andygrove opened a new pull request, #23538:
URL: https://github.com/apache/datafusion/pull/23538
> This PR was created by an LLM as a draft PR. I will mark it as ready for
review after human review.
## Which issue does this PR close?
N/A — autonomous exploratory PR.
## Rationale for this change
Vectorize greatest's nullable comparison path — replace the per-row boxed
make_comparator with cmp::gt_eq plus bitwise null masking, keeping the
comparator only for floats where NaN ordering differs.
## What changes are included in this PR?
Vectorize greatest's nullable comparison path — replace the per-row boxed
make_comparator with cmp::gt_eq plus bitwise null masking, keeping the
comparator only for floats where NaN ordering differs.
## Are these changes tested?
Correctness: unit tests + seeded differential fuzz (bit-identical Arrow
output vs `main`).
Benchmark (criterion):
- greatest_utf8_nullable: 9.023% faster (base 10651ns -> cand 9690ns)
- greatest_i64_nullable: 21.744% faster (base 7310ns -> cand 5721ns)
- greatest_i64_nullable_3_args: 21.667% faster (base 14078ns -> cand 11027ns)
- greatest_i64_no_nulls: 1.398% faster (base 3163ns -> cand 3119ns)
Full criterion output:
```text
greatest_i64_nullable time: [5.7071 µs 5.7111 µs 5.7164 µs]
change: [−21.848% −21.744% −21.617%] (p = 0.00 <
0.05)
Performance has improved.
Found 5 outliers among 100 measurements (5.00%)
1 (1.00%) high mild
4 (4.00%) high severe
greatest_i64_no_nulls time: [3.1184 µs 3.1210 µs 3.1238 µs]
change: [−1.4889% −1.3979% −1.3102%] (p = 0.00 <
0.05)
Performance has improved.
Found 3 outliers among 100 measurements (3.00%)
2 (2.00%) high mild
1 (1.00%) high severe
greatest_i64_nullable_3_args
time: [11.012 µs 11.022 µs 11.033 µs]
change: [−21.773% −21.667% −21.583%] (p = 0.00 <
0.05)
Performance has improved.
Found 13 outliers among 100 measurements (13.00%)
12 (12.00%) high mild
1 (1.00%) high severe
greatest_utf8_nullable time: [9.6840 µs 9.6950 µs 9.7074 µs]
change: [−9.2000% −9.0226% −8.8658%] (p = 0.00 <
0.05)
Performance has improved.
Found 1 outliers among 100 measurements (1.00%)
1 (1.00%) high severe
```
## Are there any user-facing changes?
<!--
If there are user-facing changes then we may require documentation to be
updated before approving the PR.
-->
<!--
If there are any breaking changes to public APIs, please add the `api
change` label.
-->
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
