polytypic
commented
1 year ago As expected this seems to give a small improvement in cases where the are no CMP operations followed by a CAS. In other cases the added instructions seem to be essentially free.
Benchmark 1: kcas-main/_build/default/test/kcas_data/hashtbl_bench.exe 1 1_000_000 1000 90
Time (mean ± σ): 38.3 ms ± 0.2 ms [User: 37.2 ms, System: 0.8 ms]
Range (min … max): 38.1 ms … 39.0 ms 76 runs
Benchmark 2: kcas-this/_build/default/test/kcas_data/hashtbl_bench.exe 1 1_000_000 1000 90
Time (mean ± σ): 37.5 ms ± 0.4 ms [User: 36.4 ms, System: 0.8 ms]
Range (min … max): 37.0 ms … 38.9 ms 79 runs
Summary
'kcas-this/_build/default/test/kcas_data/hashtbl_bench.exe 1 1_000_000 1000 90' ran
1.02 ± 0.01 times faster than 'kcas-main/_build/default/test/kcas_data/hashtbl_bench.exe 1 1_000_000 1000 90'
Benchmark 1: kcas-main/_build/default/test/kcas_data/hashtbl_bench.exe 2 1_000_000 1000 90
Time (mean ± σ): 22.8 ms ± 0.1 ms [User: 41.6 ms, System: 1.0 ms]
Range (min … max): 22.6 ms … 23.4 ms 128 runs
Benchmark 2: kcas-this/_build/default/test/kcas_data/hashtbl_bench.exe 2 1_000_000 1000 90
Time (mean ± σ): 22.7 ms ± 0.4 ms [User: 41.4 ms, System: 1.1 ms]
Range (min … max): 22.4 ms … 26.3 ms 129 runs
Warning: Statistical outliers were detected. Consider re-running this benchmark on a quiet system without any interferences from other programs. It might help to use the '--warmup' or '--prepare' options.
Summary
'kcas-this/_build/default/test/kcas_data/hashtbl_bench.exe 2 1_000_000 1000 90' ran
1.00 ± 0.02 times faster than 'kcas-main/_build/default/test/kcas_data/hashtbl_bench.exe 2 1_000_000 1000 90'
Benchmark 1: kcas-main/_build/default/test/kcas_data/hashtbl_bench.exe 4 1_000_000 1000 90
Time (mean ± σ): 15.2 ms ± 1.2 ms [User: 49.1 ms, System: 1.9 ms]
Range (min … max): 14.6 ms … 24.8 ms 199 runs
Warning: Statistical outliers were detected. Consider re-running this benchmark on a quiet system without any interferences from other programs. It might help to use the '--warmup' or '--prepare' options.
Benchmark 2: kcas-this/_build/default/test/kcas_data/hashtbl_bench.exe 4 1_000_000 1000 90
Time (mean ± σ): 14.9 ms ± 0.3 ms [User: 48.1 ms, System: 1.9 ms]
Range (min … max): 14.5 ms … 17.7 ms 198 runs
Warning: Statistical outliers were detected. Consider re-running this benchmark on a quiet system without any interferences from other programs. It might help to use the '--warmup' or '--prepare' options.
Summary
'kcas-this/_build/default/test/kcas_data/hashtbl_bench.exe 4 1_000_000 1000 90' ran
1.02 ± 0.08 times faster than 'kcas-main/_build/default/test/kcas_data/hashtbl_bench.exe 4 1_000_000 1000 90'
Benchmark 1: kcas-main/_build/default/test/kcas_data/hashtbl_bench.exe 1 1_000_000 1000 10
Time (mean ± σ): 225.3 ms ± 0.5 ms [User: 223.2 ms, System: 1.7 ms]
Range (min … max): 224.6 ms … 226.8 ms 13 runs
Benchmark 2: kcas-this/_build/default/test/kcas_data/hashtbl_bench.exe 1 1_000_000 1000 10
Time (mean ± σ): 216.1 ms ± 0.6 ms [User: 214.1 ms, System: 1.6 ms]
Range (min … max): 215.2 ms … 217.3 ms 13 runs
Summary
'kcas-this/_build/default/test/kcas_data/hashtbl_bench.exe 1 1_000_000 1000 10' ran
1.04 ± 0.00 times faster than 'kcas-main/_build/default/test/kcas_data/hashtbl_bench.exe 1 1_000_000 1000 10'
Benchmark 1: kcas-main/_build/default/test/kcas_data/hashtbl_bench.exe 2 1_000_000 1000 10
Time (mean ± σ): 115.5 ms ± 0.5 ms [User: 225.6 ms, System: 1.8 ms]
Range (min … max): 114.9 ms … 116.8 ms 26 runs
Benchmark 2: kcas-this/_build/default/test/kcas_data/hashtbl_bench.exe 2 1_000_000 1000 10
Time (mean ± σ): 114.4 ms ± 0.4 ms [User: 223.8 ms, System: 1.4 ms]
Range (min … max): 114.0 ms … 115.5 ms 26 runs
Summary
'kcas-this/_build/default/test/kcas_data/hashtbl_bench.exe 2 1_000_000 1000 10' ran
1.01 ± 0.01 times faster than 'kcas-main/_build/default/test/kcas_data/hashtbl_bench.exe 2 1_000_000 1000 10'
Benchmark 1: kcas-main/_build/default/test/kcas_data/hashtbl_bench.exe 4 1_000_000 1000 10
Time (mean ± σ): 71.0 ms ± 0.4 ms [User: 270.4 ms, System: 2.4 ms]
Range (min … max): 70.3 ms … 72.3 ms 42 runs
Benchmark 2: kcas-this/_build/default/test/kcas_data/hashtbl_bench.exe 4 1_000_000 1000 10
Time (mean ± σ): 70.6 ms ± 0.3 ms [User: 268.7 ms, System: 2.5 ms]
Range (min … max): 70.1 ms … 71.4 ms 42 runs
Summary
'kcas-this/_build/default/test/kcas_data/hashtbl_bench.exe 4 1_000_000 1000 10' ran
1.01 ± 0.01 times faster than 'kcas-main/_build/default/test/kcas_data/hashtbl_bench.exe 4 1_000_000 1000 10'
Benchmark 1: kcas-main/_build/default/test/kcas/xt_benchmark.exe 1 10000
Time (mean ± σ): 6.9 ms ± 0.1 ms [User: 6.0 ms, System: 0.7 ms]
Range (min … max): 6.9 ms … 8.2 ms 423 runs
Warning: Statistical outliers were detected. Consider re-running this benchmark on a quiet system without any interferences from other programs. It might help to use the '--warmup' or '--prepare' options.
Benchmark 2: kcas-this/_build/default/test/kcas/xt_benchmark.exe 1 10000
Time (mean ± σ): 6.9 ms ± 0.1 ms [User: 6.0 ms, System: 0.7 ms]
Range (min … max): 6.9 ms … 8.2 ms 434 runs
Warning: Statistical outliers were detected. Consider re-running this benchmark on a quiet system without any interferences from other programs. It might help to use the '--warmup' or '--prepare' options.
Summary
'kcas-this/_build/default/test/kcas/xt_benchmark.exe 1 10000' ran
1.00 ± 0.02 times faster than 'kcas-main/_build/default/test/kcas/xt_benchmark.exe 1 10000'
Benchmark 1: kcas-main/_build/default/test/kcas/xt_benchmark.exe 2 10000
Time (mean ± σ): 13.2 ms ± 0.4 ms [User: 12.2 ms, System: 0.7 ms]
Range (min … max): 13.0 ms … 18.8 ms 226 runs
Warning: Statistical outliers were detected. Consider re-running this benchmark on a quiet system without any interferences from other programs. It might help to use the '--warmup' or '--prepare' options.
Benchmark 2: kcas-this/_build/default/test/kcas/xt_benchmark.exe 2 10000
Time (mean ± σ): 13.3 ms ± 0.1 ms [User: 12.3 ms, System: 0.7 ms]
Range (min … max): 13.1 ms … 14.0 ms 227 runs
Warning: Statistical outliers were detected. Consider re-running this benchmark on a quiet system without any interferences from other programs. It might help to use the '--warmup' or '--prepare' options.
Summary
'kcas-main/_build/default/test/kcas/xt_benchmark.exe 2 10000' ran
1.00 ± 0.03 times faster than 'kcas-this/_build/default/test/kcas/xt_benchmark.exe 2 10000'
Benchmark 1: kcas-main/_build/default/test/kcas/xt_benchmark.exe 4 10000
Time (mean ± σ): 22.2 ms ± 0.1 ms [User: 21.1 ms, System: 0.7 ms]
Range (min … max): 22.0 ms … 22.8 ms 135 runs
Warning: Statistical outliers were detected. Consider re-running this benchmark on a quiet system without any interferences from other programs. It might help to use the '--warmup' or '--prepare' options.
