rafaelfranca · Nov 21, 2025 · Nov 21, 2025 · Nov 24, 2025 · Nov 27, 2025 · Nov 27, 2025
diff --git a/.github/workflows/test.yml b/.github/workflows/test.yml
@@ -152,8 +152,12 @@ jobs:
         with:
           ruby-version: head
 
-      - name: Test run_once.sh
-        run: ./run_once.sh --yjit-stats benchmarks/railsbench/benchmark.rb
+      - name: Run a benchmark once directly with Ruby
+        run: ruby --yjit-stats benchmarks/railsbench/benchmark.rb
+        env:
+          WARMUP_ITRS: '0'
+          MIN_BENCH_ITRS: '1'
+          MIN_BENCH_TIME: '0'
 
   test-windows:
     runs-on: windows-latest

diff --git a/README.md b/README.md
@@ -6,7 +6,7 @@ Small set of benchmarks and scripts for the Ruby programming language.
 The benchmarks are found in the `benchmarks` directory. Individual Ruby files
 in `benchmarks` are microbenchmarks. Subdirectories under `benchmarks` are
 larger macrobenchmarks. Each benchmark relies on a harness found in
-[./harness/harness.rb](harness/harness.rb). The harness controls the number of times a benchmark is
+[./harness/default.rb](harness/default.rb). The harness controls the number of times a benchmark is
 run, and writes timing values into an output file.
 
 The `run_benchmarks.rb` script (optional) traverses the `benchmarks` directory and
@@ -97,8 +97,8 @@ This is the easiest way to run a single benchmark.
 It requires no setup at all and assumes nothing about the Ruby you are benchmarking.
 It's also convenient for profiling, debugging, etc, especially since all benchmarked code runs in that process.
 
-```
-ruby benchmarks/some_benchmark.rb
+```bash
+ruby benchmarks/fib.rb
 ```
 
 ### Benchmark organization
@@ -138,19 +138,19 @@ There are two Ractor-related categories:
 
 * **`--category ractor`** - Runs both regular benchmarks marked with `ractor:
   true` in `benchmarks.yml` AND all benchmarks from the `benchmarks-ractor`
-  directory. The `harness-ractor` harness is used for both types of benchmark.
+  directory. The `ractor` harness is used for both types of benchmark.
 
 * **`--category ractor-only`** - Runs ONLY benchmarks from the
   `benchmarks-ractor` directory, ignoring regular benchmarks even if they are
   marked with `ractor: true`. This category also automatically uses the
-  `harness-ractor` harness.
+  `ractor` harness.
 
 ### Directory Structure
 
 The `benchmarks-ractor/` directory sits at the same level as the main
 `benchmarks` directory, and contains Ractor-specific benchmark
 implementations that are designed to test Ractor functionality. They are not
-intended to be used with any harness except `harness-ractor`.
+intended to be used with any harness except `ractor`.
 
 ### Usage Examples
 
@@ -162,7 +162,7 @@ intended to be used with any harness except `harness-ractor`.
 ./run_benchmarks.rb --category ractor-only
 ```
 
-Note: The `harness-ractor` harness is automatically selected when using these
+Note: The `ractor` harness is automatically selected when using these
 categories, so there's no need to specify `--harness` manually.
 
 ### Ractor counts
@@ -188,12 +188,12 @@ because no single process ran all counts.
 The text summary, the CSV output, and `misc/zjit_diff.rb` show one row for
 each count, with the RSS and JIT stats of that count's process.
 
-When you run a benchmark directly with `-Iharness-ractor`, the harness runs
+When you run a benchmark directly with `-r./harness/ractor`, the harness runs
 all counts in one process, one count after another.
 
 ### Ractor Scenario Benchmarks
 
-By default, `harness-ractor` spawns the worker Ractors and runs the benchmark
+By default, the `ractor` harness spawns the worker Ractors and runs the benchmark
 block inside each of them. A benchmark that calls
 `run_benchmark(n, scenario: true)` uses scenario mode instead. The harness
 calls the block one time per trial in the main Ractor, with the Ractor count as
@@ -232,7 +232,7 @@ with multiple ractors, for GC work that reclaims ractor-local memory:
 
 ```bash
 # Every count in one process
-ruby -Iharness-ractor benchmarks/ractor-dead-set/benchmark.rb
+ruby -r./harness/ractor benchmarks/ractor-dead-set/benchmark.rb
 
 # One process per count, comparing two Rubies, with Ractor GC metrics
 RUBY_BENCH_RACTORS=1,2,4,6,8 ./run_benchmarks.rb --category ractor --ractor-gc \
@@ -321,26 +321,62 @@ This file will then be passed to the underlying Ruby interpreter with
 
 ## Harnesses
 
-You can find several test harnesses in this repository:
+You can find several test harnesses in the `harness/` directory:
+
+* `default` - the normal default harness, with duration controlled by warmup iterations and time/count limits
+* `bips` - a harness that measures iterations/second until stable
+* `continuous` - a harness that adjusts the batch sizes of iterations to run in stable iteration size batches
+* `once` - a simplified harness that simply runs once
+* `perf` - a simplified harness that runs for exactly the hinted number of iterations
+* `ractor` - a harness that runs the benchmark inside Ractors to measure scaling
+* `gc` - the default harness plus GC timing, count and heap statistics
+* `stackprof` - a harness to profile the benchmark with stackprof
+* `stats` - count method calls and loop iterations
+* `vernier` - a harness to profile the benchmark with vernier
+* `warmup` - a harness which runs as long as needed to find warmed up (peak) performance
+* `chain` - a harness to chain multiple harnesses together
+* `mplr` - a harness for multiple iterations with time limits
 
-* harness - the normal default harness, with duration controlled by warmup iterations and time/count limits
-* harness-bips - a harness that measures iterations/second until stable
-* harness-continuous - a harness that adjusts the batch sizes of iterations to run in stable iteration size batches
-* harness-once - a simplified harness that simply runs once
-* harness-perf - a simplified harness that runs for exactly the hinted number of iterations
-* harness-stackprof - a harness to profile the benchmark with stackprof
-* harness-stats - count method calls and loop iterations
-* harness-vernier - a harness to profile the benchmark with vernier
-* harness-warmup - a harness which runs as long as needed to find warmed up (peak) performance
+### Selecting a harness
 
-To use it, run a benchmark script directly, specifying a harness directory with `-I`:
+**Using the `HARNESS` environment variable**
+
+```bash
+# Run with specific harness
+HARNESS=perf ruby benchmarks/fib.rb
+HARNESS=stackprof ruby benchmarks/railsbench/benchmark.rb
+HARNESS=vernier ruby benchmarks/optcarrot/benchmark.rb
 
+# Combine with Ruby options
+HARNESS=perf ruby --yjit benchmarks/fib.rb
+HARNESS=once ruby --yjit-stats benchmarks/railsbench/benchmark.rb
 ```
-ruby -Iharness benchmarks/railsbench/benchmark.rb
+
+When using `run_benchmarks.rb`, you can specify a harness with the `--harness` option:
+
+```bash
+./run_benchmarks.rb --harness=once
+./run_benchmarks.rb --harness=perf
 ```
 
 There is also a robust but complex CI harness in [the yjit-metrics repo](https://github.com/Shopify/yjit-metrics).
 
+### Chain harness
+
+The `chain` harness allows you to combine multiple harnesses together. This is useful when you want to run a benchmark through an analysis tool, such as a profiler, while another harness drives the iterations.
+
+Use the `HARNESS_CHAIN` environment variable to specify which harnesses to chain (comma-separated, at least 2 required):
+
+```bash
+# Profile the iterations of the default harness with vernier
+HARNESS=chain HARNESS_CHAIN="vernier,default" ruby benchmarks/fib.rb
+
+# Profile a benchmark running under the ractor harness with vernier
+HARNESS=chain HARNESS_CHAIN="vernier,ractor" ruby benchmarks/fib.rb
+```
+
+The harnesses are nested, not run one after another. The first harness is the outermost one: its benchmark block calls the `run_benchmark` of the next harness, and the last harness runs the benchmark itself. Every harness writes its results to the same file, and the outermost harness writes last, so its results are the ones kept.
+
 ### Iterations and duration
 
 With the default harness, the number of iterations and duration
@@ -360,25 +396,24 @@ You can also use `--warmup`, `--bench`, or `--once` to set these environment var
 ./run_benchmarks.rb railsbench --once
 ```
 
-There is also a handy script for running benchmarks just once using
-`WARMUP_ITRS=0 MIN_BENCH_ITRS=1 MIN_BENCH_TIME=0`, for example
-with the `--yjit-stats` command-line option:
+To run a single benchmark once directly with Ruby, set the same variables as `--once`:
 
-```
-./run_once.sh --yjit-stats benchmarks/railsbench/benchmark.rb
+```bash
+# Run once with YJIT stats
+WARMUP_ITRS=0 MIN_BENCH_ITRS=1 MIN_BENCH_TIME=0 ruby --yjit-stats benchmarks/railsbench/benchmark.rb
 ```
 
 ### Using perf
 
 There is also a harness to use Linux perf. By default, it only runs a fixed number of iterations.
 If `PERF` environment variable is present, it starts the perf subcommand after warmup.
 
-```sh
+```bash
 # Use `perf record` for both warmup and benchmark
-perf record ruby --yjit-perf=map -Iharness-perf benchmarks/railsbench/benchmark.rb
+HARNESS=perf perf record ruby --yjit-perf=map benchmarks/railsbench/benchmark.rb
 
 # Use `perf record` only for benchmark
-PERF=record ruby --yjit-perf=map -Iharness-perf benchmarks/railsbench/benchmark.rb
+HARNESS=perf PERF=record ruby --yjit-perf=map benchmarks/railsbench/benchmark.rb
 ```
 
 This is the only harness that uses `run_benchmark`'s argument, `num_itrs_hint`.
@@ -390,14 +425,14 @@ Linux that shows its profiles in the Firefox Profiler. Run the benchmark script
 directly under `samply record`:
 
 ```sh
-samply record ruby -Iharness benchmarks/railsbench/benchmark.rb
+samply record ruby benchmarks/railsbench/benchmark.rb
 
 # Name JIT-compiled methods. Both JITs write /tmp/perf-<pid>.map, which samply reads.
-samply record ruby --yjit --yjit-perf -Iharness benchmarks/railsbench/benchmark.rb
-samply record ruby --zjit --zjit-perf -Iharness benchmarks/railsbench/benchmark.rb
+samply record ruby --yjit --yjit-perf benchmarks/railsbench/benchmark.rb
+samply record ruby --zjit --zjit-perf benchmarks/railsbench/benchmark.rb
 
 # Save the profile without opening it, and open it later
-samply record --save-only -o railsbench.json.gz ruby -Iharness benchmarks/railsbench/benchmark.rb
+samply record --save-only -o railsbench.json.gz ruby benchmarks/railsbench/benchmark.rb
 samply load railsbench.json.gz
 ```
 
@@ -412,7 +447,7 @@ benchmark, also set `RUBY_BENCH_PROFILING=1` to keep the harness's retention
 GCs out of the profile. The worker Ractors show up as separate thread tracks.
 
 ```sh
-RUBY_BENCH_RACTORS=4 RUBY_BENCH_PROFILING=1 samply record ruby -Iharness-ractor benchmarks/ractor-dead-set/benchmark.rb
+RUBY_BENCH_RACTORS=4 RUBY_BENCH_PROFILING=1 samply record ruby -r./harness/ractor benchmarks/ractor-dead-set/benchmark.rb
 ```
 
 samply also records child processes, so `samply record ruby run_benchmarks.rb ...`
@@ -465,7 +500,7 @@ fail before warmup.
 ./run_benchmarks.rb --category ractor --chruby=base::ruby-base --ractor-gc
 
