|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC15
|
-35.96% |
27.159 |
17.392 |
0.000 |
-1.35% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC8
|
-34.78% |
16.438 |
10.721 |
0.000 |
-6.25% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC2
|
-34.76% |
7.862 |
5.129 |
0.024 |
-4.59% |
0.024 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC64
|
-34.04% |
97.197 |
64.116 |
0.002 |
-0.32% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC31
|
-33.82% |
49.315 |
32.638 |
0.000 |
-0.73% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC32
|
-33.80% |
50.745 |
33.593 |
0.002 |
-0.70% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC16
|
-33.77% |
27.874 |
18.461 |
0.001 |
-0.66% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC63
|
-33.37% |
95.057 |
63.336 |
0.006 |
-0.05% |
0.006 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC7
|
-33.34% |
15.009 |
10.006 |
0.000 |
-0.00% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC4
|
-31.25% |
11.435 |
7.862 |
0.000 |
0.00% |
0.000 |
|
MultiSource/Benchmarks/MiBench/automotive-bitcount/automotive-bitcount
Profile
|
-28.75% |
0.099 |
0.070 |
0.000 |
-28.90% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchAutoVecForLoopTC2
|
-23.08% |
9.291 |
7.147 |
0.000 |
-0.01% |
0.000 |
|
MicroBenchmarks/LCALS/SubsetCRawLoops/lcalsCRaw.test:BM_FIRST_DIFF_RAW/171
|
-20.60% |
0.373 |
0.296 |
0.003 |
-20.75% |
0.003 |
|
MultiSource/Benchmarks/Trimaran/enc-pc1/enc-pc1
Profile
|
-18.62% |
0.567 |
0.461 |
0.000 |
-18.97% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC128
|
-18.46% |
193.684 |
157.930 |
0.011 |
2.73% |
0.011 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_csa_with_cond_arith_novec_int64_t_
|
-18.06% |
845917.775 |
693154.609 |
360.569 |
0.33% |
360.569 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC127
|
-17.27% |
192.976 |
159.657 |
0.310 |
-0.08% |
0.310 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopTC2
|
-16.66% |
8.576 |
7.147 |
0.000 |
-16.66% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchAutoVecForBigLoopTC2
|
-16.66% |
8.577 |
7.147 |
0.000 |
-0.01% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:benchForTruncOrZextVecInLoopWithVW16From_uint16_t_To_uint8_t_
|
-16.62% |
14301.400 |
11924.235 |
4.452 |
-0.03% |
4.452 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:benchForTruncOrZextVecInLoopFrom_uint8_t_To_uint32_t_
|
-16.60% |
14301.346 |
11927.801 |
0.240 |
-0.05% |
0.240 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:benchForTruncOrZextVecInLoopFrom_uint16_t_To_uint32_t_
|
-16.60% |
14301.618 |
11928.108 |
0.123 |
-0.05% |
0.123 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:benchForTruncOrZextVecInLoopWithVW8From_uint16_t_To_uint8_t_
|
-16.59% |
14301.326 |
11928.668 |
0.177 |
-0.01% |
0.177 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:benchForTruncOrZextVecInLoopFrom_uint16_t_To_uint8_t_
|
-16.53% |
14301.637 |
11938.190 |
6.010 |
0.06% |
6.010 |
|
External/SPEC/CINT2017rate/525.x264_r/525.x264_r
Profile
|
-16.43% |
122.249 |
102.166 |
0.052 |
-16.53% |
0.052 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:benchForTruncOrZextVecInLoopWithVW8From_uint32_t_To_uint8_t_
|
-16.15% |
14302.107 |
11992.890 |
1.362 |
-0.07% |
1.362 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint32_t>/16
|
-15.69% |
36.450 |
30.733 |
0.001 |
-2.27% |
0.001 |
|
SingleSource/Benchmarks/Shootout-C++/Shootout-C++-matrix
Profile
|
-15.62% |
6.590 |
5.560 |
0.000 |
-15.62% |
0.000 |
|
MicroBenchmarks/ImageProcessing/Blur/blur.test:BENCHMARK_boxBlurKernel/512
|
-15.44% |
20206.086 |
17085.707 |
75.524 |
-15.27% |
75.524 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC3
|
-15.38% |
9.291 |
7.862 |
0.000 |
-0.00% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_single_csa_only_autovec_int64_t_
|
-15.03% |
795510.808 |
675983.543 |
1138.336 |
-14.82% |
1138.336 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_single_csa_cond_load_autovec_int64_t_
|
-14.98% |
797385.405 |
677972.842 |
248.325 |
-15.14% |
248.325 |
|
MicroBenchmarks/ImageProcessing/Blur/blur.test:BENCHMARK_boxBlurKernel/128
|
-13.84% |
1075.450 |
926.605 |
0.285 |
-14.06% |
0.285 |
|
