh.w r4, [r3, FreeRTOS#308] ; 0x134 ``` It is clear from assembly that the tick type is 16 bit. 3. configTICK_TYPE_WIDTH_IN_BITS is defined to TICK_TYPE_WIDTH_16_BITS. Source (function xTaskIncrementTick in tasks.c): ``` const TickType_t xConstTickCount = xTickCount + ( TickType_t ) 1; ``` Assembly: ``` 10e2: 4b53 ldr r3, [pc, FreeRTOS#332] ; (1230 <xTaskIncrementTick+0x15c>) 10e4: f8b3 4134 ldrh.w r4, [r3, FreeRTOS#308] ; 0x134 10e8: b2a4 uxth r4, r4 10ea: 3401 adds r4, #1 10ec: b2a4 uxth r4, r4 10ee: f8a3 4134 strh.w r4, [r3, FreeRTOS#308] ; 0x134 ``` It is clear from assembly that the tick type is 16 bit. 4. configTICK_TYPE_WIDTH_IN_BITS is defined to TICK_TYPE_WIDTH_32_BITS. Source (function xTaskIncrementTick in tasks.c): ``` const TickType_t xConstTickCount = xTickCount + ( TickType_t ) 1; ``` Assembly: ``` 109e: 4b50 ldr r3, [pc, FreeRTOS#320] ; (11e0 <xTaskIncrementTick+0x150>) 10a0: f8d3 4134 ldr.w r4, [r3, FreeRTOS#308] ; 0x134 10a4: 3401 adds r4, #1 10a6: f8c3 4134 str.w r4, [r3, FreeRTOS#308] ; 0x134 ``` It is clear from assembly that the tick type is 32 bit. 5. configTICK_TYPE_WIDTH_IN_BITS is defined to TICK_TYPE_WIDTH_64_BITS. ``` #error configTICK_TYPE_WIDTH_IN_BITS set to unsupported tick type width. ``` The testing was done for GCC/ARM_CM3 port which does not support 64 bit tick type. 6. Neither configUSE_16_BIT_TICKS nor configTICK_TYPE_WIDTH_IN_BITS defined. ``` #error Missing definition: One of configUSE_16_BIT_TICKS and configTICK_TYPE_WIDTH_IN_BITS must be defined in FreeRTOSConfig.h. See the Configuration section of the FreeRTOS API documentation for details. ``` 7. Both configUSE_16_BIT_TICKS and configTICK_TYPE_WIDTH_IN_BITS defined. ``` #error Only one of configUSE_16_BIT_TICKS and configTICK_TYPE_WIDTH_IN_BITS must be defined in FreeRTOSConfig.h. See the Configuration section of the FreeRTOS API documentation for details. ``` Related issue - FreeRTOS#628 Signed-off-by: Gaurav Aggarwal <[email protected]> * Feature/fixing clang gnu compiler warnings (FreeRTOS#620) * Adding in ability to support a library for freertos_config and a custom freertos_kernel_port (FreeRTOS#558) * Using single name definition for libraries everywhere. (FreeRTOS#558) * Supporting backwards compatibility with FREERTOS_CONFIG_FILE_DIRECTORY (FreeRTOS#571) * Removing compiler warnings for GNU and Clang. (FreeRTOS#571) * Added in documentation on how to consume from a main project. Added default PORT selection for native POSIX and MINGW platforms. * Only adding freertos_config if it exists. Removing auto generation of it from a FREERTOS_CONFIG_FILE_DIRECTORY. * Fixing clang and gnu compiler warnings. * Adding in project information and how to compile for GNU/clang * Fixing compiler issue with unused variable - no need to declare variable. * Adding in compile warnings for linux builds that kernel is okay with using. * Fixing more extra-semi-stmt clang warnings. * Moving definition of hooks into header files if features are enabled. * Fixing formatting with uncrustify. * Fixing merge conflicts with main merge. * Fixing compiler errors due to merge issues and formatting. * Fixing Line feeds. * Adding 'portNORETURN' into portmacros.h. Other Updates based on PR request * Further clean-up of clang and clang-tidy issues. * Removing compiler specific pragmas from common c files. * Fixing missing lexicon entry and uncrustify formatting changes. * Resolving merge issue multiple defnitions of proto for prvIdleTask * Fixing formatting issues that are not covered by uncrustify. Use clang-tidy instead if you want this level of control. * More uncrustify formatting issues. * Fixing extra bracket in #if statement. --------- Co-authored-by: Gaurav-Aggarwal-AWS <[email protected]> * POSIX port fixes (FreeRTOS#626) * Fix types in POSIX port Use TaskFunction_t and StackType_t as other ports do. * Fix portTICK_RATE_MICROSECONDS in POSIX port --------- Co-authored-by: Jacques GUILLOU <[email protected]> Co-authored-by: Gaurav-Aggarwal-AWS <[email protected]> * Cortex-M35P: Add Cortex-M35P port (FreeRTOS#631) * Cortex-M35P: Add Cortex-M35P port The Cortex-M35P support added to kernel. The port hasn't been validated yet with TF-M. Hence TF-M support is not included in this port. Signed-off-by: Devaraj Ranganna <[email protected]> * Add portNORETURN to the newly added portmacro.h Signed-off-by: Gaurav Aggarwal <[email protected]> --------- Signed-off-by: Devaraj Ranganna <[email protected]> Signed-off-by: Gaurav Aggarwal <[email protected]> Co-authored-by: Gaurav-Aggarwal-AWS <[email protected]> Co-authored-by: Gaurav Aggarwal <[email protected]> Co-authored-by: kar-rahul-aws <[email protected]> * Introduced Github Status Badge for Unit Tests (FreeRTOS#634) * Introduced Github Status Badge for Unit Tests * Github status badge to point to latest run * Github status badge UT points to latest results * Fixed URL for Github Status badge --------- Co-authored-by: kar-rahul-aws <[email protected]> * Remove C99 requirement from CMake file (FreeRTOS#633) * Remove C99 requirement from CMake file The kernel source is C89 compliant and does not need C99. Signed-off-by: Gaurav Aggarwal <[email protected]> * Explicitly set C89 requirement for kernel Signed-off-by: Gaurav Aggarwal <[email protected]> --------- Signed-off-by: Gaurav Aggarwal <[email protected]> * Add Thread Local Storage (TLS) support using Picolibc functions (FreeRTOS#343) * Pass top of stack to configINIT_TLS_BLOCK Picolibc wants to allocate the per-task TLS block within the stack segment, so it will need to modify the top of stack value. Pass the pxTopOfStack variable to make this explicit. Signed-off-by: Keith Packard <[email protected]> * Move newlib-specific definitions to separate file This reduces the clutter in FreeRTOS.h caused by having newlib-specific macros present there. Signed-off-by: Keith Packard <[email protected]> * Make TLS code depend only on configUSE_C_RUNTIME_TLS_SUPPORT Remove reference to configUSE_NEWLIB_REENTRANT as that only works when using newlib. configUSE_C_RUNTIME_TLS_SUPPORT is always set when configUSE_NEWLIB_REENTRANT is set, so using both was redundant in that case. Signed-off-by: Keith Packard <[email protected]> * portable-ARC: Adapt ARC support to use generalized TLS support With generalized thread local storage (TLS) support present in the core, the two ARC ports need to have the changes to the TCB mirrored to them. Signed-off-by: Keith Packard <[email protected]> * Add Thread Local Storage (TLS) support using Picolibc functions This patch provides definitions of the general TLS support macros in terms of the Picolibc TLS support functions. Picolibc is normally configured to use TLS internally for all variables that are intended to be task-local, so these changes are necessary for picolibc to work correctly with FreeRTOS. The picolibc helper functions rely on elements within the linker script to arrange the TLS data in memory and define some symbols. Applications wanting to use this mechanism will need changes in their linker script when migrating to picolibc. Signed-off-by: Keith Packard <[email protected]> --------- Signed-off-by: Keith Packard <[email protected]> Co-authored-by: Keith Packard <[email protected]> Co-authored-by: Gaurav-Aggarwal-AWS <[email protected]> * Interrupt priority assert improvements for CM3/4/7 (FreeRTOS#602) * Interrupt priority assert improvements for CM3/4/7 In the ARM_CM3, ARM_CM4, and ARM_CM7 ports, change the assertion that `configMAX_SYSCALL_INTERRUPT_PRIORITY` is nonzero to account for the number of priority bits implemented by the hardware. Change these ports to also use the lowest priority for PendSV and SysTick, ignoring `configKERNEL_INTERRUPT_PRIORITY`. * Remove not needed configKERNEL_INTERRUPT_PRIORITY define Signed-off-by: Gaurav Aggarwal <[email