|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_autovec_int64_t_
|
-27.50% |
1043411.064 |
756462.212 |
2222.979 |
-27.60% |
2222.979 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_novec_int64_t_
|
-27.49% |
1044293.094 |
757253.666 |
639.426 |
-27.50% |
639.426 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_autovec_int32_t_
|
-26.89% |
515642.851 |
376972.355 |
335.763 |
-26.90% |
335.763 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_novec_int32_t_
|
-26.89% |
515697.407 |
377033.013 |
549.961 |
-26.91% |
549.961 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint64_t>/10
|
-26.82% |
38.569 |
28.225 |
0.012 |
-26.95% |
0.012 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint64_t>/16
|
-25.88% |
53.259 |
39.477 |
0.009 |
-25.82% |
0.009 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_autovec_uint8_t_
|
-25.05% |
124983.119 |
93674.878 |
36.981 |
-25.01% |
36.981 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_with_cond_arith_novec_uint8_t_
|
-25.03% |
124992.432 |
93708.550 |
17.331 |
-24.98% |
17.331 |
|
SingleSource/Benchmarks/Shootout/Shootout-ary3
Profile
|
-23.50% |
0.984 |
0.753 |
0.001 |
-23.55% |
0.001 |
|
MultiSource/Benchmarks/Ptrdist/yacr2/yacr2
Profile
|
-23.03% |
2.836 |
2.183 |
0.000 |
-23.01% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint64_t>/16
|
-20.88% |
32.546 |
25.752 |
0.013 |
-20.87% |
0.013 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint16_t>/10
|
-19.86% |
9.385 |
7.521 |
0.017 |
-19.86% |
0.017 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint16_t>/16
|
-19.85% |
9.385 |
7.523 |
0.015 |
-19.85% |
0.015 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint8_t>/16
|
-19.45% |
60.718 |
48.906 |
0.013 |
-18.62% |
0.013 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint8_t>/10
|
-18.13% |
41.296 |
33.808 |
0.043 |
-18.15% |
0.043 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint64_t>/16
|
-17.91% |
10.017 |
8.223 |
0.040 |
-17.93% |
0.040 |
|
MultiSource/Benchmarks/Olden/em3d/em3d
Profile
|
-17.89% |
18.303 |
15.028 |
0.007 |
-17.90% |
0.007 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint64_t>/10
|
-17.76% |
10.017 |
8.238 |
0.019 |
-17.72% |
0.019 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint64_t>/10
|
-17.34% |
24.391 |
20.162 |
0.023 |
-17.28% |
0.023 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/28
|
-17.07% |
104.493 |
86.652 |
0.007 |
-17.08% |
0.007 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/16
|
-15.42% |
66.949 |
56.623 |
0.004 |
-15.43% |
0.004 |
|
MultiSource/Benchmarks/SciMark2-C/scimark2
Profile
|
-15.06% |
166.580 |
141.490 |
0.047 |
-15.11% |
0.047 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint16_t>/10
|
-14.00% |
35.652 |
30.660 |
0.006 |
-14.00% |
0.006 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/10
|
-13.63% |
48.179 |
41.612 |
0.000 |
-13.64% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint8_t>/16
|
-10.60% |
33.956 |
30.357 |
0.108 |
-10.69% |
0.108 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_only_novec_int64_t_
|
-9.90% |
1044938.586 |
941514.839 |
782.021 |
-9.96% |
782.021 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_only_autovec_int64_t_
|
-9.65% |
1044350.493 |
943562.749 |
624.743 |
-9.81% |
624.743 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_single_load<uint16_t>/16
|
-8.61% |
50.004 |
45.697 |
0.002 |
-8.51% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_only_autovec_int32_t_
|
-8.60% |
515233.743 |
470917.898 |
653.446 |
-8.62% |
653.446 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:BENCHMARK_multi_csa_only_novec_int32_t_
|
-8.51% |
515047.599 |
471211.333 |
549.777 |
-8.53% |
549.777 |
|
MultiSource/Benchmarks/ASC_Sequoia/AMGmk/AMGmk
Profile
|
-8.50% |
43.481 |
39.783 |
0.035 |
-8.46% |
0.035 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint8_t>/28
|
-8.47% |
157.963 |
144.588 |
0.454 |
-8.47% |
0.454 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/256
|
-8.39% |
1207.602 |
1106.238 |
0.041 |
-8.40% |
0.041 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint8_t>/51
