Optimize masking with math/bits #171
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
See golang/go#31586 (comment) Thanks @renthraysk benchmark old MB/s new MB/s speedup BenchmarkXOR/2/fast-8 470.88 492.61 1.05x BenchmarkXOR/3/fast-8 602.24 719.25 1.19x BenchmarkXOR/4/fast-8 718.82 1186.64 1.65x BenchmarkXOR/8/fast-8 1027.60 1718.71 1.67x BenchmarkXOR/16/fast-8 1413.31 3430.46 2.43x BenchmarkXOR/32/fast-8 2701.81 5585.42 2.07x BenchmarkXOR/128/fast-8 7757.97 13432.37 1.73x BenchmarkXOR/512/fast-8 15155.03 18797.79 1.24x BenchmarkXOR/4096/fast-8 20689.95 20334.61 0.98x BenchmarkXOR/16384/fast-8 21687.87 21613.94 1.00x Now its faster than basic XOR at every byte size greater than 2 on little endian amd64 machines.
|
More optimization is possible, see golang/go#31586 (comment) |
Thanks again to @renthraysk This provides another significant speedup. benchmark old MB/s new MB/s speedup Benchmark_mask/2/fast-8 405.48 513.25 1.27x Benchmark_mask/3/fast-8 518.93 661.92 1.28x Benchmark_mask/4/fast-8 1207.10 1252.39 1.04x Benchmark_mask/8/fast-8 1708.82 1655.63 0.97x Benchmark_mask/16/fast-8 3418.58 3051.25 0.89x Benchmark_mask/32/fast-8 5789.43 5813.31 1.00x Benchmark_mask/128/fast-8 12819.53 14804.50 1.15x Benchmark_mask/512/fast-8 18247.06 21659.50 1.19x Benchmark_mask/4096/fast-8 19802.31 23885.68 1.21x Benchmark_mask/16384/fast-8 20896.97 25081.11 1.20x
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
See golang/go#31586 (comment)
Thanks @renthraysk
Now its faster than basic XOR at every byte size greater than 2 on little
endian amd64 machines.
And faster at every level than gobwas/ws and gorilla/websocket.