Skip to content

[GR-79953] Optimize vectorcall dispatch. - #1200

Merged
graalvmbot merged 7 commits into
masterfrom
fa/GR-79953/optimize_vectorcall_dispatch
Oct 8, 2026
Merged

graalvmbot merged 7 commits into
masterfrom
fa/GR-79953/optimize_vectorcall_dispatch

Conversation

@graalvmbot

Copy link
Copy Markdown
Collaborator

This change improves GraalPy’s vectorcall support for native types, with Cython-generated extension code being the primary beneficiary.

It allows more calls involving native callables and method descriptors to remain on the vectorcall path, avoiding unnecessary argument adaptation and transitions between native and managed.

It also enables Cython’s fast calling conventions and improves method-call dispatch so Cython-to-Cython calls can stay in native code more often.

The optimized behavior is not Cython-specific: other native extensions using vectorcall, native method descriptors, or native tp_call implementations benefit from the same lower-overhead dispatch paths.

@oracle-contributor-agreement oracle-contributor-agreement Bot added the OCA Verified All contributors have signed the Oracle Contributor Agreement. label Oct 8, 2026
@graalvmbot
graalvmbot merged commit f4a16b4 into master Oct 8, 2026
40 of 43 checks passed
@graalvmbot
graalvmbot deleted the fa/GR-79953/optimize_vectorcall_dispatch branch October 8, 2026 09:06
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

OCA Verified All contributors have signed the Oracle Contributor Agreement.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants