AnsweredAssumed Answered

I want to see new built-in functions in VDSP++

Question asked by bookevg on Nov 15, 2011
Latest reply on Nov 25, 2011 by bookevg

I move from ADSP21992 to ADSP21489. My code in   ADSP21992 was written in asm. I used fixed point and my float point.

Then I need more perfomance processor and developing my code in C/С++. I draw attention ADSP21xxx because it have 40bit float point and it's AD.

Unfortunately the compiler don't full support 40bit float point especially because the compiler use the function "Rn = pass Rx" unpredictably.

My algorithms are sensitive to the mantissa width.

Now I deceided to use fixed point. But I want to use asm's construction because the optimization don't work.

Existing built-in function are scanty.

I need:

unsigned fract * fract

unsigned fract *   unsigned fract

fract + unsigned fract * fract

 

unsigned fract  + unsigned fract *   unsigned fract

convert unsigned fract to float

sat

It would super if VDSP++ have almost instructions of the processor

My colleague which work with Code Composer of TI's DSP has almost instructions of the processor as built-in functions (intrisinc).

Also these built-in functions in Code Composer let to use SIMD without the compiler.

And he may reach the maximum perfomance, but I cant' do these because the compiler isn't versatile.

 

Outcomes