[BUG] __VFP_FP__ Macro Does Not Correctly Indicate Floating Point Enabled for ARM Targets

[haydenridd]

Describe the bug The macro __VFP_FP__ set by the compiler is deprecated, and despite its name does not indicate that floating support is enabled. This is only used for 3 ARM targets, and should be updated to __ARM_FP which is the appropriate flag.

Target

• Development board: N/A, but affects ARM ports GCC/ARM_CM7, GCC/ARM_CM4_MPU, and GCC/ARM_CM4F
• Instruction Set Architecture: ARM Thumb ISA
• IDE and version: N/A, main branch of FreeRTOS Kernel
• Toolchain and version: gcc version 13.2.1 20231009 (Arm GNU Toolchain 13.2.rel1 (Build arm-13.7))

Host

• Host OS: Fedora 37
• Version: 6.5.12-100.fc37.x86_64

To Reproduce

• Compile a dummy file containing:

  #ifndef __VFP_FP__
  #error This port can only be used when the project options are configured to enable hardware floating point support.
  #endif

  And observe that independent of whether or not floating support is enabled, the error never triggers.

Expected behavior These ports should fail to compile when floating support is not enabled.

Additional context See the following links for an explanation of why this flag is just set to a constant 1 by both GCC and Clang:

• https://reviews.llvm.org/D100372
• https://gcc.gnu.org/legacy-ml/gcc-patches/2016-10/msg01025.html

I will create a PR to fix this!

[aggarg]

Thank you for your contribution!