RE: nRF52 and float printf

Fabien Comte — Fri, 22 Jan 2016 08:31:42 GMT

I will try, thank you

RE: nRF52 and float printf

Susheel Nuguru — Mon, 18 Jan 2016 10:18:18 GMT

In Keil Settings->Target->Floating Point Hardware -> Use single presecision

This will do nothiing but just set some compiler macros in core_cm4.h

#elif defined ( __ICCARM__ )
  #if defined __ARMVFP__
    #if (__FPU_PRESENT == 1U)
      #define __FPU_USED       1U
    #else
      #error "Compiler generates FPU instructions for a device without an FPU (check __FPU_PRESENT)"
      #define __FPU_USED       0U
    #endif
  #else
    #define __FPU_USED         0U
  #endif

This is inturn used in system_nrf52.c->SystemInit()

#if (__FPU_USED == 1)
    SCB->CPACR |= (3UL << 20) | (3UL << 22); 
    __DSB();
    __ISB();
#endif

Then your compiler will generate hardware floating point instructions and your printf shouldl work normally.

If you have MicroLib set in Settings->Target->Use MicroLIB, then i think it optimizes float instructions. I am not sure how it works here, but i suspect floating point calculations will be tricky. I never use MicroLIB when i want software float.

If you do not want to use avialable VFP hardware in nRF52(i cannot imagine why not but its your choice). Then you can add

Settings->Target->Floating Point Hardware -> Use single presecision
Settings->C/C++->Misc controls->--fpu=softvfp
Settings->Linker->Misc controls->--fpu=softvfp

RE: nRF52 and float printf

monpetit — Mon, 18 Jan 2016 03:14:17 GMT

If you use Eclipse: