roadmap: GraalPy verified — compatible, but perf is workload-dependent - #6
Merged
Conversation
Verification run 2026-07-18 (GraalPy 3.12.8 / GraalVM CE 25.1.3, container): emitted output runs unchanged, including PEP 701 nested-quote f-strings and class-pattern match. Bench numbers vs CPython 3.14: collatz 1.7x faster, map_build ~1.6x slower, expr_eval ~4x slower. Warmup probes localize the slow path to the ADT-as-classes representation (never JIT-warms in any styling), while tuple-shaped data warms to 2x faster than CPython. Docs guidance and the PyPy caveat updated accordingly. Claude-Session: https://claude.ai/code/session_011PBhfjrzsUtwupYtJtzAWu
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Records the GraalPy verification result in the "Performance beyond CPython" entry.
Compatibility: verified. Emitted output runs unchanged on GraalPy 3.12.8 — PEP 701 nested-quote f-strings, class-pattern
match, dataclass ADTs/records, full bench suite with byte-identical outputs.Performance: workload-dependent, not a blanket win. vs CPython 3.14: collatz 1.7x faster, map_build ~1.6x slower, expr_eval ~4x slower. In-process warmup probes show GraalPy JIT-warming the hand-written tuple-based baseline to 2x faster than CPython while never warming any ADT-as-classes variant (match or isinstance, dataclass or
__slots__) — allocation-heavy trees of small class instances are the slow path, and that is Pyfun's core data shape.Docs line updated: runs unchanged, measure with
bench/run.py --python graalpy, don't assume; CPython 3.14 is the best all-round host measured; the result does not auto-transfer to PyPy (different GC), pending--target 3.11. Artifacts inlocal/graalpy-verification/(untracked by design).https://claude.ai/code/session_011PBhfjrzsUtwupYtJtzAWu