|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_autovec_int64_t_
|
-27.75% |
1044897.046 |
754900.586 |
1686.469 |
-27.75% |
1686.469 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_novec_int64_t_
|
-27.72% |
1045514.943 |
755710.452 |
1041.629 |
-27.65% |
1041.629 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_novec_int32_t_
|
-27.05% |
515702.479 |
376224.319 |
849.371 |
-27.06% |
849.371 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_autovec_int32_t_
|
-26.88% |
515209.347 |
376708.422 |
457.642 |
-26.96% |
457.642 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint64_t>/10
|
-26.72% |
38.573 |
28.268 |
0.023 |
-26.84% |
0.023 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint64_t>/16
|
-25.83% |
53.253 |
39.500 |
0.012 |
-25.77% |
0.012 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_novec_uint8_t_
|
-25.28% |
124978.824 |
93382.835 |
69.180 |
-25.24% |
69.180 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_autovec_uint8_t_
|
-25.27% |
124993.786 |
93412.290 |
46.384 |
-25.22% |
46.384 |
|
SingleSource/Benchmarks/Shootout/Shootout-ary3
Profile
|
-23.54% |
0.984 |
0.752 |
0.001 |
-23.59% |
0.001 |
|
MultiSource/Benchmarks/Ptrdist/yacr2/yacr2
Profile
|
-23.02% |
2.834 |
2.182 |
0.001 |
-23.05% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint64_t>/16
|
-20.76% |
32.541 |
25.786 |
0.108 |
-20.77% |
0.108 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint16_t>/10
|
-19.89% |
9.385 |
7.518 |
0.020 |
-19.89% |
0.020 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint16_t>/16
|
-19.75% |
9.385 |
7.531 |
0.007 |
-19.76% |
0.007 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint8_t>/10
|
-18.77% |
41.302 |
33.551 |
0.162 |
-18.77% |
0.162 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint8_t>/16
|
-18.68% |
60.090 |
48.864 |
0.018 |
-18.69% |
0.018 |
|
MultiSource/Benchmarks/Olden/em3d/em3d
Profile
|
-17.86% |
18.293 |
15.026 |
0.006 |
-17.91% |
0.006 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint64_t>/16
|
-17.78% |
10.014 |
8.233 |
0.022 |
-17.83% |
0.022 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint64_t>/10
|
-17.66% |
10.012 |
8.243 |
0.018 |
-17.66% |
0.018 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint64_t>/10
|
-17.18% |
24.344 |
20.161 |
0.085 |
-17.28% |
0.085 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/28
|
-17.10% |
104.488 |
86.616 |
0.024 |
-17.11% |
0.024 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/16
|
-15.43% |
66.955 |
56.623 |
0.003 |
-15.43% |
0.003 |
|
MultiSource/Benchmarks/SciMark2-C/scimark2
Profile
|
-15.06% |
166.533 |
141.458 |
0.052 |
-15.13% |
0.052 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint16_t>/10
|
-13.77% |
35.611 |
30.707 |
0.028 |
-13.87% |
0.028 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/10
|
-13.67% |
48.200 |
41.609 |
0.001 |
-13.64% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint8_t>/16
|
-10.48% |
33.940 |
30.383 |
0.459 |
-10.62% |
0.459 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_only_novec_int64_t_
|
-10.02% |
1044801.628 |
940067.923 |
2290.901 |
-10.10% |
2290.901 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_only_autovec_int64_t_
|
-9.79% |
1045700.646 |
943342.539 |
257.738 |
-9.84% |
257.738 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_only_autovec_int32_t_
|
-8.76% |
515569.166 |
470429.134 |
156.286 |
-8.71% |
156.286 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_only_novec_int32_t_
|
-8.71% |
515948.376 |
470990.780 |
392.365 |
-8.57% |
392.365 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint8_t>/28
|
-8.51% |
157.983 |
144.536 |
0.004 |
-8.51% |
0.004 |
|
MultiSource/Benchmarks/ASC_Sequoia/AMGmk/AMGmk
Profile
|
-8.49% |
43.492 |
39.799 |
0.035 |
-8.42% |
0.035 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint16_t>/16
|
-8.47% |
49.942 |
45.710 |
0.168 |
-8.49% |
0.168 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/256
|
-8.39% |
1207.575 |
1106.240 |
0.025 |
-8.40% |
0.025 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint8_t>/51
|
-8.38% |
272.466 |
249.645 |
0.009 |
-8.38% |
0.009 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/999
|
-8.32% |
2364.747 |
2168.117 |
0.113 |
-8.33% |
0.113 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/999
