* Use unsigned types/constants where needed.
* Address MISRA 21.15 violations in FreeRTOS_Sockets.c
* Address MISRA rule violations in code (primarily Rule 2.2)
* Inline had been disabled for Coverity builds, preventing
Coverity from correctly identifying dead code; this change
removes the disabling of inline during Coverity builds.
* Added an explanation for the inline suppression of Rule
11.4 in prvSocketValid().
* Address MISRA Rule Violations (10.4 & 10.8)
* MISRA: Rule 21.15 changes
Co-authored-by: Aniruddha Kanhere <60444055+AniruddhaKanhere@users.noreply.github.com>
* Change type of usGenerateChecksum's ulAlignBits to intptr_t
Not all platforms have 32-bit pointers. 8-bit machines such as avr8 have
16-bit pointers.
This patch changes the type of ulAlignBits and renames it to uxAlignBits to
reflect the type change.
This fixes a compiler warning on machines with pointers that aren't
32-bits.
Signed-off-by: Andy Bennett <andyjpb@ashurst.eu.org>
* Fix usGenerateChecksum on odd-aligned buffers with non zero usSum
usGenerateChecksum would generate an incorrect checksum when pucNextData
was odd-aligned and usSum was non-zero.
This was caused by the byte order of usSum not matching the byte order of
the subsequent summing operation.
Odd-aligned buffers are common on 8-bit platforms such as avr8 when using
one of the FreeRTOS dynamic heap allocators.
Signed-off-by: Andy Bennett <andyjpb@ashurst.eu.org>
* Feedback from PR#122
https://github.com/FreeRTOS/FreeRTOS/pull/122
+ Use a uintptr_t rather than an intptr_t.
Changes supplied by Hein Tibosch.
Signed-off-by: Andy Bennett <andyjpb@ashurst.eu.org>
Co-authored-by: Aniruddha Kanhere <60444055+AniruddhaKanhere@users.noreply.github.com>
The rendezvous test in EventGroupsDemo assumed that one tick is not
enough for rendezvous to complete. Turns out that it is not true on a
system with 1ms tick and rendezvous does get complete within one tick.
This commit removes the test with one tick assumption.
Signed-off-by: Gaurav Aggarwal <aggarg@amazon.com>
* Add Renesas RX72N Envision Kit RTOS Demo using Renesas RXv3 port layer
* Refactor the workaround which makes CC-RX linker optimization working better
To be robust for future updates of Renesas RX SmartConfigurator until the cause of the problem is fixed by Renesas.
* Fix vApplicationMallocFailedHook()'s argument type generated by e2 studio.
No changes are caused in the MOT and SREC files which are built.
* Merge update of submodule (submodule's commit id)
Changes are caused in the MOT and SREC files which are built for Renesas RX72N Envision Kit RTOS Demo.
Theses RTOS Demo programs work after changes.
* Fix: Add checking DPFPU register in RegTest inline asm functions in main_full.c
* Fix and sync kernel pointer (submodule's commit id) to the same as 'upstream/master'
* Revert "Fix and sync kernel pointer (submodule's commit id) to the same as 'upstream/master'"
This reverts commit d684f749bc.
* Revert "Merge remote-tracking branch 'upstream/master' into rx700v3dpfpu"
This reverts commit 3a44473b4e, reversing
changes made to 0f4f024fe2.
* Merge remote-tracking branch 'upstream/master' into rx700v3dpfpu
* Fix and sync kernel pointer (submodule's commit id) to the same as 'upstream/master'
* Test: Fix unnecessary update of SiFive_HiFive1_RTOS_demo.url.
* Fix? unnecessary update of SiFive_HiFive1_RTOS_demo.url.
* Fix unnecessary update of SiFive_HiFive1_RTOS_demo.url.
The existing startup code was generating some warnings of the form
"Label 'xxxxx' is defined pubweak in a section implicitly declared
root". This change replaces the startup file with the one generated from
latest STM32CubeMX to fix the above warnings.
Signed-off-by: Gaurav Aggarwal <aggarg@amazon.com>
* Fix: Demo/RX100-RSK_Renesas_e2studio RTOS demo project build error
* Update project files for e2 studio v4.0.2.008 (the same as Demo/RX100_RX113-RSK_Renesas_e2studio RTOS demo project)
Co-authored-by: Ming Yue <mingyue86010@gmail.com>
* Use unsigned types/constants where needed.
* Address MISRA 21.15 violations in FreeRTOS_Sockets.c
* Address MISRA rule violations in code (primarily Rule 2.2)
* Inline had been disabled for Coverity builds, preventing
Coverity from correctly identifying dead code; this change
removes the disabling of inline during Coverity builds.
* Added an explanation for the inline suppression of Rule
11.4 in prvSocketValid().
Co-authored-by: Aniruddha Kanhere <60444055+AniruddhaKanhere@users.noreply.github.com>
