You are viewing a plain text version of this content. The canonical link for it is here.
Posted to github@datafusion.apache.org by "raushanprabhakar1 (via GitHub)" <gi...@apache.org> on 2026/04/23 15:32:13 UTC
[PR] Adding Use of arrow's has_true() / has_false() [datafusion]
raushanprabhakar1 opened a new pull request, #21806:
URL: https://github.com/apache/datafusion/pull/21806
## Which issue does this PR close?
Closes #21784
## Rationale for this change
Apache Arrow added `BooleanArray::has_true()` and `has_false()` so callers can answer “any true/false?” without a full bit count. That can short-circuit and avoid unnecessary work compared to patterns like `true_count() == 0` or `true_count() > 0`.
This PR applies those APIs across DataFusion where the logic is purely existential (or equivalent via null-safe “all true” / “no true” checks), matching the audit suggested in the issue.
## What changes are included in this PR?
- Replace hot-path checks that only needed existence or emptiness with `has_true()` / `has_false()` (and `null_count()` where needed), including:
- Nested/array helpers (`array_has`, list replace), Spark `array_contains` null-semantics fast path
- Physical expressions: `evaluate_selection`, binary AND/OR short-circuit, CASE/IN list loops
- `scatter` fast paths
- Top-K filter handling, sort-merge join filter, nested-loop join bitmap checks
- Parquet column stats (`metadata.rs`, `has_any_exact_match`)
- Keep `true_count()` / `false_count()` where an actual count is required (row counts, metrics, selectivity, `to_array(n)`, etc.)
- Import `arrow::array::Array` where `null_count()` is used on `BooleanArray` in trait-heavy paths
## Are these changes tested?
Existing tests cover this behavior; the edits are semantics-preserving refactors (same conditions, cheaper primitives). No new tests were added.
## Are there any user-facing changes?
No. Behavior should be unchanged; this is an internal performance/clarity improvement.
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4348225812
🤖 Benchmark completed (GKE) | [trigger](https://github.com/apache/datafusion/pull/21806#issuecomment-4348058676)
**Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB)
<details><summary>CPU Details (lscpu)</summary>
```
Architecture: aarch64
CPU op-mode(s): 64-bit
Byte Order: Little Endian
CPU(s): 16
On-line CPU(s) list: 0-15
Vendor ID: ARM
Model name: Neoverse-V2
Model: 1
Thread(s) per core: 1
Core(s) per cluster: 16
Socket(s): -
Cluster(s): 1
Stepping: r0p1
BogoMIPS: 2000.00
Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache: 1 MiB (16 instances)
L1i cache: 1 MiB (16 instances)
L2 cache: 32 MiB (16 instances)
L3 cache: 80 MiB (1 instance)
NUMA node(s): 1
NUMA node0 CPU(s): 0-15
Vulnerability Gather data sampling: Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Not affected
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed: Not affected
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1: Mitigation; __user pointer sanitization