MicroBenchmarks/ImageProcessing/Blur/blur.test:BENCHMARK_boxBlurKernel/256
|
-13.74% |
4576.471 |
3947.588 |
0.837 |
-13.85% |
0.837 |
|
MicroBenchmarks/ImageProcessing/Blur/blur.test:BENCHMARK_boxBlurKernel/1024
|
-12.93% |
84714.250 |
73764.222 |
21.677 |
-12.96% |
21.677 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint32_t>/28
|
-12.70% |
45.027 |
39.310 |
0.001 |
-1.79% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint8_t>/16
|
-11.77% |
36.452 |
32.162 |
0.001 |
2.27% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC8
|
-11.76% |
12.150 |
10.721 |
0.000 |
-34.78% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC2
|
-11.60% |
5.718 |
5.054 |
0.034 |
-41.07% |
0.034 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4LoopTC1
|
-11.11% |
5.718 |
5.083 |
0.000 |
-11.08% |
0.000 |
|
MultiSource/Benchmarks/Trimaran/enc-md5/enc-md5
Profile
|
-10.79% |
3.009 |
2.684 |
0.000 |
-10.80% |
0.000 |
|
MicroBenchmarks/LCALS/SubsetCLambdaLoops/lcalsCLambda.test:BM_FIND_FIRST_MIN_LAMBDA/5001
|
-10.71% |
12.018 |
10.731 |
0.000 |
-11.92% |
0.000 |
|
MicroBenchmarks/ImageProcessing/Dither/Dither.test:BENCHMARK_FLOYD_DITHER/256
|
-10.18% |
1104.850 |
992.394 |
0.281 |
-10.12% |
0.281 |
|
MicroBenchmarks/ImageProcessing/Dither/Dither.test:BENCHMARK_FLOYD_DITHER/512
|
-10.17% |
4442.342 |
3990.420 |
2.541 |
-10.01% |
2.541 |
|
MicroBenchmarks/LCALS/SubsetCRawLoops/lcalsCRaw.test:BM_BAND_LIN_EQ_RAW/171
|
-9.91% |
0.245 |
0.221 |
0.000 |
-1.28% |
0.000 |
|
MicroBenchmarks/LCALS/SubsetARawLoops/lcalsARaw.test:BM_ENERGY_CALC_RAW/44217
|
-9.80% |
2275.638 |
2052.655 |
64.639 |
-8.33% |
64.639 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint8_t>/28
|
-9.53% |
45.029 |
40.739 |
0.001 |
3.63% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint32_t>/51
|
-9.41% |
60.753 |
55.034 |
0.001 |
-1.28% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopEpilogueVectorizationBenchmarks.test:benchReductionAutoVec<uint8_t>/65
|
-9.37% |
91.482 |
82.906 |
0.002 |
-0.00% |
0.002 |
|
MicroBenchmarks/ImageProcessing/Dither/Dither.test:BENCHMARK_FLOYD_DITHER/128
|
-9.09% |
273.851 |
248.960 |
0.284 |
-8.98% |
0.284 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC3
|
-9.09% |
7.862 |
7.147 |
0.000 |
-0.00% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC3
|
-9.09% |
7.862 |
7.147 |
0.000 |
-23.08% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint8_t>/10
|
-8.89% |
32.163 |
29.303 |
0.001 |
0.00% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint32_t>/10
|
-8.89% |
32.163 |
29.303 |
0.001 |
2.50% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW1LoopWithReductionTC3
|
-8.34% |
8.577 |
7.862 |
0.000 |
-0.00% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC4
|
-8.33% |
8.577 |
7.862 |
0.000 |
-0.00% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC3
|
-8.33% |
8.576 |
7.862 |
0.000 |
9.99% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC4
|
-8.33% |
8.577 |
7.862 |
0.000 |
-26.66% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint8_t>/51
|
-8.24% |
60.754 |
55.750 |
0.001 |
-1.26% |
0.001 |
|
MicroBenchmarks/LCALS/SubsetCLambdaLoops/lcalsCLambda.test:BM_HYDRO_1D_LAMBDA/171
|
-8.20% |
0.497 |
0.456 |
0.000 |
0.00% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/999
|
-8.18% |
2882.552 |
2646.642 |
0.033 |
-8.21% |
0.033 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/999
|
-8.07% |
1460.900 |
1342.988 |
0.018 |
-8.03% |
0.018 |
|
MicroBenchmarks/LCALS/SubsetCRawLoops/lcalsCRaw.test:BM_FIRST_DIFF_RAW/5001
|
-7.94% |
10.874 |
10.010 |
0.000 |
-7.92% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4BigLoopTC31
|
-7.90% |
297.556 |
274.048 |
0.001 |
-4.93% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint8_t>/10
|
-7.85% |
36.452 |
33.592 |
0.000 |
-2.08% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint32_t>/10
|
-7.84% |
36.450 |
33.592 |
0.001 |
-2.09% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/256
|
-7.82% |
758.341 |
699.011 |
0.009 |
-7.91% |
0.009 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/16
|
-7.69% |
46.457 |
42.884 |
0.001 |
-4.76% |
0.001 |
|
External/SPEC/CFP2017rate/508.namd_r/508.namd_r
Profile
|
-7.38% |
83.745 |
77.568 |
0.036 |
-7.67% |
