|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_autovec_int64_t_
|
-27.64% |
1043287.122 |
754903.588 |
1740.846 |
-27.75% |
1740.846 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_novec_int64_t_
|
-27.54% |
1043701.604 |
756294.559 |
1691.627 |
-27.59% |
1691.627 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_novec_int32_t_
|
-26.96% |
515377.794 |
376415.948 |
810.605 |
-27.03% |
810.605 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint64_t>/10
|
-26.85% |
38.617 |
28.249 |
0.034 |
-26.89% |
0.034 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_autovec_int32_t_
|
-26.83% |
515260.484 |
377011.364 |
431.400 |
-26.90% |
431.400 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint64_t>/16
|
-25.77% |
53.202 |
39.494 |
0.023 |
-25.78% |
0.023 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_autovec_uint8_t_
|
-25.51% |
124927.250 |
93063.218 |
366.062 |
-25.50% |
366.062 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_novec_uint8_t_
|
-25.49% |
124922.847 |
93079.326 |
356.504 |
-25.49% |
356.504 |
|
SingleSource/Benchmarks/Shootout/Shootout-ary3
Profile
|
-23.38% |
0.983 |
0.753 |
0.001 |
-23.49% |
0.001 |
|
MultiSource/Benchmarks/Ptrdist/yacr2/yacr2
Profile
|
-23.09% |
2.838 |
2.183 |
0.000 |
-23.02% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint64_t>/16
|
-20.90% |
32.544 |
25.744 |
0.016 |
-20.90% |
0.016 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint16_t>/16
|
-19.91% |
9.386 |
7.517 |
0.012 |
-19.91% |
0.012 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint16_t>/10
|
-19.82% |
9.385 |
7.525 |
0.007 |
-19.82% |
0.007 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint8_t>/16
|
-19.66% |
60.079 |
48.267 |
0.293 |
-19.68% |
0.293 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint8_t>/10
|
-19.46% |
41.298 |
33.261 |
0.299 |
-19.47% |
0.299 |
|
MultiSource/Benchmarks/Olden/em3d/em3d
Profile
|
-17.83% |
18.292 |
15.030 |
0.003 |
-17.89% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint64_t>/10
|
-17.56% |
10.012 |
8.253 |
0.018 |
-17.56% |
0.018 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint64_t>/16
|
-17.48% |
10.015 |
8.264 |
0.023 |
-17.52% |
0.023 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint64_t>/10
|
-17.24% |
24.347 |
20.149 |
0.026 |
-17.33% |
0.026 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/28
|
-17.08% |
104.487 |
86.641 |
0.011 |
-17.09% |
0.011 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/16
|
-15.42% |
66.948 |
56.625 |
0.003 |
-15.43% |
0.003 |
|
MultiSource/Benchmarks/SciMark2-C/scimark2
Profile
|
-15.08% |
166.588 |
141.459 |
0.049 |
-15.13% |
0.049 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint16_t>/10
|
-13.81% |
35.589 |
30.672 |
0.008 |
-13.97% |
0.008 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/10
|
-13.64% |
48.180 |
41.608 |
0.001 |
-13.64% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint8_t>/16
|
-10.34% |
34.062 |
30.539 |
0.220 |
-10.16% |
0.220 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_only_novec_int64_t_
|
-9.83% |
1044635.274 |
941899.255 |
541.132 |
-9.92% |
541.132 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_only_autovec_int64_t_
|
-9.72% |
1044184.599 |
942693.337 |
538.508 |
-9.90% |
538.508 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint16_t>/16
|
-8.62% |
50.007 |
45.697 |
0.015 |
-8.51% |
0.015 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_only_novec_int32_t_
|
-8.56% |
515181.875 |
471059.089 |
218.226 |
-8.56% |
218.226 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_only_autovec_int32_t_
|
-8.54% |
515103.842 |
471099.601 |
305.598 |
-8.58% |
305.598 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint8_t>/28
|
-8.51% |
157.965 |
144.529 |
0.096 |
-8.51% |
0.096 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/256
|
-8.40% |
1207.592 |
1106.196 |
0.059 |
-8.40% |
0.059 |
|
MultiSource/Benchmarks/ASC_Sequoia/AMGmk/AMGmk
Profile
|
-8.39% |
43.491 |
39.842 |
0.007 |
-8.32% |
0.007 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/999
|
-8.38% |
4646.622 |
4257.268 |