Vulnerability Spectre v2: Mitigation; CSV2, BHB
Vulnerability Srbds: Not affected
Vulnerability Tsa: Not affected
Vulnerability Tsx async abort: Not affected
Vulnerability Vmscape: Not affected
```
</details>
<details><summary>Details</summary>
<p>
```
Comparing HEAD and feat_21784-has-true-has-false
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query ┃ HEAD ┃ feat_21784-has-true-has-false ┃ Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ QQuery 0 │ 1.25 / 4.77 ±6.89 / 18.55 ms │ 1.20 / 4.60 ±6.71 / 18.02 ms │ no change │
│ QQuery 1 │ 12.67 / 12.98 ±0.23 / 13.29 ms │ 12.16 / 12.57 ±0.27 / 12.95 ms │ no change │
│ QQuery 2 │ 37.25 / 37.63 ±0.35 / 38.07 ms │ 36.55 / 36.84 ±0.23 / 37.14 ms │ no change │
│ QQuery 3 │ 32.55 / 33.29 ±0.78 / 34.68 ms │ 31.92 / 32.54 ±0.73 / 33.71 ms │ no change │
│ QQuery 4 │ 246.79 / 251.06 ±4.19 / 258.64 ms │ 239.92 / 243.54 ±2.30 / 247.03 ms │ no change │
│ QQuery 5 │ 285.82 / 288.49 ±2.42 / 292.65 ms │ 281.72 / 284.40 ±2.22 / 287.10 ms │ no change │
│ QQuery 6 │ 6.54 / 7.46 ±0.71 / 8.60 ms │ 6.60 / 7.35 ±0.70 / 8.63 ms │ no change │
│ QQuery 7 │ 13.64 / 13.88 ±0.20 / 14.21 ms │ 13.46 / 13.78 ±0.17 / 13.94 ms │ no change │
│ QQuery 8 │ 333.20 / 335.18 ±1.49 / 336.58 ms │ 332.52 / 335.96 ±3.69 / 342.85 ms │ no change │
│ QQuery 9 │ 454.33 / 459.01 ±2.82 / 461.88 ms │ 457.98 / 460.35 ±1.92 / 462.66 ms │ no change │
│ QQuery 10 │ 73.64 / 74.36 ±0.76 / 75.32 ms │ 72.70 / 77.93 ±8.99 / 95.89 ms │ no change │
│ QQuery 11 │ 84.63 / 86.13 ±2.29 / 90.66 ms │ 84.28 / 85.45 ±1.08 / 87.12 ms │ no change │
│ QQuery 12 │ 278.16 / 280.50 ±2.81 / 286.00 ms │ 273.65 / 277.86 ±4.30 / 285.98 ms │ no change │
│ QQuery 13 │ 393.71 / 401.84 ±4.27 / 405.86 ms │ 393.61 / 401.48 ±4.98 / 406.99 ms │ no change │
│ QQuery 14 │ 288.97 / 290.99 ±1.46 / 293.35 ms │ 286.56 / 290.90 ±2.63 / 294.02 ms │ no change │
│ QQuery 15 │ 287.12 / 294.85 ±10.12 / 314.52 ms │ 285.50 / 290.59 ±5.84 / 302.00 ms │ no change │
│ QQuery 16 │ 620.86 / 628.16 ±5.32 / 634.45 ms │ 632.19 / 634.24 ±2.68 / 639.28 ms │ no change │
│ QQuery 17 │ 624.46 / 629.74 ±5.39 / 637.37 ms │ 625.50 / 638.59 ±10.23 / 651.27 ms │ no change │
│ QQuery 18 │ 1258.58 / 1277.49 ±13.92 / 1301.52 ms │ 1261.06 / 1288.98 ±29.10 / 1336.70 ms │ no change │
│ QQuery 19 │ 29.08 / 29.49 ±0.36 / 29.91 ms │ 28.77 / 30.25 ±2.36 / 34.95 ms │ no change │
│ QQuery 20 │ 531.14 / 538.57 ±9.61 / 556.15 ms │ 527.99 / 533.25 ±5.42 / 540.32 ms │ no change │
│ QQuery 21 │ 600.32 / 605.02 ±5.77 / 615.85 ms │ 601.40 / 609.00 ±6.38 / 620.28 ms │ no change │
│ QQuery 22 │ 1072.00 / 1076.98 ±6.54 / 1089.78 ms │ 1073.81 / 1077.15 ±2.29 / 1080.75 ms │ no change │
│ QQuery 23 │ 3354.87 / 3381.10 ±21.28 / 3402.25 ms │ 3352.12 / 3371.95 ±20.98 / 3410.07 ms │ no change │
│ QQuery 24 │ 42.54 / 44.70 ±2.89 / 50.43 ms │ 42.40 / 45.09 ±4.70 / 54.45 ms │ no change │
│ QQuery 25 │ 114.54 / 115.47 ±1.07 / 117.01 ms │ 114.28 / 116.66 ±3.11 / 122.71 ms │ no change │
