CI: Add benchmark scripts (#4989)

**Which issue(s) this PR fixes**: 
Fixes #

**What this PR does / why we need it**: 
This PR adds benchmark that measures performance using the `in_tail`
plugin.
It would be good if performance regression can be found in the future...

We can check the results on the summary page in Benchmark workflow.
https://github.com/fluent/fluentd/actions/runs/15407283089?pr=4989

**Docs Changes**:
Not needed.

**Release Note**: 
Same as the title.

---------

Signed-off-by: Shizuo Fujita <fujita@clear-code.com>

Author: Watson1978

.github/workflows/benchmark.yml

@@ -0,0 +1,38 @@
+name: Benchmark
+
+on:
+  push:
+    branches: [master]
+  pull_request:
+    branches: [master]
+  workflow_dispatch:
+
+permissions: read-all
+
+concurrency:
+  group: ${{ github.head_ref || github.sha }}-${{ github.workflow }}
+  cancel-in-progress: true
+
+jobs:
+  test:
+    runs-on: ${{ matrix.os }}
+    continue-on-error: false
+    strategy:
+      fail-fast: false
+      matrix:
+        os: ['ubuntu-latest', 'macos-latest', 'windows-latest']
+        ruby-version: ['3.4']
+
+    name: Ruby ${{ matrix.ruby-version }} on ${{ matrix.os }}
+    steps:
+      - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
+      - name: Set up Ruby
+        uses: ruby/setup-ruby@13e7a03dc3ac6c3798f4570bfead2aed4d96abfb # v1.244.0
+        with:
+          ruby-version: ${{ matrix.ruby-version }}
+      - name: Install dependencies
+        run: bundle install
+      - name: Run Benchmark
+        shell: bash # Ensure to use bash shell on all platforms
+        run: |
+          bundle exec rake benchmark:run:in_tail | tee -a $GITHUB_STEP_SUMMARY

.gitignore

@@ -16,6 +16,7 @@ fluent-cat
 fluent-gem
 fluentd
 pkg/*
+tmp/*
 test/tmp/*
 test/config/tmp/*
 make_dist.sh

Gemfile

@@ -2,6 +2,8 @@ source 'https://rubygems.org/'
 
 gemspec
 
+gem 'benchmark'
+
 local_gemfile = File.join(File.dirname(__FILE__), "Gemfile.local")
 if File.exist?(local_gemfile)
   puts "Loading Gemfile.local ..." if $DEBUG # `ruby -d` or `bundle -v`

Rakefile

@@ -5,6 +5,8 @@ require 'fileutils'
 require 'rake/testtask'
 require 'rake/clean'
 
+require_relative 'tasks/benchmark'
+
 task test: [:base_test]
 
 # 1. update ChangeLog and lib/fluent/version.rb

fluentd.gemspec

@@ -13,7 +13,7 @@ Gem::Specification.new do |gem|
   gem.files         = Dir.chdir(__dir__) do
     `git ls-files -z`.split("\x0").reject do |f|
       (File.expand_path(f) == __FILE__) ||
-        f.start_with?(*%w[test/ .git Gemfile])
+        f.start_with?(*%w[tasks/ test/ .git Gemfile])
     end
   end
   gem.executables   = gem.files.grep(%r{^bin/}).map{ |f| File.basename(f) }

tasks/benchmark.rb

@@ -0,0 +1,48 @@
+require "json"
+require "fileutils"
+
+BENCHMARK_FILE_SIZE = 1 * 1024 * 1024 * 1024
+BENCHMARK_FILE_PATH = File.expand_path("./tmp/benchmark/data.log")
+
+namespace :benchmark do
+  task :init do
+    # Synchronize stdout because the output order is not as intended on Windows environment
+    STDOUT.sync = true
+  end
+
+  task :prepare_1GB do
+    FileUtils.mkdir_p(File.dirname(BENCHMARK_FILE_PATH))
+    File.open(BENCHMARK_FILE_PATH, "w") do |f|
+      data = { "message": "a" * 1024 }.to_json
+
+      loop do
+        f.puts data
+        break if File.size(BENCHMARK_FILE_PATH) > BENCHMARK_FILE_SIZE
+      end
+    end
+  end
+
+  task :show_info do
+    # Output the information with markdown format
+    puts "### Environment"
+    puts "```"
+    system "bundle exec ruby --version"
+    system "bundle exec ruby bin/fluentd --version"
+    puts "```\n"
+  end
+
+  desc "Run in_tail benchmark"
+  task :"run:in_tail" => [:init, :prepare_1GB, :show_info] do
+    # Output the results with markdown format
+    puts "### in_tail with 1 GB file"
+    puts "```"
+    system "bundle exec ruby bin/fluentd -r ./tasks/benchmark/patch_in_tail.rb --no-supervisor -c ./tasks/benchmark/conf/in_tail.conf -o ./tmp/benchmark/fluent.log"
+    puts "```"
+
+    Rake::Task["benchmark:clean"].invoke
+  end
+
+  task :clean do
+    FileUtils.rm_rf(File.dirname(BENCHMARK_FILE_PATH))
+  end
+end

tasks/benchmark/conf/in_tail.conf

@@ -0,0 +1,14 @@
+<source>
+  @type tail
+  tag benchmark
+  path "#{File.expand_path './tmp/benchmark/data.log'}"
+  read_from_head true
+  <parse>
+    @type json
+  </parse>
+</source>
+
+<match **>
+  @type file
+  path "#{File.expand_path './tmp/benchmark/in_tail'}"
+</match>

tasks/benchmark/patch_in_tail.rb

@@ -0,0 +1,21 @@
+require 'benchmark'
+require 'fluent/plugin/in_tail'
+
+class Fluent::Plugin::TailInput::TailWatcher::IOHandler
+  alias_method :original_with_io, :with_io
+
+  def with_io(&block)
+    @benchmark_measured_in_tail ||= false
+    # Measure the benchmark only once.
+    return original_with_io(&block) if @benchmark_measured_in_tail
+
+    Benchmark.bm do |x|
+      x.report {
+        original_with_io(&block)
+        @benchmark_measured_in_tail = true
+      }
+    end
+
+    exit 0
+  end
+end