Skip to content

Decode RowFn constants directly - #9680

Merged
connortsui20 merged 1 commit into
developfrom
ct/row-fn-constant-decode
Aug 29, 2026
Merged

Decode RowFn constants directly#9680
connortsui20 merged 1 commit into
developfrom
ct/row-fn-constant-decode

Conversation

@connortsui20

@connortsui20 connortsui20 commented Aug 27, 2026 •
edited
Loading

Copy link
Copy Markdown
Member

Summary

Adds a constant-specific representation to InputElement, so batch constants no longer become one-row decoded columns.

Changes

decode_constant retains the logical batch length while primitive and Boolean inputs extract one native value. Complex inputs can keep ordinary column decoding behind the same hook.

API Changes

Adds InputElement::Constant, decode_constant, and get_constant.

@connortsui20 connortsui20 mentioned this pull request Aug 27, 2026
Merged
@connortsui20 connortsui20 added the changelog/performance A performance improvement label Aug 27, 2026
@connortsui20
connortsui20 marked this pull request as ready for review August 28, 2026 13:46
@connortsui20 connortsui20 mentioned this pull request Aug 28, 2026
Closed
@connortsui20
connortsui20 force-pushed the ct/row-fn-constant-decode branch from f21a243 to 0945a6c Compare August 28, 2026 19:19
@codspeed

codspeed Bot commented Aug 28, 2026 •
edited
Loading

Copy link
Copy Markdown

Merging this PR will regress 6 benchmarks

⚠️ Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.

For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

⚠️ Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.

Open the report in CodSpeed to investigate

⚡ 94 improved benchmarks
❌ 6 regressed benchmarks
✅ 2069 untouched benchmarks
⏩ 106 skipped benchmarks1
🗄️ 4 archived benchmarks run2

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

Mode Benchmark BASE HEAD Efficiency
❌ WallTime arrow_checked_add_u32_neon[16384] 13.5 µs 20.3 µs -33.68%
❌ WallTime mul_u64_nonnull_neon 15.1 µs 20.9 µs -27.81%
❌ WallTime mul_i64_nonnull_neon 17.1 µs 20 µs -14.5%
❌ WallTime multiply_shapes_neon[(16384, PerRowPerRow)] 17.3 µs 20.1 µs -13.64%
❌ WallTime mul_i8_nonnull_avx2 11.2 µs 12.7 µs -11.59%
❌ Simulation compress_fsst[(500, 64, 8)] 475.1 µs 534.3 µs -11.07%
⚡ WallTime deferred_bool_avx2[i32, PerRowConstant] 5.4 µs 2.6 µs ×2.1
⚡ WallTime deferred_bool_avx512[i32, PerRowConstant] 3.3 µs 1.9 µs +76.7%
⚡ WallTime deferred_bool_avx512[i32, ConstantPerRow] 3.2 µs 1.9 µs +71.57%
⚡ WallTime subtract_shapes_neon[(128, ConstantPerRow)] 3.1 µs 1.9 µs +65.19%
⚡ WallTime infallible_bool_avx512[i32, PerRowConstant] 2.8 µs 1.7 µs +63%
⚡ WallTime infallible_bool_avx512[i32, ConstantPerRow] 2.8 µs 1.8 µs +60.38%
⚡ WallTime add_shapes_neon[(128, ConstantPerRow)] 3 µs 1.9 µs +60.01%
⚡ WallTime add_shapes_neon[(128, PerRowConstant)] 3 µs 1.9 µs +59.7%
⚡ WallTime multiply_shapes_neon[(128, PerRowConstant)] 3.1 µs 1.9 µs +58.69%
⚡ WallTime multiply_shapes_neon[(128, ConstantPerRow)] 3.1 µs 2 µs +58.21%
⚡ WallTime subtract_shapes_neon[(128, PerRowConstant)] 2.9 µs 1.9 µs +56.57%
⚡ WallTime subtract_shapes_avx512[(128, PerRowConstant)] 2.9 µs 1.9 µs +55.29%
⚡ WallTime add_shapes_avx2[(128, PerRowConstant)] 2.9 µs 1.9 µs +54.79%
⚡ WallTime subtract_shapes_avx512[(128, ConstantPerRow)] 3 µs 1.9 µs +54.76%
... ... ... ... ... ...

ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

Comparing ct/row-fn-constant-decode (ffca59e) with develop (01f147a)3

Open in CodSpeed

Footnotes

  1. 106 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

  2. 4 benchmarks were run, but are now archived. If they were deleted in another branch, consider rebasing to remove them from the report. Instead if they were added back, click here to restore them. ↩

  3. No successful run was found on develop (3a1838c) during the generation of this report, so 01f147a was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩

@connortsui20
connortsui20 force-pushed the ct/row-fn-constant-decode branch from 0945a6c to 740cd82 Compare August 28, 2026 19:40
@codecov

codecov Bot commented Aug 28, 2026

Copy link
Copy Markdown

Codecov Report

❌ Patch coverage is 59.78261% with 37 lines in your changes missing coverage. Please review.
✅ Project coverage is 83.22%. Comparing base (b38ef22) to head (740cd82).

Files with missing lines Patch % Lines
...ex-array/src/scalar_fn/unstable/row/batch/tests.rs 0.00% 20 Missing ⚠️
...y/src/scalar_fn/unstable/row/types/element/bool.rs 0.00% 12 Missing ⚠️
.../scalar_fn/unstable/row/types/element/primitive.rs 75.00% 3 Missing ⚠️
.../unstable/row/types/element/tuple/element_tuple.rs 90.00% 1 Missing ⚠️
vortex-spatial/src/scalar_fn/row.rs 91.66% 1 Missing ⚠️

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
  • 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

Base automatically changed from ct/row-fn-bool-output-deferred to develop August 29, 2026 00:37
@connortsui20
connortsui20 force-pushed the ct/row-fn-constant-decode branch from 740cd82 to ffca59e Compare August 29, 2026 00:37
@connortsui20
connortsui20 merged commit a745604 into develop Aug 29, 2026
79 of 80 checks passed
@connortsui20
connortsui20 deleted the ct/row-fn-constant-decode branch August 29, 2026 02:00
This was referenced Sep 1, 2026
Open
Open
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@robert3005 robert3005 robert3005 approved these changes
@joseph-isaacs joseph-isaacs Awaiting requested review from joseph-isaacs

Assignees

No one assigned

Labels

changelog/performance A performance improvement

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@connortsui20 @robert3005