0.137 |
-8.39% |
0.137 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint8_t>/51
|
-8.37% |
272.470 |
249.656 |
0.543 |
-8.38% |
0.543 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/999
|
-8.32% |
2364.940 |
2168.139 |
0.716 |
-8.33% |
0.716 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint8_t>/256
|
-8.24% |
1245.697 |
1143.097 |
0.114 |
-8.24% |
0.114 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/256
|
-8.22% |
642.912 |
590.036 |
0.005 |
-8.22% |
0.005 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint8_t>/999
|
-8.22% |
4804.998 |
4409.874 |
0.359 |
-8.22% |
0.359 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/51
|
-8.11% |
266.228 |
244.649 |
0.010 |
-8.11% |
0.010 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint8_t>/256
|
-8.09% |
662.348 |
608.792 |
0.581 |
-8.08% |
0.581 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint8_t>/999
|
-8.07% |
2443.023 |
2245.817 |
0.492 |
-8.08% |
0.492 |
|
SingleSource/Benchmarks/Adobe-C++/simple_types_constant_folding
Profile
|
-8.03% |
2.366 |
2.176 |
0.002 |
-8.06% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/28
|
-7.84% |
155.481 |
143.284 |
0.004 |
-7.85% |
0.004 |
|
MultiSource/Benchmarks/DOE-ProxyApps-C++/HPCCG/HPCCG
Profile
|
-7.75% |
4.650 |
4.290 |
0.048 |
-7.79% |
0.048 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/16
|
-7.50% |
100.111 |
92.604 |
0.002 |
-7.50% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint8_t>/51
|
-7.40% |
172.979 |
160.175 |
0.498 |
-7.37% |
0.498 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/10
|
-7.28% |
81.651 |
75.709 |
0.001 |
-7.28% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/51
|
-7.21% |
169.251 |
157.056 |
0.003 |
-7.21% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/16
|
-6.66% |
89.160 |
83.219 |
0.001 |
-6.67% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/51
|
-6.65% |
51.612 |
48.178 |
0.568 |
-6.59% |
0.568 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint8_t>/28
|
-6.63% |
122.635 |
114.508 |
0.581 |
-6.63% |
0.581 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/28
|
-6.60% |
51.581 |
48.177 |
0.445 |
-6.63% |
0.445 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/999
|
-6.54% |
51.553 |
48.179 |
0.000 |
-6.68% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/256
|
-6.54% |
51.548 |
48.178 |
0.011 |
-6.66% |
0.011 |
|
MultiSource/Benchmarks/DOE-ProxyApps-C++/miniFE/miniFE
Profile
|
-6.30% |
17.743 |
16.626 |
0.047 |
-6.85% |
0.047 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/28
|
-6.25% |
120.132 |
112.625 |
0.004 |
-6.25% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint32_t>/999
|
-5.76% |
1453.709 |
1369.931 |
0.790 |
-5.76% |
0.790 |
|
MultiSource/Benchmarks/DOE-ProxyApps-C/miniGMG/miniGMG
Profile
|
-5.75% |
10.443 |
9.843 |
0.043 |
-0.42% |
0.043 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/10
|
-5.69% |
76.958 |
72.580 |
0.002 |
-5.69% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint64_t>/999
|
-5.65% |
5517.054 |
5205.085 |
1.041 |
-5.66% |
1.041 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint32_t>/256
|
-5.64% |
1087.470 |
1026.187 |
0.117 |
-5.64% |
0.117 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint32_t>/51
|
-5.57% |
241.836 |
228.374 |
0.007 |
-5.57% |
0.007 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint8_t>/999
|
-5.56% |
2847.820 |
2689.379 |
0.194 |
-5.56% |
0.194 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint16_t>/999
|
-5.56% |
1451.152 |
1370.522 |
0.280 |
-5.55% |
0.280 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint32_t>/256
|
-5.54% |
580.978 |
548.763 |
0.019 |
-5.55% |
0.019 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint8_t>/256
|
-5.54% |
744.564 |
703.286 |
0.017 |
-5.54% |
0.017 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/256
|
-5.54% |
745.200 |
703.898 |
0.014 |
-5.55% |
0.014 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/256
|
-5.54% |
745.202 |
703.931 |
0.028 |
-5.54% |
0.028 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/999
|
-5.53% |
2848.348 |
2690.751 |