│ QQuery 26 │ 42.98 / 43.96 ±0.61 / 44.59 ms │ 43.03 / 44.39 ±1.49 / 47.20 ms │ no change │
│ QQuery 27 │ 678.43 / 681.44 ±2.51 / 685.89 ms │ 676.76 / 684.06 ±8.13 / 699.63 ms │ no change │
│ QQuery 28 │ 3006.89 / 3021.10 ±11.11 / 3040.18 ms │ 3008.35 / 3016.07 ±7.62 / 3028.96 ms │ no change │
│ QQuery 29 │ 42.53 / 52.26 ±14.65 / 80.69 ms │ 42.20 / 50.30 ±11.81 / 73.47 ms │ no change │
│ QQuery 30 │ 313.46 / 322.54 ±8.22 / 332.60 ms │ 307.94 / 321.52 ±10.90 / 333.94 ms │ no change │
│ QQuery 31 │ 313.50 / 315.21 ±1.60 / 317.46 ms │ 306.75 / 324.06 ±18.87 / 360.72 ms │ no change │
│ QQuery 32 │ 1020.41 / 1030.82 ±9.28 / 1048.27 ms │ 1012.91 / 1021.36 ±9.08 / 1036.09 ms │ no change │
│ QQuery 33 │ 1448.40 / 1466.26 ±15.34 / 1488.04 ms │ 1425.62 / 1467.74 ±28.04 / 1504.81 ms │ no change │
│ QQuery 34 │ 1453.43 / 1472.32 ±23.16 / 1517.56 ms │ 1452.02 / 1484.40 ±21.10 / 1517.40 ms │ no change │
│ QQuery 35 │ 294.53 / 311.22 ±14.50 / 331.27 ms │ 286.64 / 313.16 ±26.96 / 350.86 ms │ no change │
│ QQuery 36 │ 65.04 / 73.27 ±6.12 / 81.85 ms │ 65.93 / 71.83 ±7.93 / 87.35 ms │ no change │
│ QQuery 37 │ 36.32 / 39.37 ±2.76 / 43.24 ms │ 35.67 / 39.58 ±3.18 / 43.49 ms │ no change │
│ QQuery 38 │ 40.82 / 46.22 ±4.49 / 53.43 ms │ 41.55 / 44.63 ±2.85 / 49.90 ms │ no change │
│ QQuery 39 │ 122.95 / 133.94 ±7.25 / 143.16 ms │ 126.46 / 134.35 ±4.37 / 139.85 ms │ no change │
│ QQuery 40 │ 14.32 / 15.47 ±1.17 / 17.65 ms │ 14.54 / 17.49 ±4.51 / 26.44 ms │ 1.13x slower │
│ QQuery 41 │ 13.80 / 14.21 ±0.30 / 14.53 ms │ 13.82 / 14.17 ±0.51 / 15.18 ms │ no change │
│ QQuery 42 │ 13.38 / 14.41 ±1.79 / 17.97 ms │ 13.20 / 16.67 ±3.96 / 22.57 ms │ 1.16x slower │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary ┃ ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD) │ 20253.17ms │
│ Total Time (feat_21784-has-true-has-false) │ 20267.05ms │
│ Average Time (HEAD) │ 471.00ms │
│ Average Time (feat_21784-has-true-has-false) │ 471.33ms │
│ Queries Faster │ 0 │
│ Queries Slower │ 2 │
│ Queries with No Change │ 41 │
│ Queries with Failure │ 0 │
└──────────────────────────────────────────────┴────────────┘
```
</p>
</details>
<details><summary>Resource Usage</summary>
**clickbench_partitioned — base (merge-base)**
| Metric | Value |
|--------|-------|
| Wall time | 105.0s |
| Peak memory | 29.9 GiB |
| Avg memory | 22.9 GiB |
| CPU user | 1066.5s |
| CPU sys | 68.6s |
| Peak spill | 0 B |
**clickbench_partitioned — branch**
| Metric | Value |
|--------|-------|
| Wall time | 105.0s |
| Peak memory | 29.7 GiB |
| Avg memory | 22.8 GiB |
| CPU user | 1065.4s |
| CPU sys | 69.8s |
| Peak spill | 0 B |
</details>
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "raushanprabhakar1 (via GitHub)" <gi...@apache.org>.
raushanprabhakar1 commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4355734522
Hi @adriangb , Thanks for your feedback, i have addressed all your feedbacks, please do review, Thanks.
