Make Mask execution strict about nullable booleans#8121
Merged
Conversation
Mask execution now requires a non-nullable boolean array and errors on nullable input. The new MaskNullAsFalse executable preserves the previous null-as-false coercion for filter and pruning predicates over nullable data, where SQL semantics treat NULL as not matching. Predicate-evaluation call sites (filter, prune, dict filter, is_constant) use MaskNullAsFalse; validity-array call sites keep the stricter Mask. Signed-off-by: Claude <noreply@anthropic.com>
joseph-isaacs
added
action/benchmark
github-actions
Bot
removed
the
action/benchmark
Remove the shared NullHandling enum and execute_mask helper in favor of a self-contained Executable impl for each target. Mask still requires a non-nullable boolean array; MaskNullAsFalse coerces nulls to false. Signed-off-by: Claude <noreply@anthropic.com>
Contributor
Polar Signals Profiling ResultsLatest Run
Previous Runs (2)
Powered by Polar Signals Cloud |
Contributor
Benchmarks: PolarSignals ProfilingVortex (geomean): 1.009x ➖ How to read Verdict and Engines
datafusion / vortex-file-compressed (1.009x ➖, 0↑ 0↓)
No file size changes detected. |
Contributor
File Sizes: PolarSignals ProfilingNo file size changes detected. |
Contributor
Benchmarks: TPC-H SF=1 on NVMEVerdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.988x ➖, 0↑ 0↓)
datafusion / vortex-compact (0.999x ➖, 0↑ 0↓)
datafusion / parquet (0.998x ➖, 0↑ 0↓)
datafusion / arrow (0.984x ➖, 1↑ 0↓)
duckdb / vortex-file-compressed (0.995x ➖, 0↑ 0↓)
duckdb / vortex-compact (1.003x ➖, 0↑ 0↓)
duckdb / parquet (1.001x ➖, 1↑ 1↓)
duckdb / duckdb (0.994x ➖, 0↑ 0↓)
File Size Changes (11 files changed, +0.1% overall, 6↑ 5↓)
Totals:
|
Contributor
File Sizes: TPC-H SF=1 on NVMENo file size changes detected. |
Contributor
Benchmarks: FineWeb NVMeVerdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (1.008x ➖, 0↑ 1↓)
datafusion / vortex-compact (0.987x ➖, 1↑ 1↓)
datafusion / parquet (0.969x ➖, 1↑ 0↓)
duckdb / vortex-file-compressed (0.996x ➖, 0↑ 0↓)
duckdb / vortex-compact (0.994x ➖, 0↑ 0↓)
duckdb / parquet (1.009x ➖, 0↑ 0↓)
File Size Changes (1 files changed, +0.1% overall, 1↑ 0↓)
Totals:
|
Contributor
File Sizes: FineWeb NVMeNo file size changes detected. |
Contributor
Benchmarks: TPC-DS SF=1 on NVMEVerdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.996x ➖, 2↑ 1↓)
datafusion / vortex-compact (1.001x ➖, 1↑ 2↓)
datafusion / parquet (0.991x ➖, 2↑ 1↓)
duckdb / vortex-file-compressed (1.004x ➖, 1↑ 3↓)
duckdb / vortex-compact (1.003x ➖, 1↑ 1↓)
duckdb / parquet (0.997x ➖, 1↑ 0↓)
duckdb / duckdb (1.002x ➖, 0↑ 2↓)
File Size Changes (6 files changed, +0.0% overall, 2↑ 4↓)
Totals:
|
Contributor
File Sizes: TPC-DS SF=1 on NVMENo file size changes detected. |
Contributor
Benchmarks: FineWeb S3Verdict: No clear signal (environment too noisy confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.898x ➖, 1↑ 1↓)
datafusion / vortex-compact (0.951x ➖, 1↑ 0↓)
datafusion / parquet (0.949x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (0.896x ➖, 1↑ 0↓)
duckdb / vortex-compact (1.015x ➖, 0↑ 1↓)
duckdb / parquet (0.912x ➖, 0↑ 0↓)
|
Contributor
Benchmarks: Statistical and Population GeneticsVerdict: No clear signal (low confidence) How to read Verdict and Engines
