Add intrinsics support, Add nanobench + documentation + example

[mcdubhghlas]

Title.

Summary by CodeRabbit

• New Features

  • Added optional benchmark builds and run targets, making it easier to compile and execute performance tests from the build system.
  • Benchmark files are now built as separate executables, alongside support for SIMD-based math benchmarking.

• Documentation

  • Added benchmarking guidance with setup steps, example usage, and common tuning options.

[mcdubhghlas]

This is the output for the benchmark test I added:

# compiler : clang 22.1.4
# fast-math: off
# sizeof(Vector4) = 16
# alignof(Vector4) = 16
Warning, results might be unstable:
* CPU frequency scaling enabled: CPU 0 between 3,012.5 and 5,883.2 MHz

Recommendations
* Use 'pyperf system tune' before benchmarking. See https://github.com/psf/pyperf

| relative |              ns/dot |               dot/s |    err% |         ins/dot |         cyc/dot |    IPC |        bra/dot |   miss% |     total | dot throughput
|---------:|--------------------:|--------------------:|--------:|----------------:|----------------:|-------:|---------------:|--------:|----------:|:---------------
|   100.0% |                0.55 |    1,831,818,654.25 |    0.5% |           11.77 |            2.45 |  4.796 |           0.25 |    0.4% |      0.11 | `scalar`
|   120.4% |                0.45 |    2,204,644,868.10 |    0.3% |            8.02 |            2.04 |  3.933 |           0.25 |    0.4% |      0.09 | `simd`

| relative |              ns/dot |               dot/s |    err% |         ins/dot |         cyc/dot |    IPC |        bra/dot |   miss% |     total | dot latency chain
|---------:|--------------------:|--------------------:|--------:|----------------:|----------------:|-------:|---------------:|--------:|----------:|:------------------
|   100.0% |                3.08 |      324,636,802.67 |    0.1% |            8.51 |           13.85 |  0.614 |           0.50 |    0.2% |      0.63 | `scalar`
|   124.5% |                2.47 |      404,160,867.71 |    0.0% |            8.51 |           11.12 |  0.765 |           0.50 |    0.2% |      0.51 | `simd`

| relative |              ns/dot |               dot/s |    err% |         ins/dot |         cyc/dot |    IPC |        bra/dot |   miss% |     total | dot compare/count
|---------:|--------------------:|--------------------:|--------:|----------------:|----------------:|-------:|---------------:|--------:|----------:|:------------------
|   100.0% |                0.58 |    1,712,182,462.47 |    0.2% |           10.76 |            2.63 |  4.100 |           0.13 |    0.0% |      0.12 | `scalar`
|   115.4% |                0.51 |    1,976,615,709.62 |    0.2% |           11.51 |            2.27 |  5.059 |           0.50 |    0.2% |      0.10 | `simd`

My computer:

$ sh /home/mcdubh/display.sh                                                                  
    .-----.      mcdubh@ryzen-desktop
  .`    _  `.    OS:        Gentoo Linux
  `.   (_ )   `. KERNEL:    Linux 7.0.1-gentoo
    `.        /  WM:        dwm
   .`       .`   PACKAGES:  1215
  /       .`     SHELL:     mksh
  \____.-`       CPU:       AMD Ryzen 9 7950X 16-Core Processor
                 RAM:       56746MiB/63396MiB
                 GPU:       Navi 31 Radeon RX 7900 XT
$ 

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: 77aa3534-a743-41ff-896a-be70408d69d5

📥 Commits

Reviewing files that changed from the base of the PR and between 5b19441 and 2d43b73.

📒 Files selected for processing (2)

• Docs/Benchmarking.md
• cmake/Modules.cmake

✅ Files skipped from review due to trivial changes (1)

• Docs/Benchmarking.md

🚧 Files skipped from review as they are similar to previous changes (1)

• cmake/Modules.cmake

---

📝 Walkthrough

Walkthrough

Introduces optional benchmark support: a BUILD_BENCHMARKS CMake option with benchmarks/run-benchmarks aggregation targets, Modules.cmake discovery/build of *.bench.cpp files linked to module libraries, a new Runtime/Platform/intrinsics.h header, a Vector4 dot-product nanobench, and accompanying documentation.

Changes

Benchmark infrastructure and Vector4 benchmark

Layer / File(s)	Summary
CMake benchmark option and aggregation targets CMakeLists.txt	Adds BUILD_BENCHMARKS option and defines benchmarks/run-benchmarks targets from the DRACO_BENCH_TARGETS global property.
Modules.cmake benchmark discovery and build wiring cmake/Modules.cmake	Sets NANOBENCH_INCLUDE_DIR, globs *.bench.cpp files, excludes them from module sources, and builds/links benchmark executables while registering targets in DRACO_BENCH_TARGETS.
Platform intrinsics header and Vector4 module include Engine/cpp/Runtime/Platform/intrinsics.h, Engine/cpp/Runtime/Core/Math/Vector4.cppm	Adds a header conditionally including immintrin.h/arm_neon.h and includes it from the Vector4 module.
Vector4 dot-product benchmark implementation Engine/cpp/Runtime/Core/Math/Vector4.bench.cpp	Implements scalar/SIMD dot callables, metadata printing, a throughput helper, and three nanobench benchmark groups (throughput, latency, compare/count) run from main().
Benchmarking documentation Docs/Benchmarking.md	Documents how to add *.bench.cpp files, sample nanobench usage, build/run commands, and optional configuration knobs.

Estimated code review effort: 3 (Moderate) | ~25 minutes

Possibly related PRs

• Redot-Engine/DraconicEngine#1: Both PRs touch cmake/Modules.cmake, extending the same add_modules_library logic for target discovery.
• Redot-Engine/DraconicEngine#18: Both PRs touch draco::math::Vector4 dot-product/SIMD details.

Suggested reviewers: OldDev78, JoltedJon

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title matches the main changes: intrinsics support plus nanobench benchmarking, documentation, and an example benchmark.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@cmake/Modules.cmake`:
- Around line 90-107: The benchmark executables created in the BUILD_BENCHMARKS
block are currently part of the default ALL build, which bypasses the intended
benchmarks/run-benchmarks workflow. Update the add_executable call for each
BENCH_TARGET in Modules.cmake to exclude these targets from ALL, while keeping
the existing target registration and linkage intact. Use the BENCH_TARGET
creation path in the BUILD_BENCHMARKS foreach loop to locate the change and
preserve the aggregated DRACO_BENCH_TARGETS behavior.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: e4e63d88-10ac-4805-bc36-a1bbcaf007b2

📥 Commits

Reviewing files that changed from the base of the PR and between b642f33 and 5b19441.

📒 Files selected for processing (7)

• CMakeLists.txt
• Docs/benchmarking.md
• Engine/cpp/Runtime/Core/Math/Vector4.bench.cpp
• Engine/cpp/Runtime/Core/Math/Vector4.cppm
• Engine/cpp/Runtime/Platform/intrinsics.h
• Engine/cpp/ThirdParty/nanobench/nanobench.h
• cmake/Modules.cmake