Feature #17670
Updated by S_H_ (Shun Hiraoka) about 3 years ago
Improve performance `Float#to_i` and `Float#to_int` methods(write in Ruby)
benchmark:
```yml
prelude: |
flo = 4.2
benchmark:
to_i: |
flo.to_i
to_int: |
flo.to_int
loop_count: 20000000
```
result:
```bash
sh@DESKTOP-L0NI312:~/rubydev/build$ make benchmark/benchmark.yml -e COMPARE_RUBY=~/.rbenv/shims/ruby -e BENCH_RUBY=../install/bin/ruby
generating vm_call_iseq_optimized.inc
vm_call_iseq_optimized.inc unchanged
compare-ruby: ruby 3.1.0dev (2021-02-28T11:24:42Z master 80e2c45f55) [x86_64-linux]
built-ruby: ruby 3.1.0dev (2021-02-28T11:24:42Z improve_float_to_i 80e2c45f55) [x86_64-linux]
# Iteration per second (i/s)
| |compare-ruby|built-ruby|
|:-------|-----------:|---------:|
|to_i | 69.184M| 78.294M|
| | -| 1.13x|
|to_int | 69.891M| 80.285M|
| | -| 1.15x|
```
`COMPARE_RUBY` is `ruby 3.1.0dev (2021-02-28T11:24:42Z master 80e2c45f55) [x86_64-linux]`. `BENCH_RUBY` is ahead of `ruby 3.1.0dev (2021-02-28T11:24:42Z master 80e2c45f55) [x86_64-linux]`.
pull request: https://github.com/ruby/ruby/pull/4234