|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_autovec_int64_t_
|
-27.44% |
1043411.064 |
757134.792 |
409.904 |
-27.54% |
409.904 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_novec_int64_t_
|
-27.43% |
1044293.094 |
757895.086 |
267.298 |
-27.44% |
267.298 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_novec_int32_t_
|
-26.89% |
515697.407 |
377026.536 |
401.695 |
-26.91% |
401.695 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_autovec_int32_t_
|
-26.76% |
515642.851 |
377636.305 |
664.916 |
-26.78% |
664.916 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint64_t>/10
|
-26.63% |
38.569 |
28.298 |
0.007 |
-26.76% |
0.007 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint64_t>/16
|
-25.83% |
53.259 |
39.502 |
0.104 |
-25.77% |
0.104 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_autovec_uint8_t_
|
-25.37% |
124983.119 |
93276.605 |
354.054 |
-25.33% |
354.054 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_novec_uint8_t_
|
-25.37% |
124992.432 |
93285.607 |
351.860 |
-25.32% |
351.860 |
|
SingleSource/Benchmarks/Shootout/Shootout-ary3
Profile
|
-23.55% |
0.984 |
0.752 |
0.001 |
-23.61% |
0.001 |
|
MultiSource/Benchmarks/Ptrdist/yacr2/yacr2
Profile
|
-23.01% |
2.836 |
2.184 |
0.000 |
-22.99% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint64_t>/16
|
-20.87% |
32.546 |
25.753 |
0.018 |
-20.87% |
0.018 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint16_t>/16
|
-19.91% |
9.385 |
7.517 |
0.014 |
-19.91% |
0.014 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint16_t>/10
|
-19.86% |
9.385 |
7.521 |
0.003 |
-19.86% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint8_t>/16
|
-19.52% |
60.718 |
48.868 |
0.014 |
-18.68% |
0.014 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint64_t>/16
|
-17.95% |
10.017 |
8.218 |
0.035 |
-17.98% |
0.035 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint8_t>/10
|
-17.94% |
41.296 |
33.886 |
0.008 |
-17.96% |
0.008 |
|
MultiSource/Benchmarks/Olden/em3d/em3d
Profile
|
-17.93% |
18.303 |
15.021 |
0.009 |
-17.94% |
0.009 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint64_t>/10
|
-17.57% |
10.017 |
8.257 |
0.029 |
-17.53% |
0.029 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint64_t>/10
|
-17.32% |
24.391 |
20.167 |
0.025 |
-17.26% |
0.025 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/28
|
-17.07% |
104.493 |
86.658 |
0.000 |
-17.07% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/16
|
-15.46% |
66.949 |
56.602 |
0.012 |
-15.46% |
0.012 |
|
MultiSource/Benchmarks/SciMark2-C/scimark2
Profile
|
-15.08% |
166.580 |
141.463 |
0.023 |
-15.13% |
0.023 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint16_t>/10
|
-13.97% |
35.652 |
30.673 |
0.009 |
-13.97% |
0.009 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/10
|
-13.65% |
48.179 |
41.605 |
0.002 |
-13.65% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint8_t>/16
|
-10.11% |
33.956 |
30.523 |
0.098 |
-10.21% |
0.098 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_only_novec_int64_t_
|
-9.84% |
1044938.586 |
942079.012 |
1104.741 |
-9.90% |
1104.741 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_only_autovec_int64_t_
|
-9.79% |
1044350.493 |
942102.924 |
1011.915 |
-9.95% |
1011.915 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint16_t>/16
|
-8.62% |
50.004 |
45.695 |
0.004 |
-8.52% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_only_autovec_int32_t_
|
-8.58% |
515233.743 |
471024.233 |
237.908 |
-8.60% |
237.908 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_only_novec_int32_t_
|
-8.56% |
515047.599 |
470955.105 |
453.341 |
-8.58% |
453.341 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint8_t>/28
|
-8.50% |
157.963 |
144.538 |
0.563 |
-8.51% |
0.563 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/256
|
-8.39% |
1207.602 |
1106.224 |
0.021 |
-8.40% |
0.021 |
|
MultiSource/Benchmarks/ASC_Sequoia/AMGmk/AMGmk
Profile
|
-8.39% |
43.481 |
39.833 |
0.013 |
-8.34% |
0.013 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint8_t>/51
|
-8.38% |
272.495 |
249.647 |
0.588 |
-8.38% |
0.588 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/999
|
-8.32% |
2364.872 |
2167.998 |
0.058 |
-8.33% |
0.058 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint8_t>/256
|
-8.24% |
1245.758 |
1143.127 |
0.012 |
-8.24% |
0.012 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint8_t>/999
|
-8.23% |