|
-8.31% |
4643.176 |
4257.366 |
0.342 |
-8.39% |
0.342 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint8_t>/256
|
-8.24% |
1245.749 |
1143.136 |
0.027 |
-8.24% |
0.027 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint8_t>/999
|
-8.22% |
4804.427 |
4409.298 |
0.294 |
-8.24% |
0.294 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/256
|
-8.22% |
642.916 |
590.046 |
0.028 |
-8.22% |
0.028 |
|
SingleSource/Benchmarks/Adobe-C++/simple_types_constant_folding
Profile
|
-8.11% |
2.367 |
2.175 |
0.001 |
-8.09% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/51
|
-8.11% |
266.229 |
244.641 |
0.009 |
-8.11% |
0.009 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint8_t>/999
|
-8.08% |
2443.121 |
2245.674 |
0.642 |
-8.09% |
0.642 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint8_t>/256
|
-8.07% |
662.263 |
608.805 |
0.084 |
-8.08% |
0.084 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/28
|
-7.85% |
155.490 |
143.287 |
0.007 |
-7.85% |
0.007 |
|
MultiSource/Benchmarks/DOE-ProxyApps-C++/HPCCG/HPCCG
Profile
|
-7.77% |
4.643 |
4.282 |
0.043 |
-7.95% |
0.043 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/16
|
-7.49% |
100.108 |
92.605 |
0.002 |
-7.50% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint8_t>/51
|
-7.40% |
172.983 |
160.179 |
0.005 |
-7.37% |
0.005 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/10
|
-7.28% |
81.651 |
75.708 |
0.003 |
-7.28% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/51
|
-7.21% |
169.250 |
157.052 |
0.013 |
-7.21% |
0.013 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/999
|
-6.70% |
51.636 |
48.177 |
0.002 |
-6.68% |
0.002 |
|
MultiSource/Benchmarks/DOE-ProxyApps-C++/miniFE/miniFE
Profile
|
-6.69% |
17.834 |
16.641 |
0.024 |
-6.77% |
0.024 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/256
|
-6.67% |
371.182 |
346.442 |
0.045 |
-2.49% |
0.045 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/16
|
-6.66% |
89.162 |
83.221 |
0.003 |
-6.66% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/28
|
-6.64% |
51.606 |
48.180 |
0.003 |
-6.63% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/256
|
-6.62% |
51.591 |
48.177 |
0.004 |
-6.66% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint8_t>/28
|
-6.62% |
122.630 |
114.517 |
0.022 |
-6.62% |
0.022 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/51
|
-6.53% |
51.543 |
48.178 |
0.007 |
-6.59% |
0.007 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/28
|
-6.25% |
120.136 |
112.624 |
0.003 |
-6.25% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint32_t>/999
|
-5.78% |
1453.698 |
1369.701 |
0.037 |
-5.78% |
0.037 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint64_t>/999
|
-5.69% |
5517.719 |
5203.516 |
1.267 |
-5.69% |
1.267 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/10
|
-5.69% |
76.961 |
72.582 |
0.004 |
-5.69% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint32_t>/256
|
-5.64% |
1087.470 |
1026.168 |
0.042 |
-5.64% |
0.042 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint32_t>/999
|
-5.60% |
4180.200 |
3945.922 |
0.829 |
-5.93% |
0.829 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint32_t>/51
|
-5.56% |
241.821 |
228.379 |
0.009 |
-5.56% |
0.009 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint8_t>/999
|
-5.55% |
2847.582 |
2689.448 |
0.248 |
-5.56% |
0.248 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint32_t>/256
|
-5.54% |
580.953 |
548.748 |
0.042 |
-5.55% |
0.042 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/256
|
-5.54% |
745.213 |
703.904 |
0.018 |
-5.54% |
0.018 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint8_t>/256
|
-5.54% |
744.576 |
703.311 |
0.006 |
-5.54% |
0.006 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/256
|
-5.54% |
745.207 |
703.929 |
0.024 |
-5.55% |
0.024 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/999
|
-5.54% |
2848.242 |
2690.557 |
0.251 |
-5.54% |
0.251 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint16_t>/999
|
-5.51% |
1450.891 |
1370.968 |
0.816 |
-5.52% |
0.816 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint8_t>/51
|
-5.50% |
170.185 |
160.817 |
0.003 |
-5.51% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/51
|
-5.49% |
170.814 |
161.431 |
0.006 |
-5.49% |