0.214 |
-5.54% |
0.214 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint8_t>/51
|
-5.52% |
170.189 |
160.803 |
0.004 |
-5.52% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/51
|
-5.50% |
170.828 |
161.431 |
0.004 |
-5.50% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint32_t>/28
|
-5.49% |
142.344 |
134.522 |
0.001 |
-5.50% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/51
|
-5.49% |
170.805 |
161.429 |
0.004 |
-5.50% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint32_t>/256
|
-5.48% |
400.496 |
378.548 |
0.011 |
-5.48% |
0.011 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint8_t>/256
|
-5.48% |
399.818 |
377.914 |
0.013 |
-5.48% |
0.013 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/999
|
-5.48% |
2846.863 |
2690.901 |
0.452 |
-5.48% |
0.452 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint16_t>/256
|
-5.47% |
400.457 |
378.544 |
0.012 |
-5.47% |
0.012 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint32_t>/999
|
-5.47% |
4175.196 |
3946.831 |
0.232 |
-5.91% |
0.232 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint64_t>/51
|
-5.45% |
310.338 |
293.439 |
0.017 |
-5.45% |
0.017 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint8_t>/999
|
-5.44% |
1450.597 |
1371.709 |
0.038 |
-5.44% |
0.038 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint32_t>/16
|
-5.41% |
92.604 |
87.595 |
0.001 |
-5.41% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/16
|
-5.39% |
52.248 |
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 |
49.429 |
0.002 |
-5.40% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/51
|
-5.39% |
52.247 |
49.431 |
0.001 |
-5.39% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/28
|
-5.39% |
52.244 |
49.428 |
0.002 |
-5.40% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/999
|
-5.39% |
52.244 |
49.429 |
0.002 |
-5.39% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/256
|
-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% |
2131.746 |
2017.574 |
0.025 |
-5.36% |
0.025 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint64_t>/999
|
-5.35% |
2803.857 |
2653.721 |
0.237 |
-5.38% |
0.237 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint64_t>/999
|
-5.32% |
3006.022 |
2846.103 |
0.371 |
-5.32% |
0.371 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/10
|
-5.18% |
16.894 |
16.018 |
0.145 |
-5.19% |
0.145 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint32_t>/51
|
-5.12% |
110.126 |
104.493 |
0.002 |
-5.13% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint8_t>/51
|
-5.08% |
110.738 |
105.114 |
0.005 |
-5.09% |
0.005 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint32_t>/51
|
-5.01% |
156.116 |
148.289 |
0.004 |
-5.01% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/999
|
-4.99% |
1528.946 |
1452.713 |
0.106 |
-5.00% |
0.106 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/256
|
-4.94% |
20.022 |
19.033 |
0.123 |
-4.94% |
0.123 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint64_t>/51
|
-4.92% |
178.327 |
169.546 |
0.003 |
-4.92% |
0.003 |
|
MultiSource/Benchmarks/TSVC/IndirectAddressing-flt/IndirectAddressing-flt
Profile
|
-4.90% |
15.779 |
15.006 |
0.036 |
-4.71% |
0.036 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint32_t>/16
|
-4.87% |
83.529 |
79.461 |
0.004 |
-4.87% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint64_t>/51
|
-4.77% |
196.460 |
187.085 |
0.005 |
-4.78% |
0.005 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint8_t>/256
|
-4.76% |
39.418 |
37.540 |
0.002 |
-4.77% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint8_t>/999
|
-4.76% |
39.419 |
37.542 |
0.001 |
-4.76% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint8_t>/51
|
-4.76% |
39.417 |
37.541 |
0.001 |
-4.77% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint32_t>/51
|
-4.69% |
40.044 |
38.166 |
0.002 |
-4.69% |
0.002 |
|
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>/51
|
-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>/256
|
-4.69% |
40.045 |
38.168 |
0.000 |
-4.69% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint16_t>/999
|
-4.68% |
40.043 |