0.036 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/51
|
-7.30% |
166.535 |
154.383 |
0.005 |
-6.49% |
0.005 |
|
SingleSource/Benchmarks/Shootout/Shootout-lists
Profile
|
-7.26% |
10.329 |
9.579 |
0.014 |
-7.33% |
0.014 |
|
SingleSource/Benchmarks/McGill/misr
Profile
|
-7.18% |
0.442 |
0.410 |
0.000 |
-7.19% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/256
|
-7.15% |
400.254 |
371.652 |
0.011 |
-6.98% |
0.011 |
|
MicroBenchmarks/LCALS/SubsetALambdaLoops/lcalsALambda.test:BM_ENERGY_CALC_LAMBDA/44217
|
-6.86% |
2164.619 |
2016.226 |
12.919 |
-2.47% |
12.919 |
|
MicroBenchmarks/ImageProcessing/Blur/blur.test:BENCHMARK_GAUSSIAN_BLUR/128
|
-6.80% |
3593.764 |
3349.483 |
0.281 |
-6.81% |
0.281 |
|
MicroBenchmarks/ImageProcessing/Blur/blur.test:BENCHMARK_GAUSSIAN_BLUR/256
|
-6.79% |
15346.000 |
14304.061 |
12.097 |
-6.82% |
12.097 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/28
|
-6.74% |
63.612 |
59.322 |
0.002 |
-4.60% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4LoopTC3
|
-6.67% |
10.721 |
10.006 |
0.000 |
0.00% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC7
|
-6.67% |
10.721 |
10.006 |
0.000 |
-0.00% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW1LoopWithReductionTC2
|
-6.64% |
5.437 |
5.076 |
0.008 |
-5.99% |
0.008 |
|
MicroBenchmarks/ImageProcessing/Blur/blur.test:BENCHMARK_GAUSSIAN_BLUR/512
|
-6.53% |
63417.455 |
59277.167 |
113.124 |
-6.64% |
113.124 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/28
|
-6.38% |
100.775 |
94.344 |
0.003 |
-5.04% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint32_t>/16
|
-6.35% |
45.028 |
42.167 |
0.009 |
-1.67% |
0.009 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint8_t>/16
|
-6.35% |
45.029 |
42.169 |
0.000 |
-1.67% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC8
|
-6.25% |
11.436 |
10.721 |
0.000 |
-2.17% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchAutoVecForBigLoopTC1
|
-6.25% |
5.718 |
5.360 |
0.000 |
-6.25% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopTC1
|
-6.25% |
5.718 |
5.360 |
0.000 |
-6.25% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchAutoVecForLoopTC1
|
-6.25% |
5.718 |
5.361 |
0.000 |
-6.25% |
0.000 |
|
MicroBenchmarks/LCALS/SubsetCLambdaLoops/lcalsCLambda.test:BM_FIND_FIRST_MIN_LAMBDA/44217
|
-6.11% |
125.729 |
118.048 |
0.663 |
-4.53% |
0.663 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopTC63
|
-6.07% |
147.609 |
138.653 |
0.003 |
-6.09% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC2
|
-6.05% |
5.439 |
5.110 |
0.025 |
-5.82% |
0.025 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/51
|
-6.02% |
95.059 |
89.340 |
0.004 |
-4.58% |
0.004 |
|
MultiSource/Applications/hexxagon/hexxagon
Profile
|
-5.91% |
9.582 |
9.016 |
0.007 |
-5.90% |
0.007 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4BigLoopTC15
|
-5.90% |
145.085 |
136.518 |
0.002 |
-5.90% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopTC4
|
-5.88% |
12.150 |
11.435 |
0.000 |
0.00% |
0.000 |
|
MultiSource/Benchmarks/Trimaran/enc-rc4/enc-rc4
Profile
|
-5.82% |
1.938 |
1.825 |
0.001 |
-5.70% |
0.001 |
|
SingleSource/Benchmarks/Adobe-C++/loop_unroll
Profile
|
-5.77% |
1.717 |
1.618 |
0.001 |
-5.62% |
0.001 |
|
MicroBenchmarks/SLPVectorization/SLPVectorizationBenchmarks.test:benchmark_add_xor_no_runtime_checks_needed<16, int>
|
-5.45% |
39.309 |
37.165 |
0.001 |
-5.45% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4BigLoopTC63
|
-5.44% |
591.548 |
559.341 |
0.009 |
-4.75% |
0.009 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4BigLoopTC128
|
-5.43% |
1196.506 |
1131.492 |
0.016 |
-5.08% |
0.016 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4BigLoopTC15
|
-5.40% |
144.307 |
136.521 |
0.735 |
-3.23% |
0.735 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/16
|
-5.38% |
66.471 |
62.897 |
0.001 |
-3.30% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC2
|
-5.29% |
5.435 |
5.148 |
0.000 |
-5.05% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4BigLoopTC127
|
-5.28% |
1187.801 |
1125.073 |
0.347 |
-4.95% |
0.347 |
|
MultiSource/Benchmarks/sim/sim
Profile
|
-5.19% |
6.707 |
6.359 |
0.007 |
-5.23% |
0.007 |
|
MicroBenchmarks/LCALS/SubsetCRawLoops/lcalsCRaw.test:BM_FIRST_DIFF_RAW/44217
|
