github
/
ruby
зеркало из https://github.com/github/ruby.git
1
0
Форкнуть 0
Код Задачи Пакеты Проекты Релизы Вики Активность
ruby/benchmark/string_concat.yml

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

52 строки
3.4 KiB
YAML
Исходник Обычный вид История

vm_opt_ltlt: call rb_str_buf_append directly if RHS is a String `rb_str_concat` does a lot of type checking we can easily bypass. ``` | |compare-ruby|built-ruby| |:--------------|-----------:|---------:| |string_concat | 362.007k| 398.965k| | | -| 1.10x| ```
2022-07-06 13:31:54 +03:00
prelude: |
CHUNK = "a" * 64
rb_str_buf_append: add a fast path for ENC_CODERANGE_VALID If the RHS has valid encoding, and both strings have the same encoding, we can use the fast path. However we need to update the LHS coderange. ``` compare-ruby: ruby 3.2.0dev (2022-07-21T14:46:32Z master cdbb9b8555) [arm64-darwin21] built-ruby: ruby 3.2.0dev (2022-07-25T07:25:41Z string-concat-vali.. 11a2772bdd) [arm64-darwin21] warming up... | |compare-ruby|built-ruby| |:-------------------|-----------:|---------:| |binary_concat_7bit | 554.816k| 556.460k| | | -| 1.00x| |utf8_concat_7bit | 556.367k| 555.101k| | | 1.00x| -| |utf8_concat_UTF8 | 412.555k| 556.824k| | | -| 1.35x| ```
2022-07-21 18:08:51 +03:00
UCHUNK = "é" * 32
Benchmark String interpolation across size pools
2023-01-11 00:13:18 +03:00
SHORT = "a" * (GC::INTERNAL_CONSTANTS[:BASE_SLOT_SIZE] / 2)
LONG = "a" * (GC::INTERNAL_CONSTANTS[:BASE_SLOT_SIZE] * 2)
string.c: use str_enc_fastpath in TERM_LEN Not having to fetch the rb_encoding save a significant amount of time. Additionally, even when we have to fetch it, we can do it faster using `ENCODING_GET` rather than `rb_enc_get`. ``` compare-ruby: ruby 3.2.0dev (2022-07-19T08:41:40Z master cb9fd920a3) [arm64-darwin21] built-ruby: ruby 3.2.0dev (2022-07-21T11:16:16Z faster-buffer-conc.. 4f001f0748) [arm64-darwin21] warming up... | |compare-ruby|built-ruby| |:---------------------|-----------:|---------:| |binary_concat_utf8 | 510.580k| 565.600k| | | -| 1.11x| |binary_concat_binary | 512.653k| 571.483k| | | -| 1.11x| |utf8_concat_utf8 | 511.396k| 566.879k| | | -| 1.11x| ```
2022-07-21 14:16:16 +03:00
GC.disable # GC causes a lot of variance
vm_opt_ltlt: call rb_str_buf_append directly if RHS is a String `rb_str_concat` does a lot of type checking we can easily bypass. ``` | |compare-ruby|built-ruby| |:--------------|-----------:|---------:| |string_concat | 362.007k| 398.965k| | | -| 1.10x| ```
2022-07-06 13:31:54 +03:00
benchmark:
rb_str_buf_append: add a fast path for ENC_CODERANGE_VALID If the RHS has valid encoding, and both strings have the same encoding, we can use the fast path. However we need to update the LHS coderange. ``` compare-ruby: ruby 3.2.0dev (2022-07-21T14:46:32Z master cdbb9b8555) [arm64-darwin21] built-ruby: ruby 3.2.0dev (2022-07-25T07:25:41Z string-concat-vali.. 11a2772bdd) [arm64-darwin21] warming up... | |compare-ruby|built-ruby| |:-------------------|-----------:|---------:| |binary_concat_7bit | 554.816k| 556.460k| | | -| 1.00x| |utf8_concat_7bit | 556.367k| 555.101k| | | 1.00x| -| |utf8_concat_UTF8 | 412.555k| 556.824k| | | -| 1.35x| ```
2022-07-21 18:08:51 +03:00
binary_concat_7bit: |
rb_str_buf_append: fastpath to str_buf_cat If the LHS is ASCII compatible and the RHS is 7BIT we can directly concat without being concerned about anything else. Benchmark: ``` compare-ruby: ruby 3.2.0dev (2022-07-12T15:01:11Z master 71aec68566) [arm64-darwin21] built-ruby: ruby 3.2.0dev (2022-07-13T10:13:53Z faster-buffer-conc.. a04c10476d) [arm64-darwin21] warming up... | |compare-ruby|built-ruby| |:---------------------|-----------:|---------:| |binary_append_utf8 | 385.315k| 573.663k| | | -| 1.49x| |binary_append_binary | 446.579k| 574.898k| | | -| 1.29x| |utf8_append_utf8 | 430.936k| 573.394k| | | -| 1.33x| ``` Note that in the benchmark, the RHS always have a precomputed coderange. So the benchmark never enter the slowpath of having to scan the RHS. However it's extremly likely that we'll end up scanning it anyway in rb_enc_cr_str_buf_cat
2022-06-29 11:35:41 +03:00
buffer = String.new(capacity: 4096, encoding: Encoding::BINARY)
buffer << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK
buffer << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK
buffer << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK
buffer << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK
buffer << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK
buffer << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK
buffer << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK
string.c: use str_enc_fastpath in TERM_LEN Not having to fetch the rb_encoding save a significant amount of time. Additionally, even when we have to fetch it, we can do it faster using `ENCODING_GET` rather than `rb_enc_get`. ``` compare-ruby: ruby 3.2.0dev (2022-07-19T08:41:40Z master cb9fd920a3) [arm64-darwin21] built-ruby: ruby 3.2.0dev (2022-07-21T11:16:16Z faster-buffer-conc.. 4f001f0748) [arm64-darwin21] warming up... | |compare-ruby|built-ruby| |:---------------------|-----------:|---------:| |binary_concat_utf8 | 510.580k| 565.600k| | | -| 1.11x| |binary_concat_binary | 512.653k| 571.483k| | | -| 1.11x| |utf8_concat_utf8 | 511.396k| 566.879k| | | -| 1.11x| ```
2022-07-21 14:16:16 +03:00