|
-8.38% |
272.495 |
249.665 |
0.346 |
-8.38% |
0.346 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/999
|
-8.32% |
2364.872 |
2168.155 |
0.076 |
-8.33% |
0.076 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint8_t>/256
|
-8.24% |
1245.758 |
1143.167 |
0.637 |
-8.24% |
0.637 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/256
|
-8.22% |
642.912 |
590.040 |
0.027 |
-8.22% |
0.027 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint8_t>/999
|
-8.21% |
4804.772 |
4410.379 |
0.291 |
-8.21% |
0.291 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/999
|
-8.17% |
4636.134 |
4257.383 |
0.248 |
-8.39% |
0.248 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/51
|
-8.11% |
266.234 |
244.651 |
0.031 |
-8.11% |
0.031 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint8_t>/999
|
-8.08% |
2443.100 |
2245.746 |
0.578 |
-8.08% |
0.578 |
|
SingleSource/Benchmarks/Adobe-C++/simple_types_constant_folding
Profile
|
-8.04% |
2.366 |
2.175 |
0.000 |
-8.07% |
0.000 |
|
MultiSource/Benchmarks/DOE-ProxyApps-C++/HPCCG/HPCCG
Profile
|
-7.92% |
4.648 |
4.280 |
0.039 |
-8.01% |
0.039 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint8_t>/256
|
-7.89% |
662.266 |
610.039 |
0.022 |
-7.90% |
0.022 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/28
|
-7.84% |
155.480 |
143.292 |
0.009 |
-7.85% |
0.009 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/16
|
-7.50% |
100.113 |
92.604 |
0.004 |
-7.50% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint8_t>/51
|
-7.39% |
172.961 |
160.183 |
0.568 |
-7.37% |
0.568 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint16_t>/10
|
-7.28% |
81.654 |
75.709 |
0.002 |
-7.28% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/51
|
-7.20% |
169.256 |
157.065 |
0.003 |
-7.20% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/16
|
-6.66% |
89.159 |
83.220 |
0.006 |
-6.67% |
0.006 |
|
MultiSource/Benchmarks/DOE-ProxyApps-C++/miniFE/miniFE
Profile
|
-6.57% |
17.779 |
16.611 |
0.052 |
-6.93% |
0.052 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/28
|
-6.49% |
51.523 |
48.179 |
0.552 |
-6.63% |
0.552 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/999
|
-6.25% |
51.545 |
48.322 |
0.343 |
-6.40% |
0.343 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/28
|
-6.25% |
120.136 |
112.629 |
0.004 |
-6.25% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint8_t>/28
|
-6.01% |
122.639 |
115.270 |
0.198 |
-6.01% |
0.198 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint32_t>/999
|
-5.75% |
1453.295 |
1369.796 |
0.818 |
-5.77% |
0.818 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint16_t>/10
|
-5.69% |
76.960 |
72.582 |
0.004 |
-5.69% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint64_t>/999
|
-5.67% |
5517.522 |
5204.741 |
0.804 |
-5.67% |
0.804 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint32_t>/256
|
-5.64% |
1087.508 |
1026.211 |
0.046 |
-5.64% |
0.046 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint32_t>/51
|
-5.57% |
241.836 |
228.376 |
0.012 |
-5.57% |
0.012 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint8_t>/999
|
-5.57% |
2847.643 |
2689.158 |
0.280 |
-5.57% |
0.280 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint16_t>/999
|
-5.56% |
1451.016 |
1370.410 |
0.498 |
-5.56% |
0.498 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint32_t>/256
|
-5.55% |
581.005 |
548.769 |
0.024 |
-5.55% |
0.024 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint8_t>/256
|
-5.54% |
744.581 |
703.297 |
0.021 |
-5.54% |
0.021 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/256
|
-5.54% |
745.197 |
703.918 |
0.036 |
-5.54% |
0.036 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/256
|
-5.54% |
745.209 |
703.961 |
0.007 |
-5.54% |
0.007 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/999
|
-5.53% |
2848.360 |
2690.847 |
0.084 |
-5.53% |
0.084 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint8_t>/51
|
-5.51% |
170.191 |
160.807 |
0.009 |
-5.52% |
0.009 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint32_t>/28
|
-5.50% |
142.353 |
134.521 |
0.011 |
-5.50% |