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4347924313
Hi @adriangb, thanks for the request (https://github.com/apache/datafusion/pull/21806#issuecomment-4347924201). You already have 15 pending benchmark job(s), and this request would add 3 more — exceeding the per-user queue limit of 15. Please wait for some of your current runs to finish before submitting more.
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4348155253
🤖 Benchmark running (GKE) | [trigger](https://github.com/apache/datafusion/pull/21806#issuecomment-4348058676)
**Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB) | `Linux bench-c4348058676-1945-txjlk 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux`
<details><summary>CPU Details (lscpu)</summary>
```
Architecture: aarch64
CPU op-mode(s): 64-bit
Byte Order: Little Endian
CPU(s): 16
On-line CPU(s) list: 0-15
Vendor ID: ARM
Model name: Neoverse-V2
Model: 1
Thread(s) per core: 1
Core(s) per cluster: 16
Socket(s): -
Cluster(s): 1
Stepping: r0p1
BogoMIPS: 2000.00
Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache: 1 MiB (16 instances)
L1i cache: 1 MiB (16 instances)
L2 cache: 32 MiB (16 instances)
L3 cache: 80 MiB (1 instance)
NUMA node(s): 1
NUMA node0 CPU(s): 0-15
Vulnerability Gather data sampling: Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Not affected
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed: Not affected
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1: Mitigation; __user pointer sanitization
Vulnerability Spectre v2: Mitigation; CSV2, BHB
Vulnerability Srbds: Not affected
Vulnerability Tsa: Not affected
Vulnerability Tsx async abort: Not affected
Vulnerability Vmscape: Not affected
```
</details>
Comparing feat/21784-has-true-has-false (8777c0987e264b56f007f310b4b494e6f0a39293) to 42cd2fa (merge-base) [diff](https://github.com/apache/datafusion/compare/42cd2fa90d18a8a54f433fc94917c20a4d1e7338..8777c0987e264b56f007f310b4b494e6f0a39293) using: tpcds
Results will be posted here when complete
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangb (via GitHub)" <gi...@apache.org>.
adriangb commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4305985687
run benchmarks
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4348227501
🤖 Benchmark completed (GKE) | [trigger](https://github.com/apache/datafusion/pull/21806#issuecomment-4348058676)
**Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB)
<details><summary>CPU Details (lscpu)</summary>
```
Architecture: aarch64
CPU op-mode(s): 64-bit
Byte Order: Little Endian
CPU(s): 16
On-line CPU(s) list: 0-15
Vendor ID: ARM