protected]> --------- Signed-off-by: Gaurav Aggarwal <[email protected]> Co-authored-by: Gaurav-Aggarwal-AWS <[email protected]> Co-authored-by: Gaurav Aggarwal <[email protected]> * Introduced code coverage status badge (FreeRTOS#635) * Introduced code coverage status badge * Trying to fix the URL checker issue * Fix URL check Signed-off-by: Gaurav Aggarwal <[email protected]> --------- Signed-off-by: Gaurav Aggarwal <[email protected]> Co-authored-by: Gaurav Aggarwal <[email protected]> Co-authored-by: Gaurav-Aggarwal-AWS <[email protected]> * added portPOINTER_SIZE_TYPE and SIZE_MAX definition to PIC24/dsPIC port (FreeRTOS#636) * added portPOINTER_SIZE_TYPE definition to PIC24/dsPIC port * Added SIZE_MAX definition to PIC24/dsPIC33 * Fix TLS and stack alignment when using picolibc (FreeRTOS#637) Both the TLS block and stack must be correctly aligned when using picolibc. The architecture stack alignment is represented by the portBYTE_ALIGNMENT_MASK and the TLS block alignment is provided by the Picolibc _tls_align() inline function for Picolibc version 1.8 and above. For older versions of Picolibc, we'll assume that the TLS block requires the same alignment as the stack. For downward growing stacks, this requires aligning the start of the TLS block to the maximum of the stack alignment and the TLS alignment. With this, both the TLS block and stack will now be correctly aligned. For upward growing stacks, the two areas must be aligned independently; the TLS block is aligned from the start of the stack, then the tls space is allocated, and then the stack is aligned above that. It's probably useful to know here that the linker ensures that variables within the TLS block are assigned offsets that match their alignment requirements. If the TLS block itself is correctly aligned, then everything within will also be. I have only tested the downward growing stack branch of this patch. Signed-off-by: Keith Packard <[email protected]> Co-authored-by: Keith Packard <[email protected]> Co-authored-by: Gaurav-Aggarwal-AWS <[email protected]> * Enable building the GCC Cortex-R5 port without an FPU (FreeRTOS#586) * Ensure configUSE_TASK_FPU_SUPPORT option is set correctly If one does enable the FPU of the Cortex-R5 processor, then the GCC compiler will define the macro __ARM_FP. This can be used to ensure, that the configUSE_TASK_FPU_SUPPORT is set accordingly. * Enable the implementation of vPortTaskUsesFPU only if configUSE_TASK_FPU_SUPPORT is set to 1 * Remove error case in pxPortInitialiseStack The case of configUSE_TASK_FPU_SUPPORT is 0 is now handled * Enable access to FPU registers only if FPU is enabled * Make minor formating changes * Format ARM Cortex-R5 port * Address review comments from @ChristosZosi * Minor code review suggestions Signed-off-by: Gaurav Aggarwal <[email protected]> --------- Signed-off-by: Gaurav Aggarwal <[email protected]> Co-authored-by: Christos Zosimidis <[email protected]> Co-authored-by: Gaurav-Aggarwal-AWS <[email protected]> Co-authored-by: Gaurav Aggarwal <[email protected]> * Fix freertos_kernel cmake property, Posix Port (FreeRTOS#640) * Fix freertos_kernel cmake property, Posix Port * Moves the `set_property()` call below the target definition in top level CMakeLists file * Corrects billion value from `ULL` suffix (not C90 compliant) to `UL` suffix with cast to uint64_t * Add blank line to CMakeLists.txt * Add missing FreeRTOS+ defines * Run kernel demos and unit tests for PR changes (FreeRTOS#645) * Run kernel demos and unit tests for PR changes Kernel demos check builds multiple demos from FreeRTOS/FreeRTOS and unit tests check runs unit tests in FreeRTOS/FreeRTOS. Both of these checks currently use main branch of FreeRTOS-Kernel. This commits updates these checks to use the changes in the PR. Signed-off-by: Gaurav Aggarwal <[email protected]> * Do not specify PR SHA explicitly as that is default Signed-off-by: Gaurav Aggarwal <[email protected]> * Remove explicit PR SHA from kernel checks Signed-off-by: Gaurav Aggarwal <[email protected]> --------- Signed-off-by: Gaurav Aggarwal <[email protected]> * Add functions to get the buffers of statically created objects (FreeRTOS#641) Added various ...GetStaticBuffer() functions to get the buffers of statically created objects. --------- Co-authored-by: Paul Bartell <[email protected]> Co-authored-by: Nikhil Kamath <[email protected]> Co-authored-by: Gaurav Aggarwal <[email protected]> * Cortex-M Assert when NVIC implements 8 PRIO bits (FreeRTOS#639) * Cortex-M Assert when NVIC implements 8 PRIO bits * Fix CM3 ports * Fix ARM_CM3_MPU * Fix ARM CM3 * Fix ARM_CM4_MPU * Fix ARM_CM4 * Fix GCC ARM_CM7 * Fix IAR ARM ports * Uncrustify changes * Fix MikroC_ARM_CM4F port * Fix MikroC_ARM_CM4F port-(2) * Fix RVDS ARM ports * Revert changes for Tasking/ARM_CM4F port * Revert changes for Tasking/ARM_CM4F port-(2) * Update port.c Fix GCC/ARM_CM4F port * Update port.c * update GCC\ARM_CM4F port * update port.c * Assert to check configMAX_SYSCALL_INTERRUPT_PRIORITY is set to higher priority * Fix merge error: remove duplicate code * Fix typos --------- Co-authored-by: Gaurav-Aggarwal-AWS <[email protected]> Co-authored-by: Ubuntu <[email protected]> * Remove C90 requirement from CMakeLists (FreeRTOS#649) This is needed as it is breaking projects - https://forums.freertos.org/t/freertos-gcc-cmake/16984 We will re-evaluate and accordingly add this later. Signed-off-by: Gaurav Aggarwal <[email protected]> * Only add alignment padding when needed (FreeRTOS#650) Heap 4 and Heap 5 add some padding to ensure that the allocated blocks are always aligned to portBYTE_ALIGNMENT bytes. The code until now was adding padding always even if the resulting block was already aligned. This commits updates the code to only add padding if the resulting block is not aligned. Signed-off-by: Gaurav Aggarwal <[email protected]> * add a missing comma (FreeRTOS#651) * fix conversion warning (FreeRTOS#658) FreeRTOS-Kernel/portable/GCC/ARM_CM4F/port.c:399:41: error: conversion from 'uint32_t' {aka 'long unsigned int'} to 'uint8_t' {aka 'unsigned char'} may change value [-Werror=conversion] Signed-off-by: Vo Trung Chi <[email protected]> --------- Signed-off-by: Gaurav Aggarwal <[email protected]> Signed-off-by: Cervenka Dusan <[email protected]> Signed-off-by: Devaraj Ranganna <[email protected]> Signed-off-by: Keith Packard <[email protected]> Signed-off-by: Vo Trung Chi <[email protected]> Co-authored-by: Archit Gupta <[email protected]> Co-authored-by: tcpluess <[email protected]> Co-authored-by: pluess <[email protected]> Co-authored-by: Gaurav-Aggarwal-AWS <[email protected]> Co-authored-by: Chris Copeland <[email protected]> Co-authored-by: David J. Fiddes <[email protected]> Co-authored-by: Dusan Cervenka <[email protected]> Co-authored-by: bbain <[email protected]> Co-authored-by: Ju1He1 <[email protected]> Co-authored-by: Aniruddha Kanhere <[email protected]> Co-authored-by: phelter <[email protected]> Co-authored-by: jacky309 <[email protected]> Co-authored-by: Jacques GUILLOU <[email protected]> Co-authored-by: Devaraj Ranganna <[email protected]> Co-authored-by: Gaurav Aggarwal <[email protected]> Co-authored-by: kar-rahul-aws <[email protected]> Co-authored-by: Nikhil Kamath <[email protected]> Co-authored-by: Keith Packard <[email protected]> Co-authored-by: Keith Packard <[email protected]> Co-authored-by: Joseph Julicher <[email protected]> Co-authored-by: Paul Bartell <[email protected]> Co-authored-by: Christos Zosimidis <[email protected]> Co-authored-by: Kody Stribrny <[email protected]> Co-authored-by: Holden <holden-zenithaerotech.com> Co-authored-by: Darian <[email protected]> Co-authored-by: Ubuntu <[email protected]> Co-authored-by: Nicolas <[email protected]> Co-authored-by: Vo Trung Chi <[email protected]>