duckdb / vortex-file-compressed (1.006x ➖, 0↑ 0↓)
duckdb / vortex-compact (0.994x ➖, 0↑ 0↓)
duckdb / parquet (1.003x ➖, 0↑ 0↓)
File Size Changes (1 files changed, +0.0% overall, 1↑ 0↓)
Totals:
|
Contributor
File Sizes: Statistical and Population GeneticsNo file size changes detected. |
Contributor
Benchmarks: TPC-H SF=10 on NVMEVerdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (1.050x ➖, 0↑ 0↓)
datafusion / vortex-compact (1.011x ➖, 0↑ 0↓)
datafusion / parquet (1.065x ➖, 0↑ 4↓)
datafusion / arrow (1.026x ➖, 0↑ 2↓)
duckdb / vortex-file-compressed (0.993x ➖, 0↑ 0↓)
duckdb / vortex-compact (0.994x ➖, 0↑ 0↓)
duckdb / parquet (0.995x ➖, 0↑ 0↓)
duckdb / duckdb (0.991x ➖, 0↑ 0↓)
File Size Changes (27 files changed, -0.0% overall, 10↑ 17↓)
Totals:
|
Contributor
File Sizes: TPC-H SF=10 on NVMENo file size changes detected. |
Contributor
Benchmarks: Clickbench on NVMEVerdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.990x ➖, 1↑ 0↓)
datafusion / parquet (0.993x ➖, 0↑ 1↓)
duckdb / vortex-file-compressed (0.986x ➖, 2↑ 0↓)
duckdb / parquet (1.002x ➖, 0↑ 1↓)
duckdb / duckdb (1.000x ➖, 0↑ 0↓)
File Size Changes (104 files changed, +0.0% overall, 54↑ 50↓)
Totals:
|
Contributor
File Sizes: Clickbench on NVMEFile Size Changes (1 files changed, -0.0% overall, 0↑ 1↓)
Totals:
|
Contributor
Benchmarks: TPC-H SF=1 on S3Verdict: No clear signal (environment too noisy confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.878x ➖, 3↑ 0↓)
datafusion / vortex-compact (0.775x ➖, 8↑ 1↓)
datafusion / parquet (1.047x ➖, 0↑ 1↓)
duckdb / vortex-file-compressed (0.896x ➖, 1↑ 0↓)
duckdb / vortex-compact (0.924x ➖, 0↑ 0↓)
duckdb / parquet (0.895x ➖, 0↑ 0↓)
|
Contributor
Benchmarks: CompressionVortex (geomean): 0.999x ➖ How to read Verdict and Engines
unknown / unknown (0.990x ➖, 1↑ 0↓)
|
Contributor
Benchmarks: Random AccessVortex (geomean): 1.010x ➖ How to read Verdict and Engines
unknown / unknown (1.045x ➖, 1↑ 2↓)
|
Contributor
Benchmarks: TPC-H SF=10 on S3Verdict: No clear signal (environment too noisy confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (1.021x ➖, 0↑ 1↓)
datafusion / vortex-compact (0.914x ➖, 1↑ 1↓)
datafusion / parquet (0.923x ➖, 1↑ 0↓)
duckdb / vortex-file-compressed (0.996x ➖, 0↑ 0↓)
duckdb / vortex-compact (0.989x ➖, 0↑ 0↓)
duckdb / parquet (0.943x ➖, 0↑ 0↓)
|
…ing-1eKcy Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk> # Conflicts: # vortex-array/public-api.lock # vortex/public-api.lock
joseph-isaacs
added
the
action/benchmark
github-actions
Bot
removed
the
action/benchmark
Contributor
Heap allocates what exactly? Null-as-false is a perfectly valid scalar function. We should just treat it like one. You're over-optimizing if you're trying to avoid wrapping this up in a ScalarFnArray. |
Contributor
|
This PR has been marked as stale because it has been open for 14 days with no activity. Please comment or remove the stale label if you wish to keep it active, otherwise it will be closed in 7 days |
Contributor
Author
|
@robert3005 thoughts |
Contributor
|
I recently optimised in place owned bitbuffer operations such that things like null as false will attempt to reuse the existing buffer where possible. Is there still a difference then? |
Contributor
Author
|
Yes the semantics are the ONLY important change here |
