Add additional gc benchmark with pickletools (#437)

python · pgdr · Nov 3, 2025 · Nov 5, 2025 · Nov 12, 2025 · Nov 3, 2025
commit 0147c02f2a6dc2debef62313acec3af186a7d65e
diff --git a/pyperformance/data-files/benchmarks/bm_pickle_opt/pyproject.toml b/pyperformance/data-files/benchmarks/bm_pickle_opt/pyproject.toml
@@ -0,0 +1,9 @@
+[project]
+name = "pyperformance_bm_pickle_opt"
+requires-python = ">=3.8"
+dependencies = ["pyperf"]
+urls = {repository = "https://github.com/python/pyperformance"}
+dynamic = ["version"]
+
+[tool.pyperformance]
+name = "pickle_opt"
diff --git a/pyperformance/data-files/benchmarks/bm_pickle_opt/run_benchmark.py b/pyperformance/data-files/benchmarks/bm_pickle_opt/run_benchmark.py
@@ -0,0 +1,45 @@
+"""The background for this benchmark is that the garbage collection in Python 3.14
+had a performance regression, see
+
+* https://github.com/python/cpython/issues/140175
+* https://github.com/python/cpython/issues/139951
+
+"""
+
+import tempfile
+from pathlib import Path
+
+import pyperf
+
+
+def setup(fname, N):
+    import pickle
+
+    x = {}
+    for i in range(1, N):
+        x[i] = f"ii{i:>07}"
+
+    with open(fname, "wb") as fh:
+        pickle.dump(x, fh, protocol=4)
+
+
+def run(fname):
+    import pickletools
+
+    with open(fname, "rb") as fh:
+        p = fh.read()
+
+    s = pickletools.optimize(p)
+
+    with open(fname.with_suffix(".out"), "wb") as fh:
+        fh.write(s)
+
+
+if __name__ == "__main__":
+    runner = pyperf.Runner()
+    N = 1_000_000
+    tmp_path = Path(tempfile.mkdtemp())
+    fname = tmp_path / "pickle"
+    setup(fname, N)
+    runner.metadata["description"] = "Pickletools optimize"
+    runner.bench_func("pickle_opt", run, fname)