Skip to content

Conversation

@benfdking
Copy link
Contributor

@benfdking benfdking commented Oct 30, 2025

Reenables benchmark but not in 3.9 because of issue related to python/cpython#32073 which was not backported to 3.9 which causes the deadlock

Copy link
Contributor

Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull Request Overview

This PR enables the benchmark CI step in CircleCI and configures it to run single-threaded. The benchmark was previously commented out and is now being re-enabled with optimizations for CI environments.

  • Re-enabled the benchmark-ci step in CircleCI configuration (skipping Python 3.9)
  • Added --processes=1 flag to the benchmark command to force single-process execution

Reviewed Changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.

File Description
Makefile Added --processes=1 flag to benchmark-ci target to run benchmarks in single-process mode
.circleci/continue_config.yml Uncommented and restructured the benchmark-ci step to run on all Python versions except 3.9

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Makefile Outdated

benchmark-ci:
python benchmarks/lsp_render_model_bench.py --debug-single-value
python benchmarks/lsp_render_model_bench.py --debug-single-value --processes=1
Copy link

Copilot AI Nov 2, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The --processes=1 flag is not recognized by the pyperf.Runner() in the benchmark script. The script uses pyperf.Runner() without any argument parsing, so this flag will cause the script to fail. If you need to control the number of processes, you should initialize the Runner with pyperf.Runner(processes=1) in the script itself, or use pyperf's built-in flag --processes 1 (note the space instead of equals sign) which is automatically handled by pyperf.Runner().

Suggested change
python benchmarks/lsp_render_model_bench.py --debug-single-value --processes=1
python benchmarks/lsp_render_model_bench.py --debug-single-value --processes 1

Copilot uses AI. Check for mistakes.
- by skipping in 3.9 for time being
- newer google libraries print warnings which break the
lsp stream
@izeigerman izeigerman merged commit 922ec89 into main Nov 5, 2025
36 checks passed
@izeigerman izeigerman deleted the fix_benchmark branch November 5, 2025 17:26
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants