ARM NEON header file missing float16_t type RRS feed

  • Question

  • Hello,

    According to the ARM manual


    half-precision floating point types (float16_t) (as well as vectors at this type) are supported by NEON. I am searching at the "arm_neon.h" header file which is under my visual studio/include directory and no float at 16 bit precision is there. 

    Is there any other header file where that types exist? 

    Not all the ARM architectures support half precision floating point but shouldn't be the header file include all the possible data types are supported by different ARM architectures?



    • Edited by john_la Friday, March 4, 2016 6:46 PM
    Friday, March 4, 2016 6:45 PM