38.168 |
0.000 |
-4.69% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint16_t>/256
|
-4.68% |
40.043 |
38.168 |
0.001 |
-4.69% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint32_t>/28
|
-4.45% |
112.627 |
107.619 |
0.003 |
-4.45% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/51
|
-4.41% |
113.871 |
108.852 |
0.007 |
-4.42% |
0.007 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/28
|
-4.35% |
50.372 |
48.178 |
0.005 |
-4.35% |
0.005 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/51
|
-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>/16
|
-4.35% |
50.368 |
48.178 |
0.001 |
-4.35% |
0.001 |
|
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>/999
|
-4.35% |
50.366 |
48.178 |
0.001 |
-4.35% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/16
|
-4.30% |
16.893 |
16.167 |
0.053 |
-4.30% |
0.053 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/28
|
-4.25% |
16.894 |
16.177 |
0.046 |
-4.25% |
0.046 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint16_t>/10
|
-4.13% |
22.236 |
21.318 |
0.304 |
-4.17% |
0.304 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint16_t>/51
|
-4.00% |
109.501 |
105.117 |
0.005 |
-4.00% |
0.005 |
|
SingleSource/Benchmarks/Misc-C++/oopack_v1p8
Profile
|
-3.85% |
0.478 |
0.460 |
0.000 |
-3.88% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/999
|
-3.63% |
40.045 |
38.594 |
0.088 |
-3.63% |
0.088 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint8_t>/10
|
-3.60% |
24.544 |
23.662 |
0.164 |
-3.57% |
0.164 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/51
|
-3.52% |
40.043 |
38.634 |
0.069 |
-3.52% |
0.069 |
|
MultiSource/Benchmarks/TSVC/IndirectAddressing-dbl/IndirectAddressing-dbl
Profile
|
-3.46% |
18.832 |
18.181 |
0.145 |
-1.85% |
0.145 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint64_t>/999
|
-3.30% |
56.938 |
55.061 |
0.001 |
-3.30% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint64_t>/51
|
-3.30% |
56.938 |
55.061 |
0.002 |
-3.30% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint64_t>/10
|
-3.03% |
40.671 |
39.438 |
0.092 |
-3.04% |
0.092 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC15
|
-2.94% |
21.274 |
20.649 |
0.001 |
-2.94% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC16
|
-2.89% |
21.262 |
20.649 |
0.001 |
-2.66% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/10
|
-2.78% |
41.296 |
40.147 |
0.024 |
-2.78% |
0.024 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint16_t>/16
|
-2.45% |
29.533 |
28.809 |
0.006 |
-2.75% |
0.006 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/10
|
-2.27% |
30.629 |
29.934 |
0.019 |
-2.37% |
0.019 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW1LoopWithReductionTC32
|
-2.17% |
28.783 |
28.157 |
0.001 |
-2.18% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC31
|
-2.17% |
28.783 |
28.157 |
0.001 |
-2.17% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC31
|
-2.17% |
28.781 |
28.157 |
0.001 |
-2.17% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC31
|
-2.17% |
28.781 |
28.156 |
0.001 |
-2.18% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4LoopWithReductionTC32
|
-2.17% |
28.781 |
28.157 |
0.001 |
-2.17% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4LoopWithReductionTC31
|
-2.17% |
28.780 |
28.157 |
0.001 |
-2.16% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW1LoopWithReductionTC31
|
-2.17% |
28.780 |
28.157 |
0.001 |
-2.16% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForLoopWithReductionAutoVecTC31
|
-2.16% |
28.780 |
28.157 |
0.001 |
-2.16% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC32
|
-2.15% |
28.776 |
28.158 |
0.000 |
-2.13% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC32
|
-2.15% |
28.775 |
28.157 |
0.000 |
-2.15% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC31
|
-2.12% |
28.767 |
28.157 |
0.001 |
-2.15% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint64_t>/16
|
-2.11% |
55.692 |
54.518 |
0.013 |
-2.11% |
0.013 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/16
|
-2.10% |
56.311 |
55.131 |
0.020 |
-2.10% |