0.006 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint32_t>/28
|
-5.49% |
142.349 |
134.531 |
0.004 |
-5.49% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/51
|
-5.49% |
170.821 |
161.439 |
0.004 |
-5.49% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/999
|
-5.49% |
2846.816 |
2690.654 |
0.527 |
-5.49% |
0.527 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint8_t>/256
|
-5.48% |
399.823 |
377.921 |
0.013 |
-5.48% |
0.013 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint32_t>/256
|
-5.47% |
400.463 |
378.542 |
0.037 |
-5.48% |
0.037 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint16_t>/256
|
-5.47% |
400.434 |
378.542 |
0.032 |
-5.47% |
0.032 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint8_t>/999
|
-5.45% |
1450.584 |
1371.591 |
0.071 |
-5.45% |
0.071 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint64_t>/51
|
-5.44% |
310.349 |
293.456 |
0.009 |
-5.44% |
0.009 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint32_t>/16
|
-5.41% |
92.606 |
87.596 |
0.004 |
-5.41% |
0.004 |
|
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% |
52.245 |
49.430 |
0.002 |
-5.39% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/16
|
-5.39% |
52.245 |
49.430 |
0.002 |
-5.39% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/256
|
-5.39% |
52.243 |
49.429 |
0.002 |
-5.39% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/51
|
-5.39% |
52.244 |
49.429 |
0.001 |
-5.39% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/999
|
-5.39% |
52.246 |
49.431 |
0.001 |
-5.39% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/28
|
-5.38% |
16.894 |
15.985 |
0.091 |
-5.38% |
0.091 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint32_t>/999
|
-5.35% |
2131.446 |
2017.377 |
0.037 |
-5.37% |
0.037 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint64_t>/999
|
-5.35% |
2803.732 |
2653.764 |
0.183 |
-5.38% |
0.183 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint64_t>/999
|
-5.31% |
3005.970 |
2846.276 |
0.408 |
-5.32% |
0.408 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint32_t>/51
|
-5.12% |
110.138 |
104.497 |
0.002 |
-5.13% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint8_t>/51
|
-5.08% |
110.747 |
105.118 |
0.002 |
-5.08% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint32_t>/51
|
-5.01% |
156.112 |
148.289 |
0.006 |
-5.01% |
0.006 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/999
|
-4.98% |
1528.726 |
1452.593 |
0.057 |
-5.01% |
0.057 |
|
MultiSource/Benchmarks/TSVC/IndirectAddressing-flt/IndirectAddressing-flt
Profile
|
-4.98% |
15.733 |
14.950 |
0.058 |
-5.06% |
0.058 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint64_t>/51
|
-4.92% |
178.319 |
169.548 |
0.004 |
-4.92% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/16
|
-4.90% |
16.894 |
16.065 |
0.086 |
-4.91% |
0.086 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/10
|
-4.89% |
16.894 |
16.068 |
0.114 |
-4.89% |
0.114 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint32_t>/16
|
-4.88% |
83.537 |
79.464 |
0.003 |
-4.87% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint64_t>/51
|
-4.78% |
196.468 |
187.079 |
0.008 |
-4.78% |
0.008 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint8_t>/51
|
-4.76% |
39.418 |
37.541 |
0.003 |
-4.77% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint8_t>/256
|
-4.76% |
39.417 |
37.540 |
0.002 |
-4.77% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint8_t>/999
|
-4.76% |
39.418 |
37.542 |
0.002 |
-4.76% |
0.002 |
|
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% |
40.045 |
38.167 |
0.002 |
-4.69% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint16_t>/256
|
-4.69% |
40.044 |
38.166 |
0.004 |
-4.69% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint16_t>/999
|
-4.69% |
40.044 |
38.167 |
0.003 |
-4.69% |
0.003 |
|
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 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint16_t>/51
|
-4.68% |
40.045 |
38.169 |
0.001 |
-4.69% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4BigLoopTC7
|
-4.68% |
162.304 |
154.708 |
2.211 |
-3.90% |
2.211 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4BigLoopTC7
|
-4.65% |
162.289 |
154.749 |
2.190 |
-3.68% |
2.190 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4BigLoopTC7
|
-4.60% |
162.507 |
155.027 |
2.015 |