-5.17% |
96.075 |
91.110 |
0.220 |
-5.16% |
0.220 |
|
MicroBenchmarks/SLPVectorization/SLPVectorizationBenchmarks.test:benchmark_xor_no_runtime_checks_needed<16, int>
|
-5.13% |
27.875 |
26.444 |
0.001 |
-5.13% |
0.001 |
|
MicroBenchmarks/LCALS/SubsetALambdaLoops/lcalsALambda.test:BM_DEL_DOT_VEC_2D_LAMBDA/1
|
-5.04% |
200.298 |
190.198 |
0.494 |
-3.65% |
0.494 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4BigLoopTC31
|
-4.98% |
288.415 |
274.041 |
0.007 |
-4.95% |
0.007 |
|
SingleSource/Benchmarks/Misc/salsa20
Profile
|
-4.97% |
10.977 |
10.431 |
0.000 |
-4.98% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4BigLoopTC127
|
-4.95% |
1183.654 |
1125.077 |
0.036 |
-4.68% |
0.036 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4BigLoopTC128
|
-4.90% |
1190.378 |
1132.037 |
0.021 |
-4.80% |
0.021 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4BigLoopTC31
|
-4.89% |
288.074 |
274.000 |
0.008 |
-4.92% |
0.008 |
|
MultiSource/Benchmarks/VersaBench/ecbdes/ecbdes
Profile
|
-4.85% |
4.024 |
3.828 |
0.000 |
-4.87% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4BigLoopTC64
|
-4.84% |
595.577 |
566.727 |
0.016 |
-4.85% |
0.016 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4BigLoopTC128
|
-4.84% |
1189.692 |
1132.158 |
0.733 |
-4.85% |
0.733 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4BigLoopTC63
|
-4.82% |
587.628 |
559.333 |
0.016 |
-4.82% |
0.016 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4BigLoopTC127
|
-4.79% |
1181.695 |
1125.060 |
0.010 |
-4.80% |
0.010 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4BigLoopTC64
|
-4.76% |
595.687 |
567.332 |
0.018 |
-4.73% |
0.018 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4BigLoopTC64
|
-4.76% |
595.048 |
566.732 |
0.018 |
-4.86% |
0.018 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<8, LessThanZero, Last>
|
-4.76% |
15379.279 |
14647.597 |
7.934 |
-4.75% |
7.934 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<15, LessThanZero, Mid>
|
-4.75% |
8204.351 |
7814.279 |
5.010 |
-4.76% |
5.010 |
|
MicroBenchmarks/LoopVectorization/LoopEpilogueVectorizationBenchmarks.test:benchAutoVec<uint16_t>/65
|
-4.74% |
150.805 |
143.660 |
0.004 |
-0.50% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4BigLoopTC63
|
-4.70% |
586.931 |
559.326 |
0.008 |
-5.33% |
0.008 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint8_t>/28
|
-4.60% |
62.182 |
59.322 |
0.001 |
-1.19% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint32_t>/28
|
-4.60% |
62.182 |
59.323 |
0.001 |
-1.19% |
0.001 |
|
MicroBenchmarks/LCALS/SubsetARawLoops/lcalsARaw.test:BM_DEL_DOT_VEC_2D_RAW/2
|
-4.44% |
2.455 |
2.346 |
0.000 |
-1.38% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4BigLoopTC8
|
-4.40% |
73.808 |
70.560 |
0.002 |
-4.40% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/10
|
-4.35% |
49.317 |
47.172 |
0.002 |
-1.50% |
0.002 |
|
MicroBenchmarks/LCALS/SubsetCLambdaLoops/lcalsCLambda.test:BM_MAT_X_MAT_LAMBDA/5001
|
-4.27% |
56090.083 |
53697.692 |
469.729 |
-2.99% |
469.729 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4BigLoopTC3
|
-4.18% |
28.608 |
27.411 |
0.007 |
-4.19% |
0.007 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4BigLoopTC8
|
-4.17% |
73.630 |
70.558 |
0.002 |
-4.00% |
0.002 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<3, EqZero, First>
|
-4.16% |
23421.042 |
22447.595 |
22.254 |
-0.00% |
22.254 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4BigLoopTC2
|
-4.14% |
20.873 |
20.008 |
0.001 |
-3.04% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4BigLoopTC16
|
-4.12% |
152.630 |
146.340 |
0.001 |
-4.12% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:benchForTruncOrZextVecWithAddInLoopFrom_uint16_t_To_uint32_t_
|
-4.12% |
21450.083 |
20566.970 |
1.464 |
-4.12% |
1.464 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4BigLoopTC16
|
-4.04% |
152.504 |
146.343 |
0.001 |
-4.06% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4BigLoopTC2
|
-4.04% |
20.836 |
19.995 |
0.006 |
-4.02% |
0.006 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_single_csa_cond_load_novec_int32_t_
|
-4.03% |
548817.467 |
526715.679 |
1336.902 |
-5.55% |
1336.902 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4BigLoopTC32