Model name: Neoverse-V2
Model: 1
Thread(s) per core: 1
Core(s) per cluster: 16
Socket(s): -
Cluster(s): 1
Stepping: r0p1
BogoMIPS: 2000.00
Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache: 1 MiB (16 instances)
L1i cache: 1 MiB (16 instances)
L2 cache: 32 MiB (16 instances)
L3 cache: 80 MiB (1 instance)
NUMA node(s): 1
NUMA node0 CPU(s): 0-15
Vulnerability Gather data sampling: Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Not affected
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed: Not affected
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1: Mitigation; __user pointer sanitization
Vulnerability Spectre v2: Mitigation; CSV2, BHB
Vulnerability Srbds: Not affected
Vulnerability Tsa: Not affected
Vulnerability Tsx async abort: Not affected
Vulnerability Vmscape: Not affected
```
</details>
<details><summary>Details</summary>
<p>
```
Comparing HEAD and feat_21784-has-true-has-false
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query ┃ HEAD ┃ feat_21784-has-true-has-false ┃ Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1 │ 7.27 / 7.74 ±0.79 / 9.32 ms │ 7.07 / 7.60 ±0.85 / 9.30 ms │ no change │
│ QQuery 2 │ 147.29 / 147.64 ±0.28 / 148.12 ms │ 143.96 / 144.64 ±0.37 / 145.09 ms │ no change │
│ QQuery 3 │ 113.67 / 114.63 ±0.91 / 116.36 ms │ 112.17 / 113.65 ±0.94 / 114.61 ms │ no change │
│ QQuery 4 │ 1304.13 / 1310.93 ±5.63 / 1318.78 ms │ 1283.98 / 1296.21 ±7.54 / 1307.21 ms │ no change │
│ QQuery 5 │ 173.99 / 175.62 ±1.45 / 178.29 ms │ 172.43 / 173.47 ±0.81 / 174.51 ms │ no change │
│ QQuery 6 │ 136.85 / 141.99 ±3.38 / 146.42 ms │ 135.02 / 139.03 ±3.46 / 144.49 ms │ no change │
│ QQuery 7 │ 336.40 / 338.47 ±1.60 / 341.13 ms │ 332.53 / 335.42 ±2.55 / 339.78 ms │ no change │
│ QQuery 8 │ 113.36 / 114.19 ±0.70 / 115.08 ms │ 113.00 / 113.48 ±0.55 / 114.54 ms │ no change │
│ QQuery 9 │ 97.33 / 101.56 ±2.97 / 105.23 ms │ 95.58 / 105.47 ±12.75 / 128.24 ms │ no change │
│ QQuery 10 │ 101.94 / 102.59 ±0.53 / 103.31 ms │ 101.60 / 102.32 ±0.61 / 103.40 ms │ no change │
│ QQuery 11 │ 903.54 / 913.93 ±6.40 / 923.40 ms │ 888.86 / 896.47 ±6.36 / 908.17 ms │ no change │
│ QQuery 12 │ 44.13 / 44.37 ±0.24 / 44.77 ms │ 43.08 / 43.76 ±0.44 / 44.32 ms │ no change │
│ QQuery 13 │ 388.45 / 392.26 ±2.53 / 396.31 ms │ 389.24 / 392.08 ±1.68 / 393.69 ms │ no change │
│ QQuery 14 │ 1044.85 / 1048.70 ±2.47 / 1051.28 ms │ 1027.35 / 1033.11 ±3.48 / 1038.19 ms │ no change │
│ QQuery 15 │ 14.83 / 15.23 ±0.44 / 16.08 ms │ 14.57 / 15.22 ±0.48 / 15.91 ms │ no change │
│ QQuery 16 │ 7.47 / 7.62 ±0.14 / 7.86 ms │ 7.36 / 7.42 ±0.11 / 7.64 ms │ no change │