4804.772 |
4409.316 |
0.594 |
-8.24% |
0.594 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/256
|
-8.23% |
642.912 |
590.032 |
0.009 |
-8.22% |
0.009 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/999
|
-8.17% |
4636.134 |
4257.316 |
0.110 |
-8.39% |
0.110 |
|
SingleSource/Benchmarks/Adobe-C++/simple_types_constant_folding
Profile
|
-8.12% |
2.366 |
2.174 |
0.001 |
-8.15% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/51
|
-8.11% |
266.234 |
244.651 |
0.004 |
-8.11% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint8_t>/999
|
-8.08% |
2443.100 |
2245.724 |
0.651 |
-8.08% |
0.651 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint8_t>/256
|
-8.07% |
662.266 |
608.820 |
0.590 |
-8.08% |
0.590 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/28
|
-7.85% |
155.480 |
143.282 |
0.003 |
-7.85% |
0.003 |
|
MultiSource/Benchmarks/DOE-ProxyApps-C++/HPCCG/HPCCG
Profile
|
-7.83% |
4.648 |
4.284 |
0.040 |
-7.91% |
0.040 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/16
|
-7.50% |
100.113 |
92.601 |
0.002 |
-7.51% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint8_t>/51
|
-7.39% |
172.961 |
160.178 |
0.004 |
-7.37% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/10
|
-7.28% |
81.654 |
75.709 |
0.000 |
-7.28% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/51
|
-7.21% |
169.256 |
157.053 |
0.009 |
-7.21% |
0.009 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/16
|
-6.66% |
89.159 |
83.217 |
0.001 |
-6.67% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint8_t>/28
|
-6.64% |
122.639 |
114.500 |
0.209 |
-6.64% |
0.209 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/256
|
-6.55% |
51.556 |
48.178 |
0.000 |
-6.66% |
0.000 |
|
MultiSource/Benchmarks/DOE-ProxyApps-C++/miniFE/miniFE
Profile
|
-6.54% |
17.779 |
16.616 |
0.031 |
-6.90% |
0.031 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/999
|
-6.53% |
51.545 |
48.178 |
0.020 |
-6.68% |
0.020 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/28
|
-6.50% |
51.523 |
48.176 |
0.118 |
-6.64% |
0.118 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/51
|
-6.49% |
51.524 |
48.180 |
0.007 |
-6.59% |
0.007 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/28
|
-6.25% |
120.136 |
112.627 |
0.002 |
-6.25% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/256
|
-6.06% |
20.022 |
18.808 |
0.464 |
-6.06% |
0.464 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint32_t>/999
|
-5.75% |
1453.295 |
1369.739 |
0.776 |
-5.78% |
0.776 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/10
|
-5.69% |
76.960 |
72.579 |
0.005 |
-5.69% |
0.005 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint64_t>/999
|
-5.64% |
5517.522 |
5206.159 |
0.373 |
-5.64% |
0.373 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint32_t>/256
|
-5.64% |
1087.508 |
1026.145 |
0.036 |
-5.65% |
0.036 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint32_t>/51
|
-5.57% |
241.836 |
228.376 |
0.008 |
-5.57% |
0.008 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint32_t>/256
|
-5.55% |
581.005 |
548.739 |
0.023 |
-5.55% |
0.023 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint8_t>/999
|
-5.55% |
2847.643 |
2689.643 |
0.057 |
-5.55% |
0.057 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint8_t>/256
|
-5.55% |
744.581 |
703.271 |
0.027 |
-5.55% |
0.027 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/256
|
-5.54% |
745.209 |
703.889 |
0.018 |
-5.55% |
0.018 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/256
|
-5.54% |
745.197 |
703.900 |
0.135 |
-5.55% |
0.135 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/999
|
-5.54% |
2848.360 |
2690.631 |
0.125 |
-5.54% |
0.125 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint8_t>/51
|
-5.52% |
170.191 |
160.803 |
0.004 |
-5.52% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint32_t>/28
|
-5.50% |
142.353 |
134.522 |
0.003 |
-5.50% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/51
|
-5.50% |
170.818 |
161.426 |
0.003 |
-5.50% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/51
|
-5.50% |
170.823 |
161.432 |
0.004 |
-5.50% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint16_t>/999
|
-5.49% |
1451.016 |
1371.351 |
0.524 |
-5.50% |
0.524 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint8_t>/256
|
-5.48% |
399.824 |
377.912 |
0.017 |
-5.48% |
0.017 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint32_t>/999
|
-5.48% |
4176.089 |
3947.228 |