buffer << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK
rb_str_buf_append: add a fast path for ENC_CODERANGE_VALID If the RHS has valid encoding, and both strings have the same encoding, we can use the fast path. However we need to update the LHS coderange. ``` compare-ruby: ruby 3.2.0dev (2022-07-21T14:46:32Z master cdbb9b8555) [arm64-darwin21] built-ruby: ruby 3.2.0dev (2022-07-25T07:25:41Z string-concat-vali.. 11a2772bdd) [arm64-darwin21] warming up... | |compare-ruby|built-ruby| |:-------------------|-----------:|---------:| |binary_concat_7bit | 554.816k| 556.460k| | | -| 1.00x| |utf8_concat_7bit | 556.367k| 555.101k| | | 1.00x| -| |utf8_concat_UTF8 | 412.555k| 556.824k| | | -| 1.35x| ```
2022-07-21 18:08:51 +03:00
utf8_concat_7bit: |
rb_str_buf_append: fastpath to str_buf_cat If the LHS is ASCII compatible and the RHS is 7BIT we can directly concat without being concerned about anything else. Benchmark: ``` compare-ruby: ruby 3.2.0dev (2022-07-12T15:01:11Z master 71aec68566) [arm64-darwin21] built-ruby: ruby 3.2.0dev (2022-07-13T10:13:53Z faster-buffer-conc.. a04c10476d) [arm64-darwin21] warming up... | |compare-ruby|built-ruby| |:---------------------|-----------:|---------:| |binary_append_utf8 | 385.315k| 573.663k| | | -| 1.49x| |binary_append_binary | 446.579k| 574.898k| | | -| 1.29x| |utf8_append_utf8 | 430.936k| 573.394k| | | -| 1.33x| ``` Note that in the benchmark, the RHS always have a precomputed coderange. So the benchmark never enter the slowpath of having to scan the RHS. However it's extremly likely that we'll end up scanning it anyway in rb_enc_cr_str_buf_cat
2022-06-29 11:35:41 +03:00
buffer = String.new(capacity: 4096, encoding: Encoding::UTF_8)
vm_opt_ltlt: call rb_str_buf_append directly if RHS is a String `rb_str_concat` does a lot of type checking we can easily bypass. ``` | |compare-ruby|built-ruby| |:--------------|-----------:|---------:| |string_concat | 362.007k| 398.965k| | | -| 1.10x| ```
2022-07-06 13:31:54 +03:00
buffer << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK
buffer << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK
buffer << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK
buffer << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK
buffer << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK
buffer << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK
buffer << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK
string.c: use str_enc_fastpath in TERM_LEN Not having to fetch the rb_encoding save a significant amount of time. Additionally, even when we have to fetch it, we can do it faster using `ENCODING_GET` rather than `rb_enc_get`. ``` compare-ruby: ruby 3.2.0dev (2022-07-19T08:41:40Z master cb9fd920a3) [arm64-darwin21] built-ruby: ruby 3.2.0dev (2022-07-21T11:16:16Z faster-buffer-conc.. 4f001f0748) [arm64-darwin21] warming up... | |compare-ruby|built-ruby| |:---------------------|-----------:|---------:| |binary_concat_utf8 | 510.580k| 565.600k| | | -| 1.11x| |binary_concat_binary | 512.653k| 571.483k| | | -| 1.11x| |utf8_concat_utf8 | 511.396k| 566.879k| | | -| 1.11x| ```
2022-07-21 14:16:16 +03:00
buffer << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK << CHUNK
rb_str_buf_append: add a fast path for ENC_CODERANGE_VALID If the RHS has valid encoding, and both strings have the same encoding, we can use the fast path. However we need to update the LHS coderange. ``` compare-ruby: ruby 3.2.0dev (2022-07-21T14:46:32Z master cdbb9b8555) [arm64-darwin21] built-ruby: ruby 3.2.0dev (2022-07-25T07:25:41Z string-concat-vali.. 11a2772bdd) [arm64-darwin21] warming up... | |compare-ruby|built-ruby| |:-------------------|-----------:|---------:| |binary_concat_7bit | 554.816k| 556.460k| | | -| 1.00x| |utf8_concat_7bit | 556.367k| 555.101k| | | 1.00x| -| |utf8_concat_UTF8 | 412.555k| 556.824k| | | -| 1.35x| ```
2022-07-21 18:08:51 +03:00
utf8_concat_UTF8: |
buffer = String.new(capacity: 4096, encoding: Encoding::UTF_8)
buffer << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK
buffer << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK
buffer << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK
buffer << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK
buffer << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK
buffer << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK
buffer << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK
buffer << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK << UCHUNK
rb_str_concat_literals: use rb_str_buf_append That's about 1.30x faster.
2022-09-08 13:43:03 +03:00
interpolation: |
buffer = "#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}" \
"#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}" \
"#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}" \
"#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}" \
"#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}" \
"#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}" \
"#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}" \
"#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}" \
"#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}#{CHUNK}"
Benchmark String interpolation across size pools
2023-01-11 00:13:18 +03:00
interpolation_same_size_pool: |
buffer = "#{SHORT}#{SHORT}"
interpolation_switching_size_pools: |
buffer = "#{SHORT}#{LONG}"