AnsweredAssumed Answered

NEON has same performance as C

Question asked by Alexandre Caron on Dec 10, 2019
Latest reply on Dec 10, 2019 by Victor Linnik

Hi i'm developing and image processing application on the Nxp imx7 and I want to compare performance of NEON instrutions vs pure c.

 

c: a,b,c are float32. Take 11ms to run

for(int pixIndex = 0;pixIndex<(640*480);pixIndex++)
{

      a[pixIndex] = (a[pixIndex] * b[pixIndex]) + c[pixIndex];

}

 

NEON: Take 10ms to run

for(int pixIndex = 0;pixIndex<(640*480)/2;pixIndex++)
{

      

float32x2_t dVect1, dVect2,dVect3;

dVect1 = vld1_f32(a);
dVect2 = vld1_f32(b);
dVect3 = vld1_f32(c);
dVect1 = vmla_f32(dVect3,dVect1,dVect2);
vst1_f32(a,dVect1);
a+=2;

b+=2;

c+=2;

 

}

 

Why NEON is only 1ms faster than c ? Do I miss something here ?

Outcomes