|
-3.98% |
294.660 |
282.928 |
0.006 |
-3.39% |
0.006 |
|
MultiSource/Applications/JM/lencod/lencod
Profile
|
-3.97% |
14.393 |
13.822 |
0.035 |
-4.46% |
0.035 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint64_t>/256
|
-3.87% |
412.669 |
396.680 |
0.348 |
-1.77% |
0.348 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/10
|
-3.78% |
37.882 |
36.451 |
0.001 |
-0.00% |
0.001 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<3, LessThanZero, Mid>
|
-3.71% |
26350.964 |
25374.148 |
22.418 |
-3.70% |
22.418 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<2, LessThanZero, Last>
|
-3.70% |
39530.521 |
38067.055 |
33.547 |
-3.72% |
33.547 |
|
MultiSource/Benchmarks/TSVC/CrossingThresholds-dbl/CrossingThresholds-dbl
Profile
|
-3.65% |
9.866 |
9.506 |
0.067 |
-3.35% |
0.067 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint32_t>/256
|
-3.58% |
413.640 |
398.818 |
0.011 |
-4.94% |
0.011 |
|
MicroBenchmarks/LCALS/SubsetALambdaLoops/lcalsALambda.test:BM_VOL3D_CALC_LAMBDA/2
|
-3.53% |
7.764 |
7.490 |
0.000 |
-3.55% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4BigLoopTC32
|
-3.49% |
293.092 |
282.848 |
0.003 |
-3.44% |
0.003 |
|
MicroBenchmarks/LCALS/SubsetARawLoops/lcalsARaw.test:BM_VOL3D_CALC_RAW/2
|
-3.46% |
7.764 |
7.496 |
0.000 |
-3.46% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4LoopTC8
|
-3.45% |
20.726 |
20.012 |
0.000 |
7.68% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/10
|
-3.39% |
42.168 |
40.739 |
0.001 |
-3.39% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4BigLoopTC32
|
-3.36% |
292.698 |
282.850 |
0.004 |
-3.37% |
0.004 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<2, LessThanZero, None>
|
-3.33% |
43920.316 |
42458.179 |
32.862 |
-6.45% |
32.862 |
|
MicroBenchmarks/LCALS/SubsetARawLoops/lcalsARaw.test:BM_DEL_DOT_VEC_2D_RAW/1
|
-3.27% |
195.549 |
189.152 |
0.615 |
-3.37% |
0.615 |
|
MultiSource/Benchmarks/TSVC/Expansion-dbl/Expansion-dbl
Profile
|
-3.27% |
10.830 |
10.476 |
0.063 |
-1.60% |
0.063 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint32_t>/16
|
-3.23% |
22.157 |
21.441 |
0.001 |
-0.01% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint32_t>/10
|
-3.23% |
22.157 |
21.442 |
0.000 |
0.00% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint32_t>/256
|
-3.23% |
22.157 |
21.442 |
0.001 |
-0.00% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint32_t>/51
|
-3.23% |
22.156 |
21.441 |
0.001 |
0.00% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint32_t>/28
|
-3.23% |
22.157 |
21.442 |
0.001 |
-0.01% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint32_t>/10
|
-3.23% |
22.157 |
21.442 |
0.000 |
-0.00% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint32_t>/999
|
-3.23% |
22.156 |
21.441 |
0.001 |
-0.00% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint32_t>/999
|
-3.23% |
22.156 |
21.442 |
0.001 |
-0.00% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint32_t>/51
|
-3.23% |
22.156 |
21.441 |
0.001 |
-0.01% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint32_t>/256
|
-3.23% |
22.156 |
21.441 |
0.001 |
-0.01% |
0.001 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<3, LessThanZero, Last>
|
-3.22% |
30251.901 |
29276.495 |
27.411 |
-3.22% |
27.411 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<5, GreaterThanZero, Mid>
|
-3.22% |
18153.028 |
17567.886 |
17.607 |
-0.00% |
17.607 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint32_t>/16
|
-3.22% |
22.156 |
21.442 |
0.000 |
0.00% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint32_t>/28
|
-3.22% |
22.156 |
21.443 |
0.000 |
-0.00% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint8_t>/256
|
-3.15% |
233.197 |
225.856 |
0.990 |
-4.27% |
0.990 |
|
SingleSource/Benchmarks/Misc-C++/oopack_v1p8
Profile
|
-3.14% |
0.337 |
0.327 |
0.001 |
-2.76% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4BigLoopTC3
|
-3.14% |
28.068 |
27.187 |
0.112 |
-4.44% |
0.112 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4BigLoopTC8
|
-3.14% |
72.867 |
70.581 |
0.004 |
-4.14% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/10
|
-3.13% |
22.872 |
22.156 |
0.001 |
-3.13% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/51
|
-3.13% |