0.020 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC32
|
-2.09% |
28.757 |
28.157 |
0.000 |
-2.12% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForLoopWithReductionAutoVecTC32
|
-2.05% |
28.747 |
28.157 |
0.001 |
-1.98% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC32
|
-1.99% |
28.730 |
28.157 |
0.000 |
-1.99% |
0.000 |
|
SingleSource/Benchmarks/CoyoteBench/huffbench
Profile
|
-1.76% |
50.626 |
49.737 |
0.005 |
-1.76% |
0.005 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/16
|
-1.75% |
38.120 |
37.451 |
0.020 |
-1.54% |
0.020 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4BigLoopTC1
|
-1.67% |
109.562 |
107.733 |
0.685 |
0.12% |
0.685 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint8_t>/10
|
-1.62% |
48.967 |
48.176 |
0.002 |
-1.49% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC15
|
-1.61% |
20.987 |
20.649 |
0.000 |
-1.35% |
0.000 |
|
MultiSource/Applications/sgefa/sgefa
Profile
|
-1.61% |
0.638 |
0.628 |
0.002 |
-2.26% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4BigLoopTC2
|
-1.58% |
109.638 |
107.904 |
0.530 |
-0.18% |
0.530 |
|
MultiSource/Benchmarks/TSVC/CrossingThresholds-flt/CrossingThresholds-flt
Profile
|
-1.50% |
11.168 |
11.001 |
0.049 |
-0.69% |
0.049 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC63
|
-1.43% |
43.800 |
43.174 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC63
|
-1.43% |
43.800 |
43.174 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC63
|
-1.43% |
43.798 |
43.173 |
0.002 |
-1.43% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW1LoopWithReductionTC64
|
-1.43% |
43.799 |
43.174 |
0.002 |
-1.43% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4LoopWithReductionTC63
|
-1.43% |
43.798 |
43.174 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4LoopWithReductionTC64
|
-1.43% |
43.799 |
43.174 |
0.001 |
-1.42% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC64
|
-1.43% |
43.799 |
43.174 |
0.001 |
-1.42% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC64
|
-1.43% |
43.799 |
43.175 |
0.001 |
-1.42% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW1LoopWithReductionTC63
|
-1.42% |
43.798 |
43.175 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC64
|
-1.42% |
43.799 |
43.175 |
0.000 |
-1.42% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForLoopWithReductionAutoVecTC63
|
-1.42% |
43.798 |
43.174 |
0.000 |
-1.43% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC63
|
-1.42% |
43.798 |
43.175 |
0.003 |
-1.43% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC64
|
-1.42% |
43.798 |
43.175 |
0.003 |
-1.43% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForLoopWithReductionAutoVecTC64
|
-1.42% |
43.798 |
43.176 |
0.000 |
-1.42% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4BigLoopTC4
|
-1.41% |
108.461 |
106.930 |
0.428 |
-0.19% |
0.428 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/28
|
-1.40% |
86.346 |
85.139 |
0.008 |
-1.40% |
0.008 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint64_t>/28
|
-1.38% |
85.721 |
84.536 |
0.033 |
-1.39% |
0.033 |
|
External/SPEC/CFP2017rate/511.povray_r/511.povray_r
Profile
|
-1.34% |
28.620 |
28.237 |
0.057 |
-1.39% |
0.057 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4BigLoopTC15
|
-1.33% |
267.323 |
263.777 |
1.285 |
-3.16% |
1.285 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4BigLoopTC15
|
-1.29% |
267.576 |
264.117 |
1.140 |
-3.37% |
1.140 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/28
|
-1.26% |
53.119 |
52.452 |
0.024 |
-1.14% |
0.024 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4BigLoopTC4
|
-1.19% |
108.569 |
107.272 |
0.157 |
0.17% |
0.157 |
|
MultiSource/Applications/lua/lua
Profile
|
-1.18% |
70.888 |
70.048 |
0.042 |
-1.22% |
0.042 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC15
|
-1.12% |
20.883 |
20.648 |
0.002 |
-2.05% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_single_csa_nested_cond_load_autovec_int64_t_
|
-1.09% |
618764.697 |
612044.881 |
1185.500 |
-0.89% |
1185.500 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4BigLoopTC15
|
-1.09% |
267.163 |
264.264 |
1.094 |
-2.82% |
1.094 |