From: "Freaky (Thomas Hurst) via ruby-core" <ruby-core@...>
Date: 2023-09-27T18:52:30+00:00
Subject: [ruby-core:114906] [Ruby master Bug#19875] Ruby 3.0 -> 3.1 Performance regression in String#count

Issue #19875 has been updated by Freaky (Thomas Hurst).





ahorek (Pavel Rosick�) wrote in #note-18:

> I think if the platform support is sufficient, there are multiple places where this feature could be beneficial.



I think the platform support covers the major bases.  I'm testing a PHP build with ifunc and target support enabled, after it was disabled for FreeBSD in 2018 and nobody thought to try turning it back on.



> Even without explicit SIMD code, some existing C code could be autovectorized by the compiler itself for the AVX2 target while keeping the binary compatibility with platforms that don't support these instructions.



Yes - `__attribute__((target_clones(..))` makes this particularly easy to do -- the difficulty lies in using it appropriately.



> You can always recompile Ruby with optimization flags for your own platform, but most users use precompiled binaries that can't benefit from it now.



It would be interesting to see some benchmarks comparing Ruby (and other things) targetting basline x86-64, and the -v2, -v3, and -v4 profiles.



> could prepare a proof of concept PR for the Ruby repo?



This is what I'm working towards, yes.



> also, see the previous discussion #16487 there are already other existing examples where this feature could help.



Thanks.



----------------------------------------

Bug #19875: Ruby 3.0 -> 3.1 Performance regression in String#count

https://bugs.ruby-lang.org/issues/19875#change-104772



* Author: iz (Illia Zub)

* Status: Open

* Priority: Normal

* ruby -v: 3.2.2

* Backport: 3.0: UNKNOWN, 3.1: UNKNOWN, 3.2: UNKNOWN

----------------------------------------

`String#count` became slower since Ruby 3.1. Originally found by `@Freaky`: https://github.com/ruby/ruby/pull/4001#issuecomment-1714779781



Compared using the [`benchmark-driver` gem](https://github.com/benchmark-driver/benchmark-driver).



```

$ 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:





```yml

$ cat ./tmp/string_count_benchmark_driver.yml 

loop_count: 10_000

prelude: |

  html = "\nruby\n" * 1024 * 1024

benchmark:

  count: html.count($/)

```



---



*Initially, I noticed the difference between `str.count($/)` and `str.lines.size` when working on the performance improvement: https://serpapi.com/blog/lines-count-failed-deployments/*



---Files--------------------------------

rb_str_len.fast (31.9 KB)

rb_str_len.slow (34 KB)

revert-4001.patch (1.71 KB)

rb_str_count.S (11.8 KB)

bytecount.c (7.23 KB)





-- 

https://bugs.ruby-lang.org/

 ______________________________________________
 ruby-core mailing list -- ruby-core@ml.ruby-lang.org
 To unsubscribe send an email to ruby-core-leave@ml.ruby-lang.org
 ruby-core info -- https://ml.ruby-lang.org/mailman3/postorius/lists/ruby-core.ml.ruby-lang.org/