0.116 |
-5.90% |
0.116 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/999
|
-5.48% |
2847.428 |
2691.409 |
0.287 |
-5.46% |
0.287 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint16_t>/256
|
-5.47% |
400.447 |
378.533 |
0.015 |
-5.47% |
0.015 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint32_t>/256
|
-5.47% |
400.456 |
378.564 |
0.003 |
-5.48% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint8_t>/999
|
-5.45% |
1450.598 |
1371.577 |
0.042 |
-5.45% |
0.042 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint64_t>/51
|
-5.44% |
310.359 |
293.463 |
0.019 |
-5.44% |
0.019 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint32_t>/16
|
-5.41% |
92.603 |
87.596 |
0.002 |
-5.41% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/51
|
-5.40% |
52.249 |
49.430 |
0.002 |
-5.39% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/999
|
-5.39% |
52.248 |
49.430 |
0.001 |
-5.39% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/28
|
-5.39% |
52.246 |
49.429 |
0.002 |
-5.39% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/10
|
-5.39% |
52.246 |
49.430 |
0.002 |
-5.39% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/256
|
-5.39% |
52.246 |
49.431 |
0.000 |
-5.39% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/16
|
-5.39% |
52.245 |
49.431 |
0.001 |
-5.39% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint32_t>/999
|
-5.36% |
2131.685 |
2017.451 |
0.107 |
-5.37% |
0.107 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/16
|
-5.36% |
16.894 |
15.989 |
0.126 |
-5.36% |
0.126 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint64_t>/999
|
-5.35% |
2803.602 |
2653.672 |
0.181 |
-5.38% |
0.181 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint64_t>/999
|
-5.32% |
3005.797 |
2845.956 |
0.487 |
-5.33% |
0.487 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint32_t>/51
|
-5.11% |
110.124 |
104.499 |
0.008 |
-5.12% |
0.008 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint8_t>/51
|
-5.09% |
110.749 |
105.117 |
0.000 |
-5.08% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint32_t>/51
|
-5.01% |
156.114 |
148.290 |
0.002 |
-5.01% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/999
|
-4.98% |
1528.757 |
1452.568 |
0.141 |
-5.01% |
0.141 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint64_t>/51
|
-4.92% |
178.324 |
169.555 |
0.003 |
-4.92% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint32_t>/16
|
-4.86% |
83.530 |
79.469 |
0.000 |
-4.86% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint64_t>/51
|
-4.78% |
196.475 |
187.092 |
0.002 |
-4.78% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint8_t>/51
|
-4.77% |
39.419 |
37.540 |
0.001 |
-4.77% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint8_t>/256
|
-4.77% |
39.419 |
37.540 |
0.001 |
-4.77% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint8_t>/999
|
-4.76% |
39.420 |
37.542 |
0.000 |
-4.76% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint32_t>/256
|
-4.69% |
40.045 |
38.166 |
0.001 |
-4.69% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint32_t>/999
|
-4.69% |
40.046 |
38.167 |
0.000 |
-4.69% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint16_t>/256
|
-4.69% |
40.045 |
38.166 |
0.004 |
-4.69% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint16_t>/999
|
-4.69% |
40.045 |
38.167 |
0.001 |
-4.69% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint16_t>/51
|
-4.69% |
40.045 |
38.167 |
0.001 |
-4.69% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint32_t>/51
|
-4.69% |
40.044 |
38.167 |
0.001 |
-4.69% |
0.001 |
|
MultiSource/Benchmarks/TSVC/IndirectAddressing-flt/IndirectAddressing-flt
Profile
|
-4.51% |
15.736 |
15.027 |
0.042 |
-4.58% |
0.042 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/28
|
-4.51% |
16.894 |
16.133 |
0.076 |
-4.51% |
0.076 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint32_t>/28
|
-4.45% |
112.631 |
107.624 |
0.002 |
-4.45% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/51
|
-4.41% |
113.877 |
108.858 |
0.002 |
-4.41% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/28
|
-4.35% |
50.371 |
48.180 |
0.001 |
-4.35% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/999
|
-4.35% |
50.368 |
48.177 |
0.001 |
-4.35% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/16
|
-4.35% |
50.369 |
48.179 |
0.001 |
-4.35% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/51
|
-4.35% |
50.368 |
48.178 |
