Cause:
This vectorization diagnostic is emitted starting from Intel(R) C++ Compiler 15.0 and with report level 3 and above. This diagnostic message emits the static estimate of the number of execution cycles which each iteration of the loop body will take when it executes with scalar operands. Below is a code snippet which demonstrates this.
Examples:
#include<stdio.h>
#include<math.h>
#define SIZE 100
int main(){
float a[SIZE], x;
a[:] = __sec_implicit_index(0);
x = __sec_reduce_all_zero(a[:] == 1 ? 0 : 1);
printf("%fn",x);
return 0;
}Begin optimization report for: main()
Report from: Vector optimizations [vec]
LOOP BEGIN at test11.cc(6,12)
remark #15300: LOOP WAS VECTORIZED
remark #15449: unmasked aligned unit stride stores: 1
remark #15475: --- begin vector loop cost summary ---
remark #15476: scalar loop cost: 4
remark #15477: vector loop cost: 1.000
remark #15478: estimated potential speedup: 3.960
remark #15479: lightweight vector operations: 4
remark #15487: type converts: 1
remark #15488: --- end vector loop cost summary ---
LOOP END
LOOP BEGIN at test11.cc(7,9)
remark #15300: LOOP WAS VECTORIZED
remark #15448: unmasked aligned unit stride loads: 1
remark #15475: --- begin vector loop cost summary ---
remark #15476: scalar loop cost: 165
remark #15477: vector loop cost: 43.250
remark #15478: estimated potential speedup: 3.810
remark #15479: lightweight vector operations: 7
remark #15481: heavy-overhead vector operations: 1
remark #15488: --- end vector loop cost summary ---
LOOP END