Document the benchmarks, and make them build and compare versions again

The benchmark project hasn't configured since #4793: download_test_data.cmake
compiles cmake/detect_libcpp_version.cpp relative to CMAKE_SOURCE_DIR, which
is tests/benchmarks when that is the top-level project, so try_run fails and
so does `make run_benchmarks`. The path is now relative to the module itself,
which is the same file for the main build.

The Dump benchmark discarded dump()'s result, which is [[nodiscard]] by now;
it warned, and left the optimizer free to shorten the loop. The result is
now kept with benchmark::DoNotOptimize.

A new cache variable, JSON_BENCHMARK_INCLUDE_DIR, names the directory holding
the nlohmann/json.hpp to benchmark (single_include by default, as before),
so the same benchmarks can be built against two versions and compared.

tests/benchmarks/README.md documents what is measured, how to build and run
the benchmarks, how to read the output, and how to compare two versions with
Google Benchmark's compare.py; it recommends doing so by hand before a
release rather than in CI.

Signed-off-by: Niels Lohmann <mail@nlohmann.me>

cmake/download_test_data.cmake

@@ -77,7 +77,7 @@ if(CMAKE_CROSSCOMPILING)
 endif()
 if(NOT DEFINED LIBCPP_VERSION_OUTPUT_CACHED)
     try_run(RUN_RESULT_VAR COMPILE_RESULT_VAR
-        "${CMAKE_BINARY_DIR}" SOURCES "${CMAKE_SOURCE_DIR}/cmake/detect_libcpp_version.cpp"
+        "${CMAKE_BINARY_DIR}" SOURCES "${CMAKE_CURRENT_LIST_DIR}/detect_libcpp_version.cpp"
         RUN_OUTPUT_VARIABLE LIBCPP_VERSION_OUTPUT
         COMPILE_OUTPUT_VARIABLE LIBCPP_VERSION_COMPILE_OUTPUT
     )

tests/benchmarks/CMakeLists.txt

@@ -26,9 +26,14 @@ endif()
 set(CMAKE_MODULE_PATH ${CMAKE_CURRENT_SOURCE_DIR}/../../cmake ${CMAKE_MODULE_PATH})
 include(download_test_data)
 
+# the header to benchmark; point this at a directory holding another version's
+# nlohmann/json.hpp to compare versions (see README.md)
+set(JSON_BENCHMARK_INCLUDE_DIR "${CMAKE_CURRENT_SOURCE_DIR}/../../single_include" CACHE PATH
+    "directory containing the nlohmann/json.hpp to benchmark")
+
 # benchmark binary
 add_executable(json_benchmarks src/benchmarks.cpp)
 target_compile_features(json_benchmarks PRIVATE cxx_std_11)
 target_link_libraries(json_benchmarks benchmark ${CMAKE_THREAD_LIBS_INIT})
 add_dependencies(json_benchmarks download_test_data)
-target_include_directories(json_benchmarks PRIVATE ${CMAKE_SOURCE_DIR}/../../single_include ${CMAKE_BINARY_DIR}/include)
+target_include_directories(json_benchmarks PRIVATE ${JSON_BENCHMARK_INCLUDE_DIR} ${CMAKE_BINARY_DIR}/include)

