Integer Sum combine1: Maximum use of data abstraction:
Best: 9.08 (4%), Overall Best: 9.09 40-most: 9.10 cycles/element
Integer Sum combine2: Take vec_length() out of loop:
Best: 7.01 (4%), Overall Best: 7.02 40-most: 7.02 cycles/element
Integer Sum combine3: Array reference to vector data:
Best: 1.58 (4%), Overall Best: 1.61 40-most: 1.61 cycles/element
Integer Sum combine3w: Update *dest within loop only with write:
Best: 1.57 (2%), Overall Best: 1.63 40-most: 1.62 cycles/element
Integer Sum combine4: Array reference, accumulate in temporary:
Best: 0.17 (4%), Overall Best: 0.19 40-most: 0.19 cycles/element
Integer Sum combine4b: Include bonds check in loop:
Best: 1.74 (14%), Overall Best: 1.75 40-most: 1.75 cycles/element
Integer Sum combine4p: Pointer reference, accumulate in temporary:
Best: 0.17 (6%), Overall Best: 0.19 40-most: 0.19 cycles/element
Integer Sum combine5: Array code, unrolled by 2:
Best: 0.28 (2%), Overall Best: 0.29 40-most: 0.30 cycles/element
Integer Sum combine5p: Pointer code, unrolled by 2, for loop:
Best: 0.28 (2%), Overall Best: 0.29 40-most: 0.30 cycles/element
Integer Sum unroll2aw: Array code, unrolled by 2, while loop:
Best: 0.28 (10%), Overall Best: 0.30 40-most: 0.30 cycles/element
Integer Sum unroll3a: Array code, unrolled by 3:
Best: 1.00 (18%), Overall Best: 1.01 40-most: 1.01 cycles/element
Integer Sum unroll4a: Array code, unrolled by 4:
Best: 0.28 (8%), Overall Best: 0.29 40-most: 0.30 cycles/element
Integer Sum unroll5a: Array code, unrolled by 5:
Best: 1.00 (20%), Overall Best: 1.01 40-most: 1.01 cycles/element
Integer Sum unroll6a: Array code, unrolled by 6:
Best: 1.00 (8%), Overall Best: 1.01 40-most: 1.01 cycles/element
Integer Sum unroll7a: Array code, unrolled by 7:
Best: 1.00 (24%), Overall Best: 1.01 40-most: 1.01 cycles/element
Integer Sum unroll8a: Array code, unrolled by 8:
Best: 0.28 (2%), Overall Best: 0.28 40-most: 0.30 cycles/element
Integer Sum unroll9a: Array code, unrolled by 9:
Best: 1.00 (12%), Overall Best: 1.01 40-most: 1.01 cycles/element
Integer Sum unroll10a: Array code, unrolled by 10:
Best: 1.00 (8%), Overall Best: 1.00 40-most: 1.01 cycles/element
Integer Sum unroll16a: Array code, unrolled by 16:
Best: 0.21 (2%), Overall Best: 0.22 40-most: 0.23 cycles/element
Integer Sum unroll2: Pointer code, unrolled by 2:
Best: 0.28 (8%), Overall Best: 0.29 40-most: 0.30 cycles/element
Integer Sum unroll3: Pointer code, unrolled by 3:
Best: 1.00 (46%), Overall Best: 1.00 40-most: 1.01 cycles/element
Integer Sum unroll4: Pointer code, unrolled by 4:
Best: 0.28 (4%), Overall Best: 0.30 40-most: 0.30 cycles/element
Integer Sum unroll8: Pointer code, unrolled by 8:
Best: 0.28 (28%), Overall Best: 0.28 40-most: 0.30 cycles/element
Integer Sum unroll16: Pointer code, unrolled by 16:
Best: 0.21 (28%), Overall Best: 0.22 40-most: 0.22 cycles/element
Integer Sum combine6: Array code, unrolled by 2, Superscalar x2:
Best: 0.28 (2%), Overall Best: 0.29 40-most: 0.30 cycles/element
Integer Sum unroll4x2a: Array code, unrolled by 4, Superscalar x2:
Best: 0.68 (14%), Overall Best: 0.69 40-most: 0.69 cycles/element
Integer Sum unroll8x2a: Array code, unrolled by 8, Superscalar x2:
Best: 0.53 (18%), Overall Best: 0.54 40-most: 0.54 cycles/element
Integer Sum unroll3x3a: Array code, unrolled by 3, Superscalar x3:
Best: 0.72 (2%), Overall Best: 0.73 40-most: 0.74 cycles/element
Integer Sum unroll4x4a: Array code, unrolled by 4, Superscalar x4:
Best: 0.28 (2%), Overall Best: 0.30 40-most: 0.30 cycles/element
Integer Sum unroll5x5a: Array code, unrolled by 5, Superscalar x5:
Best: 0.57 (2%), Overall Best: 0.58 40-most: 0.59 cycles/element
Integer Sum unroll6x6a: Array code, unrolled by 6, Superscalar x6:
Best: 0.58 (44%), Overall Best: 0.58 40-most: 0.59 cycles/element
Integer Sum unroll7x7a: Array code, unrolled by 7, Superscalar x7:
Best: 0.54 (22%), Overall Best: 0.55 40-most: 0.55 cycles/element
