Title: Performance regression in anext() after moving its implementation from C to Python · Issue #158407 · python/cpython · GitHub
Open Graph Title: Performance regression in anext() after moving its implementation from C to Python · Issue #158407 · python/cpython
X Title: Performance regression in anext() after moving its implementation from C to Python · Issue #158407 · python/cpython
Description: Bug report Bug description: I have a project where I build CPython from individual commits into OCI images and run them on AWS Lambda to detect performance changes before they reach a runtime release. While comparing merged commits on ma...
Open Graph Description: Bug report Bug description: I have a project where I build CPython from individual commits into OCI images and run them on AWS Lambda to detect performance changes before they reach a runtime relea...
X Description: Bug report Bug description: I have a project where I build CPython from individual commits into OCI images and run them on AWS Lambda to detect performance changes before they reach a runtime relea...
Opengraph URL: https://github.com/python/cpython/issues/158407
X: @github
Domain: github.com
{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"Performance regression in anext() after moving its implementation from C to Python","articleBody":"# Bug report\n\n### Bug description:\n\nI have a project where I build CPython from individual commits into OCI images and run them on AWS Lambda to detect performance changes before they reach a runtime release.\n\nWhile comparing merged commits on `main`, I found that #157362, commit `0a6c1ed34118b091230ee38fc047bcc2df8e5c5e`, makes explicit calls to `anext()` slower. I compared that commit directly with its parent, `8df6f077115484361b40952751fb7d703843c192`.\n\nThe change is visible with a regular GIL build and does not require free-threading.\n\nReproducer:\n\n```python\nimport asyncio\nimport statistics\nimport time\n\nITEMS = 1_000_000\nREPEATS = 7\n\n\nasync def items():\n for value in range(ITEMS):\n yield value\n\n\nasync def consume_with_anext():\n stream = items()\n started = time.perf_counter_ns()\n for _ in range(ITEMS):\n await anext(stream)\n return (time.perf_counter_ns() - started) / ITEMS\n\n\nasync def consume_with_async_for():\n started = time.perf_counter_ns()\n async for _ in items():\n pass\n return (time.perf_counter_ns() - started) / ITEMS\n\n\nasync def main():\n for name, benchmark in (\n (\"anext\", consume_with_anext),\n (\"async for\", consume_with_async_for),\n ):\n samples = [await benchmark() for _ in range(REPEATS)]\n print(f\"{name}: {statistics.median(samples):.1f} ns/item\")\n\n\nasyncio.run(main())\n```\n\nOn macOS arm64, using identically configured builds and seven interleaved outer repetitions, I measured `await anext(async_generator)` as:\n\n| Revision | Median |\n| --- | ---: |\n| parent `8df6f0771154` | 141.8 ns/item |\n| `0a6c1ed34118` | 228.4 ns/item |\n\nThis is a 61.1% regression. The `async for` control changed by 0.7%.\n\nI also isolated the builtin from the async generator itself:\n\n| Revision | `await generator.__anext__()` | `await anext(generator)` |\n| --- | ---: | ---: |\n| parent | 170.7 ns/item | 137.0 ns/item |\n| commit | 167.3 ns/item | 215.4 ns/item |\n\nThe direct method did not regress. The additional time comes from the new Python implementation of the builtin, which performs `type()` and `__anext__` lookup followed by a Python call, while the previous C implementation called the async iteration slot directly.\n\nI reproduced the result in two independent AWS Lambda experiments using optimized x86_64 CPython builds with LTO. Each experiment used 12 environments and 24 paired warm samples:\n\n| Experiment | Parent median | Commit median | Control-adjusted regression |\n| --- | ---: | ---: | ---: |\n| 1 | 180.0 ms | 248.1 ms | 39.8% |\n| 2 | 179.3 ms | 245.0 ms | 37.0% |\n\nThis affects code that manually consumes ready or buffered async iterators with `await anext(stream)`. This benchmark isolates CPU overhead and does not model network latency. Code using `async for` does not use this path.