Use size_t for RSTRING_LEN in String#count

[nobu]

https://hackerone.com/reports/1042722

[ioquatix]

Would it make sense to standardise on size_t for this kind of use case?

[ilyazub]

String#count became 3.4 times slower after this PR: https://www.reddit.com/r/ruby/comments/16d4ha6/comment/k04ohqu/?utm_source=share&utm_medium=web2x&context=3

Benchmark 1: ./ruby test.rb
  Time (mean ± σ):      95.5 ms ±   1.2 ms    [User: 93.9 ms, System: 1.9 ms]
  Range (min … max):    93.4 ms …  98.5 ms    30 runs

Benchmark 2: ./ruby.master test.rb
  Time (mean ± σ):     326.5 ms ±   2.0 ms    [User: 323.1 ms, System: 3.6 ms]
  Range (min … max):   324.2 ms … 329.1 ms    10 runs

Summary
  './ruby test.rb' ran
    3.42 ± 0.05 times faster than './ruby.master test.rb'

[Freaky]

Slightly-tweaked test.rb that demonstrates the difference just as well:

html = IO.read('/dev/zero', 1024 * 1024)
Array.new(1000) { html.count($/) }

AMD 5700X, FreeBSD 13.2 (clang 14.0.5), tested against b1f0d00 and same with this patch reverted.

[ilyazub]

Here're the benchmark results using the benchmark-driver gem.

$ benchmark-driver tmp/string_count_benchmark_driver.yml --rbenv '3.1.1;3.1.4;2.7.2;3.2.2;3.0.6'                                                 
Calculating -------------------------------------
                          3.1.1       3.1.4       2.7.2       3.2.2       3.0.6
               count    465.804     463.741     865.783     462.711     857.395 i/s -     10.000k times in 21.468251s 21.563768s 11.550239s 21.611783s 11.663235s

Comparison:
                            count
               2.7.2:       865.8 i/s 
               3.0.6:       857.4 i/s - 1.01x  slower
               3.1.1:       465.8 i/s - 1.86x  slower
               3.1.4:       463.7 i/s - 1.87x  slower
               3.2.2:       462.7 i/s - 1.87x  slower

Benchmark:

# cat tmp/string_count_benchmark_driver.yml
loop_count: 10_000
prelude: |
  html = IO.read('/dev/zero', 1024 * 1024)
benchmark:
  count: html.count($/)

i7-6700HQ CPU @ 2.60GHz, Debian 10 (gcc 8.3.0)

[ilyazub]

I created a Redmine ticket: https://bugs.ruby-lang.org/issues/19875