0.011 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint16_t>/51
|
-5.50% |
170.823 |
161.426 |
0.009 |
-5.50% |
0.009 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/51
|
-5.50% |
170.818 |
161.429 |
0.004 |
-5.50% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint32_t>/999
|
-5.49% |
4176.089 |
3946.797 |
0.516 |
-5.91% |
0.516 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/999
|
-5.48% |
2847.428 |
2691.417 |
0.226 |
-5.46% |
0.226 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint8_t>/256
|
-5.48% |
399.824 |
377.929 |
0.011 |
-5.48% |
0.011 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint32_t>/256
|
-5.47% |
400.456 |
378.552 |
0.013 |
-5.48% |
0.013 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint16_t>/256
|
-5.47% |
400.447 |
378.557 |
0.016 |
-5.46% |
0.016 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint64_t>/51
|
-5.45% |
310.359 |
293.448 |
0.025 |
-5.45% |
0.025 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint8_t>/999
|
-5.45% |
1450.598 |
1371.580 |
0.018 |
-5.45% |
0.018 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_three_loads<uint32_t>/16
|
-5.40% |
92.603 |
87.599 |
0.004 |
-5.41% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/16
|
-5.39% |
16.894 |
15.982 |
0.111 |
-5.40% |
0.111 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/51
|
-5.39% |
52.249 |
49.431 |
0.003 |
-5.39% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/999
|
-5.39% |
52.248 |
49.431 |
0.003 |
-5.39% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/10
|
-5.39% |
52.246 |
49.430 |
0.001 |
-5.39% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/256
|
-5.39% |
52.246 |
49.431 |
0.001 |
-5.39% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint16_t>/16
|
-5.39% |
52.245 |
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 |
49.432 |
0.001 |
-5.39% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint32_t>/999
|
-5.36% |
2131.685 |
2017.401 |
0.170 |
-5.37% |
0.170 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint64_t>/999
|
-5.35% |
2803.602 |
2653.544 |
0.271 |
-5.38% |
0.271 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint64_t>/999
|
-5.30% |
3005.797 |
2846.366 |
0.104 |
-5.31% |
0.104 |
|
MicroBenchmarks/ImageProcessing/Dilate/Dilate.test:BENCHMARK_DILATE/128
|
-5.29% |
331.055 |
313.528 |
1.834 |
-3.45% |
1.834 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/51
|
-5.27% |
51.524 |
48.807 |
0.179 |
-5.37% |
0.179 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint8_t>/256
|
-5.21% |
51.556 |
48.870 |
0.130 |
-5.32% |
0.130 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint32_t>/51
|
-5.11% |
110.124 |
104.497 |
0.003 |
-5.13% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint8_t>/51
|
-5.09% |
110.749 |
105.117 |
0.002 |
-5.08% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint32_t>/51
|
-5.01% |
156.114 |
148.290 |
0.003 |
-5.01% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/999
|
-4.99% |
1528.757 |
1452.517 |
0.151 |
-5.01% |
0.151 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint64_t>/51
|
-4.92% |
178.324 |
169.554 |
0.004 |
-4.92% |
0.004 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint32_t>/16
|
-4.87% |
83.530 |
79.464 |
0.006 |
-4.87% |
0.006 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint64_t>/51
|
-4.78% |
196.475 |
187.085 |
0.009 |
-4.78% |
0.009 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint8_t>/51
|
-4.76% |
39.419 |
37.541 |
0.003 |
-4.77% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint8_t>/999
|
-4.76% |
39.420 |
37.544 |
0.001 |
-4.76% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint8_t>/256
|
-4.76% |
39.419 |
37.542 |
0.004 |
-4.76% |
0.004 |
|
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<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>/999
|
-4.69% |
40.045 |
38.168 |
0.001 |
-4.69% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint16_t>/256
|
-4.69% |
40.045 |
38.169 |
0.001 |
-4.69% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint16_t>/51
|
-4.69% |
40.045 |
38.169 |
0.002 |
-4.69% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint32_t>/256