Benchmark 2: kcas-this/_build/default/test/kcas/xt_benchmark.exe 4 10000
Time (mean ± σ): 22.2 ms ± 0.1 ms [User: 21.2 ms, System: 0.7 ms]
Range (min … max): 22.0 ms … 22.6 ms 134 runs
Summary
'kcas-main/_build/default/test/kcas/xt_benchmark.exe 4 10000' ran
1.00 ± 0.01 times faster than 'kcas-this/_build/default/test/kcas/xt_benchmark.exe 4 10000'
Benchmark 1: kcas-main/_build/default/test/kcas/xt_parallel_cmp_bench.exe 100000
Time (mean ± σ): 19.0 ms ± 1.5 ms [User: 30.4 ms, System: 1.6 ms]
Range (min … max): 17.6 ms … 21.7 ms 142 runs
Warning: The first benchmarking run for this command was significantly slower than the rest (21.0 ms). This could be caused by (filesystem) caches that were not filled until after the first run. You are already using the '--warmup' option which helps to fill these caches before the actual benchmark. You can either try to increase the warmup count further or re-run this benchmark on a quiet system in case it was a random outlier. Alternatively, consider using the '--prepare' option to clear the caches before each timing run.
Benchmark 2: kcas-this/_build/default/test/kcas/xt_parallel_cmp_bench.exe 100000
Time (mean ± σ): 18.7 ms ± 1.9 ms [User: 29.7 ms, System: 1.7 ms]
Range (min … max): 16.9 ms … 27.5 ms 173 runs
Warning: Statistical outliers were detected. Consider re-running this benchmark on a quiet system without any interferences from other programs. It might help to use the '--warmup' or '--prepare' options.
Summary
'kcas-this/_build/default/test/kcas/xt_parallel_cmp_bench.exe 100000' ran
1.02 ± 0.13 times faster than 'kcas-main/_build/default/test/kcas/xt_parallel_cmp_bench.exe 100000'
Benchmark 1: kcas-main/_build/default/test/kcas/xt_parallel_cmp_bench.exe 200000
Time (mean ± σ): 35.5 ms ± 3.0 ms [User: 58.9 ms, System: 2.4 ms]
Range (min … max): 33.1 ms … 43.2 ms 90 runs
Warning: Statistical outliers were detected. Consider re-running this benchmark on a quiet system without any interferences from other programs. It might help to use the '--warmup' or '--prepare' options.
Benchmark 2: kcas-this/_build/default/test/kcas/xt_parallel_cmp_bench.exe 200000
Time (mean ± σ): 34.2 ms ± 3.4 ms [User: 56.7 ms, System: 2.3 ms]
Range (min … max): 31.7 ms … 40.4 ms 92 runs
Warning: Statistical outliers were detected. Consider re-running this benchmark on a quiet system without any interferences from other programs. It might help to use the '--warmup' or '--prepare' options.
Summary
'kcas-this/_build/default/test/kcas/xt_parallel_cmp_bench.exe 200000' ran
1.04 ± 0.14 times faster than 'kcas-main/_build/default/test/kcas/xt_parallel_cmp_bench.exe 200000'
Benchmark 1: kcas-main/_build/default/test/kcas/xt_parallel_cmp_bench.exe 400000
Time (mean ± σ): 69.6 ms ± 6.5 ms [User: 119.4 ms, System: 3.6 ms]
Range (min … max): 63.7 ms … 78.2 ms 38 runs
Warning: The first benchmarking run for this command was significantly slower than the rest (77.9 ms). This could be caused by (filesystem) caches that were not filled until after the first run. You are already using the '--warmup' option which helps to fill these caches before the actual benchmark. You can either try to increase the warmup count further or re-run this benchmark on a quiet system in case it was a random outlier. Alternatively, consider using the '--prepare' option to clear the caches before each timing run.
Benchmark 2: kcas-this/_build/default/test/kcas/xt_parallel_cmp_bench.exe 400000
Time (mean ± σ): 67.8 ms ± 7.5 ms [User: 115.5 ms, System: 3.5 ms]
Range (min … max): 61.4 ms … 78.3 ms 48 runs
Warning: Statistical outliers were detected. Consider re-running this benchmark on a quiet system without any interferences from other programs. It might help to use the '--warmup' or '--prepare' options.
Summary
'kcas-this/_build/default/test/kcas/xt_parallel_cmp_bench.exe 400000' ran
1.03 ± 0.15 times faster than 'kcas-main/_build/default/test/kcas/xt_parallel_cmp_bench.exe 400000'
Benchmark 1: kcas-main/_build/default/test/kcas/benchmark.exe 1 10000
Time (mean ± σ): 3.0 ms ± 0.0 ms [User: 2.0 ms, System: 0.6 ms]
Range (min … max): 2.9 ms … 3.2 ms 996 runs
Benchmark 2: kcas-this/_build/default/test/kcas/benchmark.exe 1 10000
Time (mean ± σ): 3.0 ms ± 0.0 ms [User: 2.1 ms, System: 0.6 ms]
Range (min … max): 2.9 ms … 3.5 ms 919 runs
Warning: Statistical outliers were detected. Consider re-running this benchmark on a quiet system without any interferences from other programs. It might help to use the '--warmup' or '--prepare' options.
