Cortex-M3 Data type: signed int

Question

I have studied the dsPIC33 for FFT operation and realised the int16 is native fractional type Q15 and int32 Q31.

I now studying using math in NXP Cortex-M3, I have looked into document about data type and could not find reference (including CMSIS) to define what is int32_t actually is.

My question: is this int32_t a (native) fractional type (Q31?), if not what are they.

Is there (easy) presentation material that details data type defined by CMSIS or NXP to be used for math in general?, if so can you provide me links.

BTW, I use NXPxpresso and CMSIS-3.

Answer 1

define what is int32_t actually is

Look in <stdint.h>, it is part of C99. On Cortex M3 (ARM) compilers usually defined as long.

is int32_t a (native) fractional type

No, it is just a normal integer type.