Integer Sum unroll8x4a: Array code, unrolled by 8, Superscalar x4:
Best: 0.53 (8%), Overall Best: 0.54 40-most: 0.54 cycles/element
Integer Sum unroll8x8a: Array code, unrolled by 8, Superscalar x8:
Best: 0.53 (8%), Overall Best: 0.54 40-most: 0.55 cycles/element
Integer Sum unroll9x9a: Array code, unrolled by 9, Superscalar x9:
Best: 0.54 (28%), Overall Best: 0.54 40-most: 0.55 cycles/element
Integer Sum unroll10x10a: Array code, unrolled by 10, Superscalar x10:
Best: 0.53 (6%), Overall Best: 0.54 40-most: 0.55 cycles/element
Integer Sum unroll2x6a: Array code, unrolled by 12, Superscalar x6:
Best: 0.55 (36%), Overall Best: 0.55 40-most: 0.56 cycles/element
Integer Sum unroll12x12a: Array code, unrolled by 12, Superscalar x12:
Best: 0.57 (2%), Overall Best: 0.60 40-most: 0.60 cycles/element
Integer Sum unroll8x2: Pointer code, unrolled by 8, Superscalar x2:
Best: 0.51 (4%), Overall Best: 0.52 40-most: 0.53 cycles/element
Integer Sum unroll8x4: Pointer code, unrolled by 8, Superscalar x4:
Best: 0.52 (56%), Overall Best: 0.53 40-most: 0.53 cycles/element
Integer Sum unroll8x8: Pointer code, unrolled by 8, Superscalar x8:
Best: 0.50 (10%), Overall Best: 0.51 40-most: 0.51 cycles/element
Integer Sum unroll9x3: Pointer code, unrolled by 9, Superscalar x3:
Best: 0.53 (10%), Overall Best: 0.54 40-most: 0.55 cycles/element
Integer Sum unrollx2as: Array code, Unroll x2, Superscalar x2, noninterleaved:
Best: 0.20 (8%), Overall Best: 0.22 40-most: 0.22 cycles/element
Integer Sum combine7: Array code, unrolled by 2, different associativity:
Best: 0.66 (80%), Overall Best: 0.66 40-most: 0.66 cycles/element
Integer Sum unroll3aa: Array code, unrolled by 3, Different Associativity:
Best: 0.74 (36%), Overall Best: 0.74 40-most: 0.75 cycles/element
Integer Sum unroll4aa: Array code, unrolled by 4, Different Associativity:
Best: 1.28 (4%), Overall Best: 1.28 40-most: 1.30 cycles/element
Integer Sum unroll5aa: Array code, unrolled by 5, Different Associativity:
Best: 0.59 (12%), Overall Best: 0.59 40-most: 0.60 cycles/element
Integer Sum unroll6aa: Array code, unrolled by 6, Different Associativity:
Best: 0.58 (34%), Overall Best: 0.59 40-most: 0.59 cycles/element
Integer Sum unroll7aa: Array code, unrolled by 7, Different Associativity:
Best: 0.54 (6%), Overall Best: 0.56 40-most: 0.56 cycles/element
Integer Sum unroll8aa: Array code, unrolled by 8, Different Associativity:
Best: 1.91 (20%), Overall Best: 1.92 40-most: 1.92 cycles/element
Integer Sum unroll9aa: Array code, unrolled by 9, Different Associativity:
Best: 0.54 (6%), Overall Best: 0.56 40-most: 0.56 cycles/element
Integer Sum unroll10aa: Array code, unrolled by 10, Different Associativity:
Best: 0.54 (12%), Overall Best: 0.55 40-most: 0.56 cycles/element
Integer Sum unroll12aa: Array code, unrolled by 12, Different Associativity:
Best: 0.55 (4%), Overall Best: 0.56 40-most: 0.57 cycles/element
Integer Sum simd_v1: SSE code, 1*VSIZE-way parallelism:
Best: 1.00 (98%), Overall Best: 1.00 40-most: 1.00 cycles/element
Integer Sum simd_v2: SSE code, 2*VSIZE-way parallelism:
Best: 0.14 (62%), Overall Best: 0.14 40-most: 0.15 cycles/element
Integer Sum simd_v4: SSE code, 4*VSIZE-way parallelism:
Best: 0.06 (8%), Overall Best: 0.08 40-most: 0.08 cycles/element
Integer Sum simd_v8: SSE code, 8*VSIZE-way parallelism:
Best: 0.05 (38%), Overall Best: 0.05 40-most: 0.06 cycles/element
Integer Sum simd_v12: SSE code, 12*VSIZE-way parallelism:
Best: 0.04 (56%), Overall Best: 0.04 40-most: 0.05 cycles/element
Integer Sum simd_v2a: SSE code, 2*VSIZE-way parallelism, reassociate:
Best: 0.56 (48%), Overall Best: 0.56 40-most: 0.57 cycles/element
Integer Sum simd_v4a: SSE code, 4*VSIZE-way parallelism, reassociate:
Best: 0.07 (10%), Overall Best: 0.07 40-most: 0.09 cycles/element
Integer Sum simd_v8a: SSE code, 8*VSIZE-way parallelism, reassociate:
Best: 0.05 (26%), Overall Best: 0.06 40-most: 0.06 cycles/element