Double Product combine1: Maximum use of data abstraction:
28.07 cycles/element
Double Product combine2: Take vec_length() out of loop:
28.00 cycles/element
Double Product combine3: Array reference to vector data:
7.09 cycles/element
Double Product combine4: Array reference, accumulate in temporary:
7.00 cycles/element
Double Product combine4p: Pointer reference, accumulate in temporary:
7.67 cycles/element
Double Product Array code, unrolled by 2:
7.00 cycles/element
Double Product combine5p: Pointer code, unrolled by 3, for loop:
7.00 cycles/element
Double Product Array code, unrolled by 3, while loop:
7.00 cycles/element
Double Product Array code, unrolled by 4:
7.01 cycles/element
Double Product Array code, unrolled by 8:
7.00 cycles/element
Double Product Array code, unrolled by 16:
7.01 cycles/element
Double Product Pointer code, unrolled by 2:
7.00 cycles/element
Double Product Pointer code, unrolled by 3:
7.01 cycles/element
Double Product Pointer code, unrolled by 4:
6.99 cycles/element
Double Product Pointer code, unrolled by 8:
7.00 cycles/element
Double Product Pointer code, unrolled by 16:
6.99 cycles/element
Double Product combine6: Array code, unrolled by 2, Superscalar x2:
3.50 cycles/element
Double Product Array code, unrolled by 4, Superscalar x2:
3.52 cycles/element
Double Product Array code, unrolled by 8, Superscalar x2:
3.50 cycles/element
Double Product Array code, unrolled by 3, Superscalar x3:
2.34 cycles/element
Double Product Array code, unrolled by 4, Superscalar x4:
2.00 cycles/element
Double Product Array code, unrolled by 8, Superscalar x4:
2.00 cycles/element
Double Product Array code, unrolled by 6, Superscalar x6:
2.00 cycles/element
Double Product Array code, unrolled by 8, Superscalar x8:
2.00 cycles/element
Double Product Array code, unrolled by 10, Superscalar x10:
2.00 cycles/element
Double Product Array code, unrolled by 12, Superscalar x6:
2.00 cycles/element
Double Product Array code, unrolled by 12, Superscalar x12:
2.01 cycles/element
Double Product Pointer code, unrolled by 8, Superscalar x2:
3.50 cycles/element
Double Product Pointer code, unrolled by 8, Superscalar x4:
2.00 cycles/element
Double Product Pointer code, unrolled by 8, Superscalar x8:
1.99 cycles/element
Double Product Pointer code, unrolled by 9, Superscalar x3:
2.33 cycles/element
Double Product Array code, Unroll x2, Superscalar x2, noninterleaved:
3.50 cycles/element
Double Product Array code, unrolled by 2, different associativity:
6.51 cycles/element
Double Product Array code, unrolled by 3, Different Associativity:
6.66 cycles/element
Double Product Array code, unrolled by 4, Different Associativity:
5.00 cycles/element
Double Product Array code, unrolled by 6, Different Associativity:
4.67 cycles/element
Double Product Array code, unrolled by 8, Different Associativity:
3.38 cycles/element
Double Product SSE code, 1*VSIZE-way parallelism:
3.52 cycles/element
Double Product SSE code, 2*VSIZE-way parallelism:
2.45 cycles/element
Double Product SSE code, 4*VSIZE-way parallelism:
1.84 cycles/element
Double Product SSE code, 8*VSIZE-way parallelism:
1.15 cycles/element
Double Product SSE code, 12*VSIZE-way parallelism:
1.04 cycles/element
Double Product SSE code, 2*VSIZE-way parallelism, reassociate:
2.09 cycles/element
Double Product SSE code, 4*VSIZE-way parallelism, reassociate:
1.73 cycles/element
Double Product SSE code, 8*VSIZE-way parallelism, reassociate:
1.03 cycles/element