n9wxu requested a review from a team as a code owner November 4, 2022 00:01

archigup mentioned this pull request Nov 30, 2022

[BUG] Compilation fails with -Wall -Werror -Wextra -O3 when using arm gcc 11.1 and up FreeRTOS/FreeRTOS#858

Closed

archigup force-pushed the add_check branch from 9684b36 to 44f0774 Compare December 1, 2022 00:14

archigup changed the title ~~Add check for empty list to listGET_OWNER_OF_NEXT_ENTRY~~ Fix array-bounds compiler warning on gcc11+ in list.h Dec 1, 2022

archigup force-pushed the add_check branch from 44f0774 to eb4819d Compare December 1, 2022 00:27

paulbartell approved these changes Dec 1, 2022

View reviewed changes

n9wxu approved these changes Dec 1, 2022

View reviewed changes

archigup merged commit 99d3d54 into FreeRTOS:main Dec 15, 2022

archigup deleted the add_check branch December 16, 2022 17:42

aggarg added a commit to chinglee-iot/FreeRTOS-Kernel that referenced this pull request Apr 18, 2023

Smp dev merge main 20230410 (#74

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Fix array-bounds compiler warning on gcc11+ in list.h #580

Fix array-bounds compiler warning on gcc11+ in list.h #580

Uh oh!

archigup commented Oct 21, 2022 •

edited

Loading

Uh oh!

codecov bot commented Oct 21, 2022 •

edited

Loading

Uh oh!

RichardBarry commented Oct 21, 2022 •

edited

Loading

Uh oh!

cobusve commented Oct 21, 2022

Uh oh!

RichardBarry commented Oct 21, 2022

Uh oh!

archigup commented Oct 21, 2022

Uh oh!

archigup commented Nov 30, 2022

Uh oh!

archigup commented Nov 30, 2022 •

edited

Loading

Uh oh!

archigup commented Dec 1, 2022

Uh oh!

Fix array-bounds compiler warning on gcc11+ in list.h #580

Fix array-bounds compiler warning on gcc11+ in list.h #580

Uh oh!

Conversation

archigup commented Oct 21, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Test Steps

Related Issue

Uh oh!

codecov bot commented Oct 21, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

RichardBarry commented Oct 21, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

cobusve commented Oct 21, 2022

Uh oh!

RichardBarry commented Oct 21, 2022

Uh oh!

archigup commented Oct 21, 2022

Uh oh!

archigup commented Nov 30, 2022

Uh oh!

archigup commented Nov 30, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

archigup commented Dec 1, 2022

Uh oh!

archigup commented Oct 21, 2022 •

edited

Loading

codecov bot commented Oct 21, 2022 •

edited

Loading

RichardBarry commented Oct 21, 2022 •

edited

Loading

archigup commented Nov 30, 2022 •

edited

Loading