-3.90% |
2.015 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint32_t>/28
|
-4.44% |
112.625 |
107.621 |
0.004 |
-4.45% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/51
|
-4.41% |
113.877 |
108.854 |
0.003 |
-4.42% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/999
|
-4.35% |
50.369 |
48.178 |
0.002 |
-4.35% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/256
|
-4.35% |
50.368 |
48.178 |
0.002 |
-4.35% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/28
|
-4.35% |
50.368 |
48.178 |
0.004 |
-4.35% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/51
|
-4.35% |
50.368 |
48.178 |
0.002 |
-4.35% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/16
|
-4.34% |
50.368 |
48.181 |
0.000 |
-4.35% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint16_t>/51
|
-4.00% |
109.496 |
105.120 |
0.003 |
-4.00% |
0.003 |
|
SingleSource/Benchmarks/Misc-C++/oopack_v1p8
Profile
|
-3.92% |
0.478 |
0.460 |
0.000 |
-3.95% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint16_t>/10
|
-3.75% |
22.140 |
21.310 |
0.013 |
-4.21% |
0.013 |
|
MultiSource/Applications/JM/lencod/lencod
Profile
|
-3.70% |
35.696 |
34.376 |
0.035 |
-3.84% |
0.035 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint64_t>/51
|
-3.30% |
56.937 |
55.059 |
0.002 |
-3.30% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint64_t>/999
|
-3.30% |
56.939 |
55.062 |
0.002 |
-3.30% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/51
|
-3.21% |
40.044 |
38.758 |
0.008 |
-3.22% |
0.008 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/999
|
-3.20% |
40.044 |
38.762 |
0.003 |
-3.21% |
0.003 |
|
External/SPEC/CINT2017rate/525.x264_r/525.x264_r
Profile
|
-3.13% |
142.371 |
137.917 |
0.214 |
-3.00% |
0.214 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint64_t>/10
|
-3.01% |
40.681 |
39.458 |
0.017 |
-2.99% |
0.017 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC15
|
-2.94% |
21.274 |
20.648 |
0.000 |
-2.94% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC16
|
-2.88% |
21.259 |
20.648 |
0.000 |
-2.67% |
0.000 |
|
MultiSource/Benchmarks/MiBench/consumer-lame/consumer-lame
Profile
|
-2.82% |
0.681 |
0.661 |
0.003 |
-2.99% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/10
|
-2.76% |
41.297 |
40.157 |
0.004 |
-2.76% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint16_t>/16
|
-2.52% |
29.550 |
28.806 |
0.017 |
-2.76% |
0.017 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/10
|
-2.42% |
30.623 |
29.882 |
0.020 |
-2.54% |
0.020 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC15
|
-2.37% |
21.149 |
20.647 |
0.000 |
-1.36% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint8_t>/10
|
-2.26% |
24.589 |
24.032 |
0.145 |
-2.06% |
0.145 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW1LoopWithReductionTC32
|
-2.17% |
28.782 |
28.156 |
0.001 |
-2.18% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC31
|
-2.17% |
28.782 |
28.157 |
0.001 |
-2.17% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC31
|
-2.17% |
28.781 |
28.157 |
0.000 |
-2.17% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4LoopWithReductionTC32
|
-2.17% |
28.780 |
28.156 |
0.001 |
-2.18% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForLoopWithReductionAutoVecTC31
|
-2.16% |
28.780 |
28.157 |
0.001 |
-2.16% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC32
|
-2.16% |
28.778 |
28.156 |
0.001 |
-2.13% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC31
|
-2.16% |
28.778 |
28.157 |
0.000 |
-2.15% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC32
|
-2.16% |
28.778 |
28.157 |
0.001 |
-2.15% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW1LoopWithReductionTC31
|
-2.15% |
28.777 |
28.157 |
0.000 |
-2.16% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4LoopWithReductionTC31
|
-2.15% |
28.775 |
28.156 |
0.000 |
-2.16% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC32
|
-2.12% |
28.767 |
28.156 |
0.000 |
-2.14% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint64_t>/16
|
-2.12% |
55.693 |
54.514 |
0.007 |
-2.12% |
0.007 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForLoopWithReductionAutoVecTC32
|
-2.08% |
28.755 |
28.158 |
0.000 |
-1.98% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/16
|
-2.06% |
56.312 |
55.154 |
0.003 |
-2.06% |
0.003 |
|