tests/benchmarks/README.md

@@ -0,0 +1,128 @@
+# Benchmarks
+
+Micro-benchmarks for parsing, serialization and the binary formats, written with
+[Google Benchmark](https://github.com/google/benchmark). They are not run by CI; see
+[When to run them](#when-to-run-them).
+
+## What is measured
+
+| benchmark | what it does |
+|---|---|
+| `ParseFile`, `ParseString` | parse JSON from a file stream or a string |
+| `ParseIndented` | parse the large files re-indented by 4 spaces, for the lexer's whitespace handling |
+| `Dump` | serialize, compact (`-`) and indented (`4`) |
+| `ToCbor`, `BinaryToCbor` | write CBOR; `BinaryToCbor` writes binary values of growing size |
+| `FromMsgpack` | read MessagePack; unchanged over the years, so its numbers stay comparable across releases |
+| `FromBinaryBuffer`, `FromBinaryFile` | read CBOR, MessagePack, UBJSON, BJData and BSON from a buffer or a `FILE*` |
+| `FromBinaryShape` | read deeply nested, container-heavy and scalar-heavy documents in every binary format |
+| `FromCborChunkedString` | read CBOR strings split into indefinite-length chunks |
+
+The input files are those of [nativejson-benchmark](https://github.com/miloyip/nativejson-benchmark) (`canada`,
+`citm_catalog`, `twitter`), a large `jeopardy` file, and number-heavy files (`floats`, `signed_ints`, ...).
+`bytes_per_second` counts the bytes read or written: the JSON text when parsing, the output when serializing.
+
+## Requirements
+
+- CMake 3.11 or later, a C++11 compiler, and Ninja for the `make` target.
+- Network access on the first configure: CMake fetches Google Benchmark and downloads the
+  [test data](https://github.com/nlohmann/json_test_data) into the build directory. To reuse a download, pass
+  `-DJSON_TestDataDirectory=<build directory>/test_files`.
+- The benchmarks include `single_include/nlohmann/json.hpp`, so run `make amalgamate` after changing anything in
+  `include/`.
+
+GCC and Clang builds use `-O3 -flto -DNDEBUG`.
+
+## Running them
+
+From the repository root, this builds everything from scratch in `cmake-build-benchmarks` and runs all benchmarks:
+
+```sh
+make run_benchmarks
+```
+
+To build once and run selectively:
+
+```sh
+cmake -S tests/benchmarks -B build-benchmarks -G Ninja -DCMAKE_BUILD_TYPE=Release
+cmake --build build-benchmarks
+build-benchmarks/json_benchmarks --benchmark_filter='ParseString|Dump'
+```
+
+Useful options of `json_benchmarks`:
+
+| option | effect |
+|---|---|
+| `--benchmark_list_tests` | list the benchmarks instead of running them |
+| `--benchmark_filter=<regex>` | run only the benchmarks whose names match |
+| `--benchmark_repetitions=<n>` | run every benchmark `n` times and add mean, median, standard deviation and coefficient of variation |
+| `--benchmark_enable_random_interleaving=true` | run the repetitions in random order, which spreads out drifts such as thermal throttling |
+| `--benchmark_min_time=<seconds>s` | run each benchmark at least this long (e.g. `2s`) |
+| `--benchmark_out=<file> --benchmark_out_format=json` | also write the results to a file, e.g. for `compare.py` |
+
+## Reading the output
+
+Each line shows the wall-clock `Time` and the `CPU` time per iteration, the number of `Iterations` Google Benchmark
+chose, and the throughput in `bytes_per_second`. With repetitions, the lines ending in `_median` are the ones to
+compare. A `_cv` (coefficient of variation) above a few percent means the machine was too noisy for small
+differences to mean anything.
+
+## Comparing two versions
+
+To see what a change or a release did, build the same benchmarks twice: once against the header of the version to
+compare with, and once against the current one. `JSON_BENCHMARK_INCLUDE_DIR` names the directory holding the
+`nlohmann/json.hpp` to benchmark. For example, to compare the current checkout with 3.12.0:
+
+```sh
+# the header of the version to compare with
+mkdir -p build-baseline-header/nlohmann
+git show v3.12.0:single_include/nlohmann/json.hpp > build-baseline-header/nlohmann/json.hpp
+
+# the same benchmarks, built against either header
+cmake -S tests/benchmarks -B build-baseline -G Ninja -DCMAKE_BUILD_TYPE=Release \
+      -DJSON_BENCHMARK_INCLUDE_DIR="$PWD/build-baseline-header"
+cmake -S tests/benchmarks -B build-current -G Ninja -DCMAKE_BUILD_TYPE=Release
+cmake --build build-baseline
+cmake --build build-current
+
+# run both, back to back
+build-baseline/json_benchmarks --benchmark_repetitions=10 --benchmark_enable_random_interleaving=true \
+                               --benchmark_out=build-baseline/results.json --benchmark_out_format=json
+build-current/json_benchmarks --benchmark_repetitions=10 --benchmark_enable_random_interleaving=true \
+                              --benchmark_out=build-current/results.json --benchmark_out_format=json
+```
+
+Google Benchmark ships a tool to compare the two result files. It needs NumPy and SciPy:
+
+```sh
+python3 -m venv build-venv
+build-venv/bin/pip install numpy scipy
+build-venv/bin/python build-current/_deps/benchmark-src/tools/compare.py -a benchmarks build-baseline/results.json build-current/results.json
+```
+
+The tool's own `tools/requirements.txt` pins the newest NumPy and SciPy, which may need a newer Python than yours;
+unpinned versions work as well. In its output:
+
+- the `Time` and `CPU` columns are relative changes: `-0.35` means 35% faster, `+0.10` means 10% slower;
+- `_pvalue` lines report a Mann-Whitney U test of whether the two versions differ. It needs at least 9
+  repetitions, and a p-value below 0.05 means the difference is unlikely to be noise;
+- `OVERALL_GEOMEAN` summarizes all benchmarks;
+- `-a` shows only the aggregates, not every repetition.
+
+The header you compare with must support everything the benchmarks use. The current benchmarks build against 3.12.0.
+Only benchmarks present in both result files are compared, so for older releases, either filter the benchmarks or
+build that release's own `tests/benchmarks` against its own header.
+
+## Getting stable numbers
+
+- Build and run both versions on the same machine, one right after the other.
+- Keep the machine otherwise idle: no builds, no browser, and a laptop plugged in.
+- On Linux, set the CPU frequency governor to `performance`, e.g. `sudo cpupower frequency-set --governor performance`.
+  Google Benchmark prints a warning when frequency scaling is enabled. Pinning the process to a core
+  (`taskset -c 2 ...`) helps as well.
+- Use 10 or more repetitions with random interleaving, compare medians, and treat changes within the `_cv` as noise.
+
+## When to run them
+
+They are a manual step, not part of CI: shared CI runners vary more between runs than most of the effects measured.
+Run the comparison above before a release, comparing the previous release tag with `develop`, and for pull requests
+that claim to change performance.

tests/benchmarks/src/benchmarks.cpp

@@ -131,7 +131,8 @@ static void Dump(benchmark::State& state, const char* filename, int indent)
 
     while (state.KeepRunning())
     {
-        j.dump(indent);
+        std::string output = j.dump(indent);
+        benchmark::DoNotOptimize(output);
     }
 
     state.SetBytesProcessed(state.iterations() * j.dump(indent).size());