0.000 |
-4.35% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/256
|
-4.35% |
50.368 |
48.179 |
0.001 |
-4.35% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4BigLoopTC15
|
-4.33% |
272.577 |
260.768 |
3.752 |
-4.60% |
3.752 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:benchForTruncOrZextVecInLoopFrom_uint64_t_To_uint32_t_
|
-4.28% |
18102.330 |
17328.309 |
185.903 |
-0.44% |
185.903 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/10
|
-4.23% |
16.894 |
16.179 |
0.085 |
-4.24% |
0.085 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4BigLoopTC15
|
-4.20% |
272.646 |
261.196 |
3.205 |
-4.11% |
3.205 |
|
MultiSource/Benchmarks/DOE-ProxyApps-C/miniAMR/miniAMR
Profile
|
-4.09% |
3.791 |
3.635 |
0.038 |
-1.63% |
0.038 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint16_t>/51
|
-4.00% |
109.499 |
105.117 |
0.001 |
-4.00% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4BigLoopTC15
|
-3.97% |
272.195 |
261.376 |
3.456 |
-3.88% |
3.456 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4BigLoopTC16
|
-3.94% |
270.325 |
259.672 |
3.139 |
-2.22% |
3.139 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4BigLoopTC16
|
-3.93% |
270.592 |
259.946 |
3.228 |
-2.12% |
3.228 |
|
SingleSource/Benchmarks/Misc-C++/oopack_v1p8
Profile
|
-3.90% |
0.479 |
0.460 |
0.000 |
-3.91% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4BigLoopTC16
|
-3.88% |
270.403 |
259.914 |
3.153 |
-2.21% |
3.153 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint16_t>/10
|
-3.83% |
22.182 |
21.332 |
0.013 |
-4.11% |
0.013 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/999
|
-3.76% |
40.045 |
38.538 |
0.107 |
-3.77% |
0.107 |
|
MultiSource/Applications/JM/lencod/lencod
Profile
|
-3.67% |
35.653 |
34.345 |
0.031 |
-3.92% |
0.031 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/256
|
-3.60% |
359.343 |
346.394 |
0.041 |
-2.51% |
0.041 |
|
MultiSource/Benchmarks/TSVC/IndirectAddressing-dbl/IndirectAddressing-dbl
Profile
|
-3.47% |
18.870 |
18.216 |
0.132 |
-1.65% |
0.132 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/51
|
-3.44% |
40.044 |
38.665 |
0.048 |
-3.45% |
0.048 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint64_t>/51
|
-3.30% |
56.940 |
55.061 |
0.002 |
-3.30% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint64_t>/999
|
-3.30% |
56.939 |
55.061 |
0.006 |
-3.30% |
0.006 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4BigLoopTC7
|
-3.11% |
161.337 |
156.317 |
1.594 |
-2.90% |
1.594 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4BigLoopTC7
|
-3.11% |
161.272 |
156.262 |
1.669 |
-2.74% |
1.669 |
|
External/SPEC/CINT2017rate/525.x264_r/525.x264_r
Profile
|
-3.08% |
142.431 |
138.047 |
0.135 |
-2.90% |
0.135 |
|
MultiSource/Benchmarks/MiBench/consumer-lame/consumer-lame
Profile
|
-3.06% |
0.681 |
0.661 |
0.001 |
-3.13% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4BigLoopTC7
|
-3.03% |
161.155 |
156.265 |
1.560 |
-3.13% |
1.560 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC15
|
-2.94% |
21.273 |
20.648 |
0.001 |
-2.94% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC16
|
-2.92% |
21.269 |
20.648 |
0.001 |
-2.67% |
0.001 |
|
MultiSource/Benchmarks/TSVC/LoopRerolling-flt/LoopRerolling-flt
Profile
|
-2.90% |
5.409 |
5.253 |
0.042 |
0.04% |
0.042 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint64_t>/10
|
-2.86% |
40.676 |
39.514 |
0.017 |
-2.85% |
0.017 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/10
|
-2.76% |
41.297 |
40.158 |
0.008 |
-2.76% |
0.008 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint16_t>/16
|
-2.47% |
29.537 |
28.807 |
0.010 |
-2.76% |
0.010 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/10
|
-2.31% |
30.641 |
29.932 |
0.023 |
-2.38% |
0.023 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC31
|
-2.18% |
28.783 |
28.156 |
0.001 |
-2.18% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC31
|
-2.18% |
28.784 |
28.157 |
0.001 |
-2.17% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC31
|
-2.17% |
28.782 |
28.157 |
0.001 |
-2.17% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW1LoopWithReductionTC32
|
-2.17% |
28.782 |
28.158 |
0.000 |
-2.17% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW1LoopWithReductionTC31
|
-2.17% |
28.780 |
28.157 |
0.001 |
-2.16% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4LoopWithReductionTC32
|
-2.17% |
28.780 |
28.157 |
0.000 |