Summary
'kcas-main/_build/default/test/kcas/benchmark.exe 1 10000' ran
1.00 ± 0.02 times faster than 'kcas-this/_build/default/test/kcas/benchmark.exe 1 10000'
Benchmark 1: kcas-main/_build/default/test/kcas/benchmark.exe 2 10000
Time (mean ± σ): 8.2 ms ± 0.1 ms [User: 7.2 ms, System: 0.7 ms]
Range (min … max): 8.1 ms … 8.4 ms 363 runs
Benchmark 2: kcas-this/_build/default/test/kcas/benchmark.exe 2 10000
Time (mean ± σ): 8.2 ms ± 0.1 ms [User: 7.3 ms, System: 0.7 ms]
Range (min … max): 8.1 ms … 8.4 ms 362 runs
Summary
'kcas-main/_build/default/test/kcas/benchmark.exe 2 10000' ran
1.00 ± 0.01 times faster than 'kcas-this/_build/default/test/kcas/benchmark.exe 2 10000'
Benchmark 1: kcas-main/_build/default/test/kcas/benchmark.exe 4 10000
Time (mean ± σ): 13.2 ms ± 0.1 ms [User: 12.2 ms, System: 0.7 ms]
Range (min … max): 13.1 ms … 13.4 ms 228 runs
Benchmark 2: kcas-this/_build/default/test/kcas/benchmark.exe 4 10000
Time (mean ± σ): 13.2 ms ± 0.0 ms [User: 12.2 ms, System: 0.7 ms]
Range (min … max): 13.1 ms … 13.5 ms 226 runs
Summary
'kcas-main/_build/default/test/kcas/benchmark.exe 4 10000' ran
1.00 ± 0.01 times faster than 'kcas-this/_build/default/test/kcas/benchmark.exe 4 10000'
Benchmark 1: kcas-main/_build/default/test/kcas/benchmark.exe 1 200000
Time (mean ± σ): 23.1 ms ± 0.1 ms [User: 22.1 ms, System: 0.7 ms]
Range (min … max): 23.0 ms … 23.4 ms 129 runs
Benchmark 2: kcas-this/_build/default/test/kcas/benchmark.exe 1 200000
Time (mean ± σ): 23.1 ms ± 0.1 ms [User: 22.1 ms, System: 0.7 ms]
Range (min … max): 23.0 ms … 23.5 ms 129 runs
Summary
'kcas-main/_build/default/test/kcas/benchmark.exe 1 200000' ran
1.00 ± 0.00 times faster than 'kcas-this/_build/default/test/kcas/benchmark.exe 1 200000'
Benchmark 1: kcas-main/_build/default/test/kcas/benchmark.exe 2 200000
Time (mean ± σ): 126.9 ms ± 0.5 ms [User: 125.6 ms, System: 1.0 ms]
Range (min … max): 126.2 ms … 127.7 ms 23 runs
Benchmark 2: kcas-this/_build/default/test/kcas/benchmark.exe 2 200000
Time (mean ± σ): 127.4 ms ± 0.4 ms [User: 126.1 ms, System: 0.9 ms]
Range (min … max): 126.6 ms … 128.4 ms 23 runs
Summary
'kcas-main/_build/default/test/kcas/benchmark.exe 2 200000' ran
1.00 ± 0.01 times faster than 'kcas-this/_build/default/test/kcas/benchmark.exe 2 200000'
Benchmark 1: kcas-main/_build/default/test/kcas/benchmark.exe 4 200000
Time (mean ± σ): 226.2 ms ± 0.3 ms [User: 224.7 ms, System: 1.1 ms]
Range (min … max): 225.6 ms … 226.8 ms 13 runs
Benchmark 2: kcas-this/_build/default/test/kcas/benchmark.exe 4 200000
Time (mean ± σ): 226.7 ms ± 0.3 ms [User: 225.2 ms, System: 1.1 ms]
Range (min … max): 226.2 ms … 227.1 ms 13 runs
Summary
'kcas-main/_build/default/test/kcas/benchmark.exe 4 200000' ran
1.00 ± 0.00 times faster than 'kcas-this/_build/default/test/kcas/benchmark.exe 4 200000'
The pass to
verifyread-only CMP operations is only necessary when a CMP is followed by a CAS. In all other cases it can be skipped as the CMPs are already verified as part of thedeterminepass.