|
-4.68% |
40.045 |
38.169 |
0.000 |
-4.69% |
0.000 |
|
MultiSource/Benchmarks/TSVC/IndirectAddressing-flt/IndirectAddressing-flt
Profile
|
-4.66% |
15.736 |
15.002 |
0.025 |
-4.73% |
0.025 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/10
|
-4.47% |
16.894 |
16.139 |
0.060 |
-4.47% |
0.060 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_three_loads<uint32_t>/28
|
-4.45% |
112.631 |
107.621 |
0.005 |
-4.45% |
0.005 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/51
|
-4.41% |
113.877 |
108.854 |
0.008 |
-4.42% |
0.008 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/28
|
-4.39% |
16.894 |
16.152 |
0.025 |
-4.39% |
0.025 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/28
|
-4.35% |
50.371 |
48.179 |
0.002 |
-4.35% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/256
|
-4.35% |
50.368 |
48.177 |
0.001 |
-4.36% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/16
|
-4.35% |
50.369 |
48.180 |
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 |
0.001 |
-4.35% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint32_t>/999
|
-4.35% |
50.368 |
48.178 |
0.001 |
-4.35% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint16_t>/51
|
-4.00% |
109.499 |
105.114 |
0.004 |
-4.01% |
0.004 |
|
SingleSource/Benchmarks/Misc-C++/oopack_v1p8
Profile
|
-3.97% |
0.479 |
0.460 |
0.000 |
-3.98% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/256
|
-3.93% |
20.022 |
19.234 |
0.250 |
-3.94% |
0.250 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/256
|
-3.60% |
359.343 |
346.408 |
0.047 |
-2.50% |
0.047 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/999
|
-3.51% |
40.045 |
38.641 |
0.058 |
-3.51% |
0.058 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint64_t>/51
|
-3.30% |
56.940 |
55.062 |
0.002 |
-3.30% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_three_loads<uint64_t>/999
|
-3.29% |
56.939 |
55.065 |
0.002 |
-3.29% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_two_loads<uint64_t>/51
|
-3.18% |
40.044 |
38.769 |
0.005 |
-3.19% |
0.005 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint64_t>/10
|
-3.06% |
40.676 |
39.433 |
0.020 |
-3.05% |
0.020 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4BigLoopTC15
|
-3.04% |
272.646 |
264.362 |
1.587 |
-2.95% |
1.587 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4BigLoopTC15
|
-3.03% |
272.577 |
264.309 |
2.053 |
-3.30% |
2.053 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC15
|
-2.94% |
21.273 |
20.649 |
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/IndirectAddressing-dbl/IndirectAddressing-dbl
Profile
|
-2.92% |
18.870 |
18.320 |
0.109 |
-1.09% |
0.109 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4BigLoopTC15
|
-2.89% |
272.195 |
264.334 |
2.053 |
-2.79% |
2.053 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/10
|
-2.78% |
41.297 |
40.150 |
0.023 |
-2.77% |
0.023 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/10
|
-2.48% |
30.641 |
29.882 |
0.022 |
-2.55% |
0.022 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint16_t>/16
|
-2.42% |
29.537 |
28.821 |
0.128 |
-2.71% |
0.128 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_single_load<uint8_t>/10
|
-2.31% |
24.552 |
23.986 |
0.083 |
-2.25% |
0.083 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC31
|
-2.18% |
28.784 |
28.156 |
0.001 |
-2.17% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint64_t>/16
|
-2.18% |
55.690 |
54.476 |
0.084 |
-2.19% |
0.084 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC31
|
-2.18% |
28.783 |
28.157 |
0.001 |
-2.18% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW1LoopWithReductionTC32
|
-2.17% |
28.782 |
28.157 |
0.001 |
-2.18% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC31
|
-2.17% |
28.782 |
28.158 |
0.000 |
-2.17% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4LoopWithReductionTC32
|
-2.17% |
28.780 |
28.157 |
0.001 |
-2.17% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC32
|
-2.16% |
28.779 |
28.156 |
0.001 |
-2.14% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW1LoopWithReductionTC31
|
-2.16% |
28.780 |