…ing-1eKcy Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk> # Conflicts: # vortex-buffer/src/bit/buf.rs # vortex-layout/src/layouts/zoned/zone_map.rs
Contributor
|
Ok by what are we saving by this not being a scalar function? I think we should change the semantics of mask to require non nullable bool but then... I think I agree with @gatesn, why is this another executable target? |
Remove the NullAsFalse executor adapter and its supporting buffer helpers. Mask::execute keeps rejecting nullable boolean arrays; callers that need null-as-false semantics now call array.fill_null(false) before executing into a Mask. Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk>
onursatici
approved these changes
Jun 15, 2026
robert3005
approved these changes
Jun 15, 2026
Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk>
joseph-isaacs
added
changelog/break
joseph-isaacs
changed the title
Mask execution strict about nullable booleans
Merging this PR will improve performance by 10.85%
|
| Mode | Benchmark | BASE |
HEAD |
Efficiency | |
|---|---|---|---|---|---|
| ⚡ | Simulation | encode_varbin[(1000, 2)] |
175.3 µs | 158.2 µs | +10.85% |
Tip
Curious why this is faster? Comment @codspeedbot explain why this is faster on this PR, or directly use the CodSpeed MCP with your agent.
Comparing claude/awesome-turing-1eKcy (7306d0d) with develop (739b745)
Footnotes
-
10 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. ↩
joseph-isaacs
added
the
action/benchmark
github-actions
Bot
removed
the
action/benchmark
Contributor
Benchmarks: Appian on NVMEVerdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.999x ➖, 0↑ 0↓)
datafusion / parquet (0.913x ➖, 2↑ 0↓)
duckdb / vortex-file-compressed (1.032x ➖, 0↑ 0↓)
duckdb / parquet (1.004x ➖, 0↑ 0↓)
duckdb / duckdb (0.966x ➖, 0↑ 0↓)
File Size Changes (4 files changed, -0.1% overall, 0↑ 4↓)
Totals:
|
joseph-isaacs
added a commit
that referenced
this pull request
Jun 17, 2026
Replace the execute_mask_coercing_nulls helper with an Executor-style `ArrayRef::null_as_false()` adapter (matching the design in #8121's history): canonicalize once and fold validity into the value bits with a single AND that reuses the value buffer when uniquely owned, deferring non-nullable input to the strict `Mask::execute`. Add the mask_fill_null microbenchmark and a one-line note that `fill_null` on a lazy `ScalarFn` array is currently slow. Verified on CI ClickBench: clickbench_q22/duckdb:vortex 1126ms -> 580ms (0.51, ~2x). Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk> Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
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
5 participants
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.
Summary
Previously
Mask::executeaccepted nullable boolean arrays and silentlycoerced null elements to
false(folding validity into the value bits). Thatbehaviour was flagged with a
TODOquestioning whether the implicit coercionwas correct.
This PR makes
Maskexecution strict:Mask::executenow requires anon-nullable boolean array (
Bool(NonNullable)) and errors on nullable input.The null-as-
falsecoercion is no longer implicit — callers that wantSQL-style
NULL-as-not-matching semantics opt in explicitly viaarray.fill_null(false)before executing into aMask.