22.872 |
22.156 |
0.004 |
-3.13% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/256
|
-3.13% |
22.872 |
22.156 |
0.000 |
-0.00% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint8_t>/999
|
-3.13% |
22.872 |
22.156 |
0.001 |
-3.13% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/28
|
-3.13% |
22.871 |
22.156 |
0.001 |
-3.13% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint32_t>/256
|
-3.13% |
228.721 |
221.567 |
0.005 |
-4.47% |
0.005 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint8_t>/256
|
-3.13% |
22.871 |
22.156 |
0.001 |
-0.01% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/10
|
-3.13% |
22.871 |
22.156 |
0.001 |
-0.00% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint8_t>/256
|
-3.13% |
22.871 |
22.156 |
0.000 |
-3.12% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint8_t>/10
|
-3.13% |
22.871 |
22.156 |
0.000 |
-3.13% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint8_t>/16
|
-3.13% |
22.871 |
22.156 |
0.001 |
-0.00% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/16
|
-3.13% |
22.871 |
22.156 |
0.003 |
-3.13% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint8_t>/999
|
-3.13% |
22.871 |
22.156 |
0.001 |
-0.01% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint8_t>/51
|
-3.13% |
22.871 |
22.156 |
0.000 |
-3.12% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/51
|
-3.13% |
22.872 |
22.157 |
0.000 |
0.00% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/999
|
-3.13% |
22.872 |
22.157 |
0.000 |
-0.00% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/256
|
-3.13% |
22.871 |
22.156 |
0.001 |
-3.13% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint8_t>/10
|
-3.13% |
22.871 |
22.156 |
0.000 |
0.00% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint8_t>/28
|
-3.13% |
22.870 |
22.155 |
0.001 |
-0.01% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/999
|
-3.12% |
22.871 |
22.157 |
0.001 |
-3.13% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/28
|
-3.12% |
22.871 |
22.157 |
0.001 |
0.00% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint8_t>/16
|
-3.12% |
22.871 |
22.156 |
0.000 |
-3.13% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint8_t>/28
|
-3.12% |
22.871 |
22.156 |
0.001 |
-3.12% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint8_t>/51
|
-3.12% |
22.870 |
22.156 |
0.001 |
-0.01% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/16
|
-3.12% |
22.871 |
22.157 |
0.001 |
-0.00% |
0.001 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<3, LessThanZero, None>
|
-3.12% |
31225.563 |
30251.491 |
22.593 |
-3.12% |
22.593 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<3, GreaterThanZero, None>
|
-3.12% |
31225.687 |
30251.826 |
22.251 |
0.00% |
22.251 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint32_t>/51
|
-3.01% |
95.059 |
92.199 |
0.023 |
-1.53% |
0.023 |
|
External/SPEC/CINT2017rate/505.mcf_r/505.mcf_r
Profile
|
-2.97% |
158.001 |
153.306 |
0.966 |
-1.92% |
0.966 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/16
|
-2.94% |
24.300 |
23.584 |
0.001 |
-0.01% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/256
|
-2.94% |
24.301 |
23.586 |
0.001 |
-0.00% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/10
|
-2.94% |
24.300 |
23.585 |
0.001 |
-0.01% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/51
|
-2.94% |
24.300 |
23.586 |
0.000 |
-0.00% |
0.000 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<7, GreaterThanZero, Mid>
|
-2.94% |
14224.042 |
13805.996 |
10.491 |
3.11% |
10.491 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<4, LessThanZero, None>
|
-2.94% |
24891.437 |
24159.936 |
21.812 |
-2.93% |
21.812 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/999
|
-2.94% |
24.300 |
23.586 |
0.000 |
0.00% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/28
|
-2.94% |
24.300 |
23.586 |
0.001 |
-0.00% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC128
|
-2.93% |
160.134 |
155.448 |
0.015 |
-19.45% |
0.015 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<8, GreaterThanZero, Mid>
|
-2.86% |
12817.175 |
12450.708 |
8.162 |
-0.00% |
8.162 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<32, EqZero, Mid>
|
-2.81% |
3209.914 |
3119.759 |
0.647 |