-2.17% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForLoopWithReductionAutoVecTC31
|
-2.16% |
28.779 |
28.156 |
0.000 |
-2.16% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4LoopWithReductionTC31
|
-2.16% |
28.779 |
28.156 |
0.001 |
-2.16% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC31
|
-2.16% |
28.779 |
28.157 |
0.001 |
-2.15% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC32
|
-2.16% |
28.779 |
28.157 |
0.000 |
-2.13% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC32
|
-2.16% |
28.778 |
28.158 |
0.002 |
-2.15% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint64_t>/16
|
-2.13% |
55.690 |
54.505 |
0.016 |
-2.14% |
0.016 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC32
|
-2.11% |
28.762 |
28.157 |
0.001 |
-2.12% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/16
|
-2.08% |
56.313 |
55.143 |
0.030 |
-2.08% |
0.030 |
|
MultiSource/Benchmarks/TSVC/ControlFlow-dbl/ControlFlow-dbl
Profile
|
-2.00% |
21.873 |
21.437 |
0.161 |
0.84% |
0.161 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC15
|
-2.00% |
21.069 |
20.649 |
0.002 |
-2.05% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC15
|
-1.95% |
21.058 |
20.648 |
0.001 |
-1.36% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForLoopWithReductionAutoVecTC32
|
-1.92% |
28.707 |
28.158 |
0.000 |
-1.98% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC32
|
-1.85% |
28.686 |
28.157 |
0.001 |
-1.99% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/16
|
-1.79% |
38.103 |
37.422 |
0.020 |
-1.62% |
0.020 |
|
SingleSource/Benchmarks/CoyoteBench/huffbench
Profile
|
-1.76% |
50.630 |
49.740 |
0.007 |
-1.76% |
0.007 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint8_t>/10
|
-1.71% |
24.552 |
24.131 |
0.078 |
-1.66% |
0.078 |
|
MultiSource/Applications/sgefa/sgefa
Profile
|
-1.68% |
0.640 |
0.630 |
0.003 |
-1.96% |
0.003 |
|
SingleSource/Benchmarks/Shootout/Shootout-matrix
Profile
|
-1.60% |
4.693 |
4.617 |
0.000 |
-1.61% |
0.000 |
|
External/SPEC/CINT2017rate/523.xalancbmk_r/523.xalancbmk_r
Profile
|
-1.58% |
212.761 |
209.399 |
0.166 |
-0.07% |
0.166 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint8_t>/10
|
-1.50% |
48.909 |
48.177 |
0.002 |
-1.49% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC64
|
-1.44% |
43.805 |
43.173 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW1LoopWithReductionTC64
|
-1.43% |
43.802 |
43.174 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC64
|
-1.43% |
43.801 |
43.173 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4LoopWithReductionTC63
|
-1.43% |
43.800 |
43.174 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC64
|
-1.43% |
43.799 |
43.173 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4LoopWithReductionTC64
|
-1.43% |
43.798 |
43.173 |
0.003 |
-1.43% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC63
|
-1.43% |
43.799 |
43.173 |
0.003 |
-1.43% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForLoopWithReductionAutoVecTC63
|
-1.43% |
43.799 |
43.174 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC63
|
-1.43% |
43.798 |
43.173 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForLoopWithReductionAutoVecTC64
|
-1.43% |
43.799 |
43.174 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW1LoopWithReductionTC63
|
-1.43% |
43.797 |
43.173 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC64
|
-1.42% |
43.798 |
43.174 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC63
|
-1.42% |
43.798 |
43.175 |
0.000 |
-1.42% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC63
|
-1.42% |
43.797 |
43.174 |
0.004 |
-1.43% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/28
|
-1.39% |
86.349 |
85.146 |
0.009 |
-1.39% |
0.009 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint64_t>/28
|
-1.39% |
85.731 |
84.538 |
0.000 |
-1.39% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/28
|
-1.31% |
53.143 |
52.446 |
0.020 |
-1.15% |
0.020 |
|
MicroBenchmarks/ImageProcessing/BilateralFiltering/BilateralFilter.test:BENCHMARK_BILATERAL_FILTER/64/2
|
-1.31% |
2571.072 |
2537.453 |
11.401 |
-0.55% |
11.401 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_csa_with_in_loop_arith_autovec_int64_t_
|
-1.21% |
728846.367 |
720031.170 |
2365.240 |
-0.47% |
2365.240 |
|
MultiSource/Applications/lua/lua
Profile
|
-1.20% |
70.925 |
70.072 |
0.003 |
-1.19% |
0.003 |
|
External/SPEC/CFP2017rate/511.povray_r/511.povray_r
Profile
|
-1.20% |
28.674 |
28.329 |
0.021 |
-1.07% |
0.021 |