MultiSource/Benchmarks/TSVC/IndirectAddressing-dbl/IndirectAddressing-dbl
Profile
|
-2.04% |
18.661 |
18.280 |
0.091 |
-1.31% |
0.091 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_csa_with_in_loop_arith_autovec_int32_t_
|
-1.95% |
377245.796 |
369877.637 |
2745.858 |
-1.63% |
2745.858 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC32
|
-1.89% |
28.699 |
28.157 |
0.001 |
-1.99% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/16
|
-1.84% |
38.098 |
37.395 |
0.022 |
-1.69% |
0.022 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC31
|
-1.82% |
28.680 |
28.157 |
0.000 |
-2.17% |
0.000 |
|
SingleSource/Benchmarks/CoyoteBench/huffbench
Profile
|
-1.76% |
50.630 |
49.741 |
0.002 |
-1.75% |
0.002 |
|
SingleSource/Benchmarks/Shootout/Shootout-matrix
Profile
|
-1.60% |
4.692 |
4.617 |
0.000 |
-1.61% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint8_t>/10
|
-1.58% |
48.947 |
48.174 |
0.004 |
-1.50% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4BigLoopTC8
|
-1.57% |
162.886 |
160.332 |
0.856 |
2.19% |
0.856 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4BigLoopTC8
|
-1.55% |
162.791 |
160.260 |
0.959 |
2.27% |
0.959 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC63
|
-1.43% |
43.799 |
43.171 |
0.001 |
-1.44% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForLoopWithReductionAutoVecTC63
|
-1.43% |
43.800 |
43.172 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4LoopWithReductionTC64
|
-1.43% |
43.799 |
43.172 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC63
|
-1.43% |
43.800 |
43.173 |
0.000 |
-1.43% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC64
|
-1.43% |
43.799 |
43.172 |
0.002 |
-1.43% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC64
|
-1.43% |
43.797 |
43.172 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC64
|
-1.43% |
43.799 |
43.174 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC63
|
-1.43% |
43.798 |
43.173 |
0.003 |
-1.43% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4LoopWithReductionTC63
|
-1.43% |
43.797 |
43.173 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW1LoopWithReductionTC64
|
-1.43% |
43.798 |
43.173 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC64
|
-1.42% |
43.798 |
43.174 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC63
|
-1.42% |
43.798 |
43.174 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW1LoopWithReductionTC63
|
-1.42% |
43.798 |
43.174 |
0.000 |
-1.43% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForLoopWithReductionAutoVecTC64
|
-1.42% |
43.799 |
43.176 |
0.001 |
-1.42% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint64_t>/28
|
-1.40% |
85.724 |
84.521 |
0.019 |
-1.41% |
0.019 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4BigLoopTC8
|
-1.39% |
162.771 |
160.502 |
0.740 |
2.36% |
0.740 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/28
|
-1.38% |
86.344 |
85.154 |
0.009 |
-1.38% |
0.009 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC16
|
-1.34% |
20.929 |
20.649 |
0.001 |
-1.09% |
0.001 |
|
MultiSource/Benchmarks/McCat/17-bintr/bintr
Profile
|
-1.32% |
1.219 |
1.203 |
0.001 |
0.39% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC15
|
-1.31% |
20.923 |
20.648 |
0.002 |
-2.06% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_csa_with_in_loop_arith_novec_int64_t_
|
-1.31% |
726903.051 |
717361.180 |
3613.057 |
-1.15% |
3613.057 |
|
MultiSource/Applications/lua/lua
Profile
|
-1.25% |
70.901 |
70.016 |
0.047 |
-1.27% |
0.047 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/28
|
-1.22% |
53.128 |
52.481 |
0.006 |
-1.09% |
0.006 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_single_csa_nested_cond_load_autovec_int64_t_
|
-1.21% |
617389.979 |
609908.662 |
2301.729 |
-1.24% |
2301.729 |
|
External/SPEC/CFP2017rate/511.povray_r/511.povray_r
Profile
|
-1.18% |
28.685 |
28.345 |
0.013 |
-1.01% |
0.013 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:benchForTruncOrZextVecInLoopWithVW16From_uint32_t_To_uint64_t_
|
-1.14% |
17601.479 |
17400.671 |
37.075 |
2.09% |
37.075 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_csa_with_in_loop_arith_autovec_int64_t_
|
-1.09% |
728286.809 |
720316.449 |
2385.093 |
-0.43% |
2385.093 |
|
MultiSource/Applications/lambda-0.1.3/lambda
Profile
|
-1.03% |
11.800 |
11.678 |
0.013 |
-1.10% |
0.013 |