28.157 |
0.002 |
-2.16% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4LoopWithReductionTC31
|
-2.16% |
28.779 |
28.156 |
0.000 |
-2.16% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForLoopWithReductionAutoVecTC31
|
-2.16% |
28.779 |
28.157 |
0.000 |
-2.16% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC32
|
-2.16% |
28.778 |
28.156 |
0.001 |
-2.16% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC31
|
-2.16% |
28.779 |
28.157 |
0.000 |
-2.15% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC32
|
-2.11% |
28.762 |
28.156 |
0.001 |
-2.13% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/16
|
-2.05% |
56.313 |
55.157 |
0.010 |
-2.06% |
0.010 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC15
|
-2.00% |
21.069 |
20.648 |
0.000 |
-2.05% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC15
|
-1.95% |
21.058 |
20.648 |
0.000 |
-1.36% |
0.000 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/16
|
-1.93% |
38.103 |
37.366 |
0.041 |
-1.76% |
0.041 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForLoopWithReductionAutoVecTC32
|
-1.92% |
28.707 |
28.157 |
0.001 |
-1.98% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC32
|
-1.84% |
28.686 |
28.157 |
0.000 |
-1.99% |
0.000 |
|
SingleSource/Benchmarks/CoyoteBench/huffbench
Profile
|
-1.76% |
50.630 |
49.740 |
0.001 |
-1.75% |
0.001 |
|
External/SPEC/CINT2017rate/523.xalancbmk_r/523.xalancbmk_r
Profile
|
-1.73% |
212.761 |
209.072 |
0.487 |
-0.22% |
0.487 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint8_t>/10
|
-1.51% |
48.909 |
48.171 |
0.002 |
-1.50% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC64
|
-1.44% |
43.805 |
43.174 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW1LoopWithReductionTC64
|
-1.44% |
43.802 |
43.173 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC64
|
-1.44% |
43.801 |
43.172 |
0.002 |
-1.43% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4LoopWithReductionTC63
|
-1.43% |
43.800 |
43.172 |
0.002 |
-1.43% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForLoopWithReductionAutoVecTC63
|
-1.43% |
43.799 |
43.172 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForLoopWithReductionAutoVecTC64
|
-1.43% |
43.799 |
43.172 |
0.002 |
-1.43% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC63
|
-1.43% |
43.799 |
43.172 |
0.002 |
-1.43% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW1LoopWithReductionTC63
|
-1.43% |
43.798 |
43.172 |
0.002 |
-1.43% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC64
|
-1.43% |
43.798 |
43.172 |
0.003 |
-1.43% |
0.003 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC2VW4LoopWithReductionTC64
|
-1.43% |
43.798 |
43.173 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW4LoopWithReductionTC63
|
-1.43% |
43.798 |
43.173 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW4LoopWithReductionTC64
|
-1.43% |
43.799 |
43.174 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC4VW1LoopWithReductionTC63
|
-1.42% |
43.797 |
43.173 |
0.002 |
-1.43% |
0.002 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint64_t>/28
|
-1.42% |
85.731 |
84.512 |
0.049 |
-1.42% |
0.049 |
|
MicroBenchmarks/LoopVectorization/LoopInterleavingBenchmarks.test:benchForIC1VW1LoopWithReductionTC63
|
-1.42% |
43.797 |
43.175 |
0.001 |
-1.43% |
0.001 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_no_early_exit_two_loads<uint32_t>/28
|
-1.40% |
86.349 |
85.143 |
0.005 |
-1.40% |
0.005 |
|
External/SPEC/CFP2017rate/511.povray_r/511.povray_r
Profile
|
-1.31% |
28.674 |
28.299 |
0.017 |
-1.17% |
0.017 |
|
MicroBenchmarks/ImageProcessing/BilateralFiltering/BilateralFilter.test:BENCHMARK_BILATERAL_FILTER/64/2
|
-1.30% |
2571.072 |
2537.660 |
1.033 |
-0.54% |
1.033 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_mid_two_loads<uint64_t>/28
|
-1.27% |
53.143 |
52.467 |
0.011 |
-1.11% |
0.011 |
|
MultiSource/Applications/lua/lua
Profile
|
-1.27% |
70.925 |
70.024 |
0.026 |
-1.26% |
0.026 |
|
MicroBenchmarks/LoopVectorization/LoopVectorizationBenchmarks.test:autovec_early_exit_taken_first_single_load<uint8_t>/16
|
-1.08% |
9.908 |
9.801 |
0.025 |
1.75% |
0.025 |