-0.00% |
0.647 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<32, GreaterThanZero, Mid>
|
-2.78% |
3209.120 |
3119.860 |
0.639 |
-2.82% |
0.639 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint64_t>/256
|
-2.78% |
224.222 |
217.994 |
0.003 |
-1.30% |
0.003 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<5, LessThanZero, None>
|
-2.78% |
21081.341 |
20495.842 |
18.643 |
-2.77% |
18.643 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<5, LessThanZero, Last>
|
-2.77% |
21080.834 |
20496.194 |
9.442 |
-0.00% |
9.442 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<5, GreaterThanZero, None>
|
-2.77% |
21079.551 |
20496.018 |
18.702 |
-2.77% |
18.702 |
|
SingleSource/Benchmarks/Misc/revertBits
Profile
|
-2.75% |
0.411 |
0.400 |
0.000 |
-2.73% |
0.000 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<6, GreaterThanZero, Last>
|
-2.70% |
18045.062 |
17558.005 |
10.920 |
-0.00% |
10.920 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4BigLoopTC15
|
-2.70% |
140.301 |
136.516 |
0.005 |
-4.43% |
0.005 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<6, LessThanZero, None>
|
-2.62% |
18529.639 |
18043.950 |
14.727 |
-2.63% |
14.727 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/16
|
-2.60% |
55.035 |
53.604 |
0.001 |
-2.60% |
0.001 |
|
MultiSource/Applications/aha/aha
Profile
|
-2.58% |
3.874 |
3.774 |
0.001 |
-3.05% |
0.001 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<7, EqZero, None>
|
-2.50% |
16732.264 |
16314.529 |
13.483 |
0.00% |
13.483 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4BigLoopTC2
|
-2.47% |
20.509 |
20.003 |
0.018 |
-3.00% |
0.018 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<7, LessThanZero, None>
|
-2.44% |
17150.534 |
16731.834 |
10.317 |
-0.00% |
10.317 |
|
MultiSource/Benchmarks/mafft/pairlocalalign
Profile
|
-2.41% |
47.339 |
46.199 |
0.011 |
-2.74% |
0.011 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<8, LessThanZero, None>
|
-2.38% |
15377.688 |
15012.417 |
8.098 |
-2.38% |
8.098 |
|
MicroBenchmarks/ImageProcessing/Blur/blur.test:BENCHMARK_GAUSSIAN_BLUR/1024
|
-2.36% |
261643.333 |
255458.667 |
145.994 |
-2.37% |
145.994 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint64_t>/10
|
-2.35% |
29.275 |
28.588 |
0.001 |
-2.45% |
0.001 |
|
MultiSource/Benchmarks/MiBench/telecomm-CRC32/telecomm-CRC32
Profile
|
-2.32% |
0.592 |
0.578 |
0.002 |
-3.46% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint8_t>/51
|
-2.26% |
95.058 |
92.914 |
0.002 |
0.00% |
0.002 |
|
MicroBenchmarks/LCALS/SubsetARawLoops/lcalsARaw.test:BM_VOL3D_CALC_RAW/1
|
-2.25% |
209.626 |
204.915 |
0.457 |
-2.06% |
0.457 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint8_t>/28
|
-2.24% |
95.775 |
93.628 |
0.001 |
-1.50% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint8_t>/16
|
-2.20% |
65.040 |
63.610 |
0.001 |
0.00% |
0.001 |
|
MicroBenchmarks/LCALS/SubsetALambdaLoops/lcalsALambda.test:BM_VOL3D_CALC_LAMBDA/1
|
-2.15% |
210.109 |
205.592 |
0.711 |
-1.49% |
0.711 |
|
SingleSource/Benchmarks/Adobe-C++/functionobjects
Profile
|
-2.09% |
6.063 |
5.937 |
0.003 |
-2.10% |
0.003 |
|
SingleSource/Benchmarks/Stanford/FloatMM
Profile
|
-2.03% |
0.769 |
0.753 |
0.001 |
-2.00% |
0.001 |
|
MicroBenchmarks/LCALS/SubsetALambdaLoops/lcalsALambda.test:BM_DEL_DOT_VEC_2D_LAMBDA/0
|
-1.98% |
1223.677 |
1199.466 |
1.328 |
-1.80% |
1.328 |
|
MultiSource/Benchmarks/DOE-ProxyApps-C++/miniFE/miniFE
Profile
|
-1.89% |
22.880 |
22.448 |
0.125 |
-1.72% |
0.125 |
|
MicroBenchmarks/LCALS/SubsetCRawLoops/lcalsCRaw.test:BM_BAND_LIN_EQ_RAW/44217
|
-1.87% |
86.191 |
84.576 |
0.090 |
-2.64% |
0.090 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint32_t>/16
|
-1.85% |
38.596 |
37.880 |
0.001 |
-0.00% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC128
|
-1.79% |
160.815 |
157.932 |
0.008 |
1.59% |
0.008 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/28
|
-1.77% |
80.766 |
79.334 |
0.003 |
-1.77% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_asinf_autovec_float_
|
-1.76% |
392.172 |
385.279 |
0.063 |
-0.02% |
0.063 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<31, LessThanZero, None>
|
-1.70% |
5252.240 |
5163.141 |
7.206 |
-1.96% |
7.206 |
|