\n\n### CPython versions tested on:\n\nCPython main branch\n\n### Operating systems tested on:\n\nLinux, macOS\n\n\n\u003c!-- gh-linked-prs -->\n### Linked PRs\n* gh-158964\n\u003c!-- /gh-linked-prs -->\n","author":{"url":"https://github.com/leandrodamascena","@type":"Person","name":"leandrodamascena"},"datePublished":"2026-09-29T13:19:38Z","url":"https://github.com/python/cpython/issues/158407"}
| react-profiling | 0 |
| react-import-map | react,react-dom,react-dom/client,react-dom/profiling,react-is,react/compiler-runtime,react/jsx-dev-runtime,react/jsx-runtime,scheduler@b8488ee2c322 |
| route-pattern | /_view_fragments/issues/show/:user_id/:repository/:id/issue_layout(.:format) |
| route-controller | voltron_issues_fragments |
| route-action | issue_layout |
| fetch-nonce | v2:7a4520db-62b3-5906-a0b2-29b29d1775af |
| current-catalog-service-hash | 81bb79d38c15960b92d99bca9288a9108c7a47b18f2423d0f6438c5b7bcd2114 |
| request-id | AA4C:34CAB:42852AC:54ED497:6B023A97 |
| html-safe-nonce | 2b0d49f4703a9fa82c5989e7253fc7c1f052ed13f252cbf85cb97087ef8ffae5 |
| visitor-payload | eyJyZWZlcnJlciI6IiIsInJlcXVlc3RfaWQiOiJBQTRDOjM0Q0FCOjQyODUyQUM6NTRFRDQ5Nzo2QjAyM0E5NyIsInZpc2l0b3JfaWQiOiIzOTk5NDQ5ODQ4MjAwOTY4ODU1IiwicmVnaW9uX2VkZ2UiOiJpYWQiLCJyZWdpb25fcmVuZGVyIjoiaWFkIn0= |
| visitor-hmac | 3b48e7beaf14035782c34559bda873856affa57f1ead45624eb7e7722db66c10 |
| hovercard-subject-tag | issue:5631222465 |
| github-keyboard-shortcuts | repository,issues,copilot |
| google-site-verification | Apib7-x98H0j5cPqHWwSMm6dNU4GmODRoqxLiDzdx9I |
| octolytics-url | https://collector.github.com/github/collect |
| analytics-location | / |
| fb:app_id | 1401488693436528 |
| apple-itunes-app | app-id=1477376905, app-argument=https://github.com/_view_fragments/issues/show/python/cpython/158407/issue_layout |
| twitter:image | https://opengraph.githubassets.com/9208e40fae47ebe026fddd5ff81657e5e05c840bf73a6cdde2ce0c6ecbff7d78/python/cpython/issues/158407 |
| twitter:card | summary_large_image |
| og:image | https://opengraph.githubassets.com/9208e40fae47ebe026fddd5ff81657e5e05c840bf73a6cdde2ce0c6ecbff7d78/python/cpython/issues/158407 |
| og:image:alt | Bug report Bug description: I have a project where I build CPython from individual commits into OCI images and run them on AWS Lambda to detect performance changes before they reach a runtime relea... |
| og:image:width | 1200 |
| og:image:height | 600 |
| og:site_name | GitHub |
| og:type | object |
| og:author:username | leandrodamascena |
| hostname | github.com |
| expected-hostname | github.com |
| None | 87e774ee7c6b3bee0490394e4d3e57ec9dbc956f74b87ea685dc8996c1676456 |
| turbo-cache-control | no-preview |
| go-import | github.com/python/cpython git https://github.com/python/cpython.git |
| octolytics-dimension-user_id | 1525981 |
| octolytics-dimension-user_login | python |
| octolytics-dimension-repository_id | 81598961 |
| octolytics-dimension-repository_nwo | python/cpython |
| octolytics-dimension-repository_public | true |
| octolytics-dimension-repository_is_fork | false |
| octolytics-dimension-repository_network_root_id | 81598961 |
| octolytics-dimension-repository_network_root_nwo | python/cpython |
| turbo-body-classes | logged-out env-production page-responsive |
| disable-turbo | false |
| browser-stats-url | https://api.github.com/_private/browser/stats |
| browser-errors-url | https://api.github.com/_private/browser/errors |
| release | 423dbb54c5e7bfecc838eb387c1435eb3d0e760d |
| ui-target | full |
| theme-color | #1e2327 |
| color-scheme | light dark |
| react-app-name | issues |
Links:
Viewport: width=device-width