Why is the speed of operations not constant? Is the cache the reason?

If your system runs at 100MHz, then you will have a clock cycle of 10ns.

you say that the 1. & 2. and 4th take 20ns = 2 clocks for all 512 iterations but the 3. takes 100ns = 10 cycles for all 512 loop steps.

if some interrupt or refresh cycle or whatsoever occcurs then you would measure this additional time within all 4 loops and not only in loop 3.

this is indeed strange

have you setup signaltap and monitored what is giong on on each of the 4 loop cycles ?

especially those custum instructions.

another question

you wrote

Real_f32 = CI_INTTOFLOAT(Real_s16);

Imag_f32 = CI_INTTOFLOAT(Imag_s16);

Result1_f32a[SampleCounter_u16] = CI_FPSQRT(Real_f32*Real_f32 +

Imag_f32*Imag_f32)/STEP_SIZE_1MILIG;

so you have 2 float multiplikation and one float addition

why don't you write

Result1_f32a[SampleCounter_u16] = CI_FPSQRT( CI_INTTOFLOAT( Real_s16 * Real_s16 + Imag_s16 * Imag_s16 ) )/STEP_SIZE_1MILIG;

now you have 4 integer multiplikation and one integer addition, this should be a bit faster as also only one CI_INTTOFLOAT operation is needed.