MicroBenchmarks/LCALS/SubsetCLambdaLoops/lcalsCLambda.test:BM_PIC_2D_LAMBDA/171
|
-1.67% |
5.138 |
5.052 |
0.011 |
-1.32% |
0.011 |
|
MicroBenchmarks/LCALS/SubsetCLambdaLoops/lcalsCLambda.test:BM_MAT_X_MAT_LAMBDA/44217
|
-1.66% |
1040017.000 |
1022719.000 |
6279.524 |
-2.09% |
6279.524 |
|
SingleSource/Benchmarks/Polybench/medley/deriche/deriche
Profile
|
-1.63% |
2.926 |
2.879 |
0.014 |
-0.53% |
0.014 |
|
MicroBenchmarks/LCALS/SubsetCRawLoops/lcalsCRaw.test:BM_FIND_FIRST_MIN_RAW/44217
|
-1.55% |
131.352 |
129.317 |
0.152 |
1.42% |
0.152 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_csa_with_in_loop_arith_novec_uint8_t_
|
-1.50% |
522074.102 |
514250.184 |
1227.521 |
0.11% |
1227.521 |
|
MicroBenchmarks/MemFunctions/MemFunctions.test:BM_MemCmp<31, GreaterThanZero, None>
|
-1.49% |
5253.515 |
5175.489 |
8.306 |
-0.18% |
8.306 |
|
MultiSource/Benchmarks/nbench/nbench
Profile
|
-1.46% |
3.804 |
3.748 |
0.002 |
-0.73% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:benchForTruncOrZextVecWithAddInLoopWithVW8From_uint16_t_To_uint64_t_
|
-1.45% |
22198.642 |
21877.851 |
8.561 |
-1.31% |
8.561 |
|
MicroBenchmarks/LCALS/SubsetARawLoops/lcalsARaw.test:BM_DEL_DOT_VEC_2D_RAW/0
|
-1.44% |
1215.755 |
1198.271 |
2.343 |
-1.34% |
2.343 |
|
MultiSource/Benchmarks/Ptrdist/anagram/anagram
Profile
|
-1.41% |
1.912 |
1.885 |
0.001 |
-1.33% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint32_t>/28
|
-1.39% |
51.461 |
50.747 |
0.001 |
-0.00% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint64_t>/999
|
-1.37% |
758.674 |
748.311 |
0.018 |
-0.27% |
0.018 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC15
|
-1.35% |
17.630 |
17.391 |
0.000 |
-1.35% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW1LoopWithReductionTC15
|
-1.35% |
17.629 |
17.391 |
0.001 |
-1.35% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC15
|
-1.35% |
17.630 |
17.392 |
0.000 |
-34.23% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint8_t>/10
|
-1.35% |
47.816 |
47.171 |
0.001 |
-2.95% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC15
|
-1.35% |
17.629 |
17.392 |
0.000 |
-1.36% |
0.000 |
|
MicroBenchmarks/LCALS/SubsetCLambdaLoops/lcalsCLambda.test:BM_INT_PREDICT_LAMBDA/171
|
-1.30% |
2.092 |
2.064 |
0.004 |
-1.31% |
0.004 |
|
SingleSource/Benchmarks/Misc-C++/Large/ray
Profile
|
-1.28% |
7.625 |
7.527 |
0.006 |
-1.90% |
0.006 |
|
SingleSource/Benchmarks/BenchmarkGame/recursive
Profile
|
-1.28% |
2.337 |
2.307 |
0.009 |
-1.01% |
0.009 |
|
MicroBenchmarks/LCALS/SubsetCRawLoops/lcalsCRaw.test:BM_INT_PREDICT_RAW/171
|
-1.17% |
2.092 |
2.067 |
0.008 |
-1.18% |
0.008 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_only_novec_int32_t_
|
-1.16% |
1426859.470 |
1410241.449 |
1941.763 |
-0.41% |
1941.763 |
|
MicroBenchmarks/LCALS/SubsetCRawLoops/lcalsCRaw.test:BM_EOS_RAW/44217
|
-1.15% |
401.127 |
396.495 |
1.001 |
-0.64% |
1.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_sinf_novec_float_
|
-1.11% |
606.545 |
599.811 |
0.264 |
0.06% |
0.264 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/51
|
-1.10% |
130.083 |
128.650 |
0.003 |
-1.11% |
0.003 |
|
MicroBenchmarks/Builtins/Int128/Builtins.test:BM_DivideIntrinsic128UniformDivisor<__uint128_t>
|
-1.09% |
37.807 |
37.394 |
0.037 |
0.07% |
0.037 |
|
MicroBenchmarks/ImageProcessing/Interpolation/Interpolation.test:BENCHMARK_BICUBIC_INTERPOLATION/32
|
-1.08% |
599.456 |
592.997 |
0.193 |
-1.09% |
0.193 |
|
External/SPEC/CFP2017rate/544.nab_r/544.nab_r
Profile
|
-1.04% |
279.872 |
276.963 |
0.170 |
-1.11% |
0.170 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint8_t>/999
|
-1.03% |
764.060 |
756.196 |
0.917 |
-1.33% |
0.917 |
|
External/SPEC/CINT2017rate/541.leela_r/541.leela_r
Profile
|
-1.02% |
170.162 |
168.426 |
0.243 |
-1.23% |
0.243 |
|
MicroBenchmarks/ImageProcessing/AnisotropicDiffusion/AnisotropicDiffusion.test:BENCHMARK_ANISTROPIC_DIFFUSION/64
|
-1.02% |
4209.735 |
4166.804 |
4.508 |
-0.44% |
4.508 |
|
MicroBenchmarks/ImageProcessing/Interpolation/Interpolation.test:BENCHMARK_BICUBIC_INTERPOLATION/64
|
-1.01% |
2649.894 |
2623.101 |
0.213 |
-1.02% |
0.213 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint32_t>/999
|
-1.01% |
1475.776 |
1460.926 |
0.032 |
-1.40% |
0.032 |