 # One benchmark, run directly. --ractor-gc sets RUBY_BENCH_RACTOR_GC=1.
-RUBY_BENCH_RACTOR_GC=1 ruby -Iharness-ractor benchmarks/ractor-dead-set/benchmark.rb
+RUBY_BENCH_RACTOR_GC=1 ruby -r./harness/ractor benchmarks/ractor-dead-set/benchmark.rb
 ```
 
 `run_benchmarks.rb` only runs Ractor harness benchmarks with `--category ractor`

diff --git a/benchmarks.yml b/benchmarks.yml
@@ -81,7 +81,7 @@ gcbench:
   desc: Ellis-Kovac-Boehm GCBench builds binary trees of various depths to exercise GC marking, sweeping, and write barriers.
   category: gc
   single_file: true
-  default_harness: harness-gc
+  default_harness: gc
 erubi:
   desc: erubi compiles a simple Erb template into a method with erubi, then times evaluating that method.
   ractor: true
@@ -105,7 +105,7 @@ knucleotide-ractor:
   desc: k-nucleotide from the Computer Language Benchmarks Game - counts nucleotide frequencies using hash tables in parallel using Ractors
   ractor: true
   ractor_only: true
-  default_harness: harness
+  default_harness: default
 lee:
   desc: lee is a circuit-board layout solver, deployed in a plausibly reality-like way
 matmul:
@@ -145,7 +145,7 @@ string_malloc_pressure:
   desc: Allocates and retains many empty Strings with large reserved capacity to exercise GC under malloc-backed String buffer pressure.
   category: gc
   single_file: true
-  default_harness: harness-gc
+  default_harness: gc
 tinygql:
   desc: TinyGQL gem parsing a large file in pure Ruby
 
@@ -285,37 +285,37 @@ gvl_release_acquire:
   desc: microbenchmark designed to test how fast the gvl can be acquired and released between ractors.
   ractor: true
   ractor_only: true
-  default_harness: harness-ractor
+  default_harness: ractor
 json_parse_float:
   desc: test the performance of parsing multiple lists of json floats with ractors.
   ractor: true
   ractor_only: true
-  default_harness: harness-ractor
+  default_harness: ractor
 json_parse_string:
   desc: test the performance of parsing multiple lists of strings with ractors.
   ractor: true
   ractor_only: true
-  default_harness: harness-ractor
+  default_harness: ractor
 ractor-dead-set:
   desc: multiple ractors each build a large live set and terminate, then memory retention after their death is measured.
   ractor: true
   ractor_only: true
   ractor_scenario: true
-  default_harness: harness-ractor
+  default_harness: ractor
 ractor-idle-garbage:
   desc: multiple ractors each build and drop a large garbage set, then idle uncollected while retention is measured.
   ractor: true
   ractor_only: true
   ractor_scenario: true
-  default_harness: harness-ractor
+  default_harness: ractor
 ractor-msg-backlog:
   desc: unshareable message payloads flood the queues of gated consumer ractors, duplicating data per consumer and driving peak RSS and retention.
   ractor: true
   ractor_only: true
   ractor_scenario: true
-  default_harness: harness-ractor
+  default_harness: ractor
 symbol-name-ractor:
   desc: repeatedly calls Symbol#name on a static symbol under the ractor harness to stress ID-to-string lookup.
   ractor: true
   ractor_only: true
-  default_harness: harness-ractor
+  default_harness: ractor
diff --git a/benchmarks/30k_ifelse.rb b/benchmarks/30k_ifelse.rb
@@ -1,4 +1,4 @@
-require_relative '../harness/loader'
+require_relative '../lib/harness/loader'
 if ENV["RUBY_BENCH_RACTOR_HARNESS"]
