AnsweredAssumed Answered

About using AMMCLIB

Question asked by dongdong dong on Oct 27, 2019
Latest reply on Nov 12, 2019 by stanish


I recently had a problem with AMMCLIB math library. The version of AMMCLIB math library is MPC577xK_AMMCLIB_v1.1.9 - s32ds_ppc, MCU is S32R274.

I tested the sine function, and I found that  GFLIB_Sin_FLT is OK, but  GFLIB_Sin_F32 and  GFLIB_Sin_F16 are not OK when I use MPC577xK_AMMCLIB.a in e200z7.If I want to use the GFLIB_Sin_F32 function, what do I need to do?

Another, when I use MPC577xK_AMMCLIB.a in e200z4, I found that the three functions are all OK, but GFLIB_Sin_F32 and GFLIB_Sin_F16 running time is more than GFLIB_Sin_FLT, I can't understand that because I think integral types should be less computation-intensive than floating point types.

What is the answer to the above question? 

Or are there other ways to improve software performance?


I am looking forward to your reply!


Best wishes