   eval_recv = Object.new
   eval_meth = :instance_eval

diff --git a/benchmarks/30k_methods.rb b/benchmarks/30k_methods.rb
@@ -121011,7 +121011,7 @@ def fun_l29_n999()
   end
 end
 
-require_relative '../harness/loader'
+require_relative '../lib/harness/loader'
 
 INTERNAL_ITRS = Integer(ENV.fetch("INTERNAL_ITRS", 200))
 

diff --git a/benchmarks/30k_variables.rb b/benchmarks/30k_variables.rb
@@ -1,6 +1,6 @@
 # frozen_string_literal: true
 # Generated by misc/gen_30k_variables.rb: 300 methods x 100 local variables = 30000 local variables
-require_relative '../harness/loader'
+require_relative '../lib/harness/loader'
 
 def fun_l0_n0(x)
   v0 = x

diff --git a/benchmarks/activerecord/benchmark.rb b/benchmarks/activerecord/benchmark.rb
@@ -1,4 +1,4 @@
-require_relative "../../harness/loader"
+require_relative "../../lib/harness/loader"
 
 Dir.chdir __dir__
 use_gemfile

diff --git a/benchmarks/addressable/equality.rb b/benchmarks/addressable/equality.rb
@@ -1,4 +1,4 @@
-require_relative "../../harness/loader"
+require_relative "../../lib/harness/loader"
 
 Dir.chdir __dir__
 use_gemfile

diff --git a/benchmarks/addressable/getters.rb b/benchmarks/addressable/getters.rb
@@ -1,4 +1,4 @@
-require_relative "../../harness/loader"
+require_relative "../../lib/harness/loader"
 
 Dir.chdir __dir__
 use_gemfile

diff --git a/benchmarks/addressable/join.rb b/benchmarks/addressable/join.rb
@@ -1,4 +1,4 @@
-require_relative "../../harness/loader"
+require_relative "../../lib/harness/loader"
 
 Dir.chdir __dir__
 use_gemfile

diff --git a/benchmarks/addressable/merge.rb b/benchmarks/addressable/merge.rb
@@ -1,4 +1,4 @@
-require_relative "../../harness/loader"
+require_relative "../../lib/harness/loader"
 
 Dir.chdir __dir__
 use_gemfile

diff --git a/benchmarks/addressable/new.rb b/benchmarks/addressable/new.rb
@@ -1,4 +1,4 @@
-require_relative "../../harness/loader"
+require_relative "../../lib/harness/loader"
 
 Dir.chdir __dir__
 use_gemfile

diff --git a/benchmarks/addressable/normalize.rb b/benchmarks/addressable/normalize.rb
@@ -1,4 +1,4 @@
-require_relative "../../harness/loader"
+require_relative "../../lib/harness/loader"
 
 Dir.chdir __dir__
 use_gemfile

diff --git a/benchmarks/addressable/parse.rb b/benchmarks/addressable/parse.rb
@@ -1,4 +1,4 @@
-require_relative "../../harness/loader"
+require_relative "../../lib/harness/loader"
 
 Dir.chdir __dir__
 use_gemfile

diff --git a/benchmarks/addressable/setters.rb b/benchmarks/addressable/setters.rb
@@ -1,4 +1,4 @@
-require_relative "../../harness/loader"
+require_relative "../../lib/harness/loader"
 
 Dir.chdir __dir__
 use_gemfile

diff --git a/benchmarks/addressable/to-s.rb b/benchmarks/addressable/to-s.rb
@@ -1,4 +1,4 @@
-require_relative "../../harness/loader"
+require_relative "../../lib/harness/loader"
 
 Dir.chdir __dir__
 use_gemfile

diff --git a/benchmarks/attr_accessor.rb b/benchmarks/attr_accessor.rb
@@ -1,4 +1,4 @@
-require_relative '../harness/loader'
+require_relative '../lib/harness/loader'
 
 class TheClass
   attr_accessor :levar
-Original file line number
+Diff line change
@@ Expand Up / @@ -121011,7 +121011,7 @@ def fun_l29_n999() @@
       end
     end
-    require_relative '../harness/loader'
+    require_relative '../lib/harness/loader'
     INTERNAL_ITRS = Integer(ENV.fetch("INTERNAL_ITRS", 200))
@@ Expand Down @@