│ QQuery 17 │ 221.74 / 223.60 ±1.28 / 225.59 ms │ 220.97 / 222.20 ±1.90 / 225.95 ms │ no change │
│ QQuery 18 │ 127.48 / 128.32 ±0.73 / 129.38 ms │ 125.57 / 126.86 ±1.21 / 128.69 ms │ no change │
│ QQuery 19 │ 154.56 / 157.39 ±2.51 / 161.93 ms │ 152.63 / 154.48 ±1.48 / 157.14 ms │ no change │
│ QQuery 20 │ 13.32 / 13.47 ±0.14 / 13.70 ms │ 12.87 / 13.33 ±0.32 / 13.77 ms │ no change │
│ QQuery 21 │ 19.83 / 19.91 ±0.09 / 20.03 ms │ 18.93 / 19.07 ±0.12 / 19.19 ms │ no change │
│ QQuery 22 │ 477.08 / 482.82 ±3.29 / 487.03 ms │ 479.17 / 481.97 ±2.30 / 485.29 ms │ no change │
│ QQuery 23 │ 1100.12 / 1103.74 ±3.31 / 1109.28 ms │ 1091.12 / 1096.98 ±5.21 / 1104.20 ms │ no change │
│ QQuery 24 │ 714.70 / 718.61 ±2.52 / 722.40 ms │ 705.34 / 711.20 ±3.50 / 715.98 ms │ no change │
│ QQuery 25 │ 333.31 / 334.63 ±1.37 / 337.23 ms │ 332.11 / 336.36 ±3.82 / 342.14 ms │ no change │
│ QQuery 26 │ 78.06 / 79.56 ±1.48 / 82.34 ms │ 77.85 / 78.21 ±0.41 / 78.93 ms │ no change │
│ QQuery 27 │ 7.31 / 7.42 ±0.18 / 7.77 ms │ 6.98 / 7.10 ±0.17 / 7.43 ms │ no change │
│ QQuery 28 │ 149.39 / 150.39 ±0.97 / 151.98 ms │ 148.71 / 151.21 ±2.93 / 156.53 ms │ no change │
│ QQuery 29 │ 274.22 / 278.42 ±4.13 / 285.45 ms │ 273.14 / 274.68 ±1.93 / 278.30 ms │ no change │
│ QQuery 30 │ 42.22 / 42.61 ±0.45 / 43.20 ms │ 41.95 / 42.74 ±1.12 / 44.95 ms │ no change │
│ QQuery 31 │ 167.05 / 169.78 ±3.05 / 174.98 ms │ 164.49 / 165.59 ±1.03 / 167.00 ms │ no change │
│ QQuery 32 │ 13.94 / 14.24 ±0.21 / 14.48 ms │ 13.40 / 13.63 ±0.27 / 14.07 ms │ no change │
│ QQuery 33 │ 139.14 / 140.98 ±2.00 / 144.70 ms │ 139.19 / 140.93 ±2.23 / 145.17 ms │ no change │
│ QQuery 34 │ 7.22 / 7.38 ±0.19 / 7.75 ms │ 6.91 / 7.18 ±0.25 / 7.65 ms │ no change │
│ QQuery 35 │ 101.34 / 102.27 ±0.64 / 103.29 ms │ 100.32 / 102.92 ±2.45 / 107.52 ms │ no change │
│ QQuery 36 │ 6.86 / 7.09 ±0.19 / 7.34 ms │ 6.62 / 6.82 ±0.14 / 6.99 ms │ no change │
│ QQuery 37 │ 8.32 / 8.46 ±0.14 / 8.73 ms │ 8.35 / 8.57 ±0.18 / 8.80 ms │ no change │
│ QQuery 38 │ 86.38 / 88.17 ±2.03 / 91.92 ms │ 84.52 / 85.95 ±0.97 / 86.89 ms │ no change │
│ QQuery 39 │ 119.02 / 120.67 ±1.30 / 122.88 ms │ 118.24 / 122.24 ±6.14 / 134.26 ms │ no change │
│ QQuery 40 │ 103.23 / 105.42 ±1.75 / 107.39 ms │ 102.33 / 106.72 ±5.19 / 116.52 ms │ no change │
│ QQuery 41 │ 14.83 / 15.14 ±0.27 / 15.59 ms │ 14.51 / 14.75 ±0.22 / 15.17 ms │ no change │
│ QQuery 42 │ 107.46 / 107.98 ±0.49 / 108.81 ms │ 106.17 / 108.69 ±2.01 / 111.84 ms │ no change │
│ QQuery 43 │ 5.89 / 6.03 ±0.13 / 6.27 ms │ 5.75 / 5.91 ±0.17 / 6.23 ms │ no change │
│ QQuery 44 │ 11.24 / 12.38 ±2.06 / 16.49 ms │ 10.75 / 11.01 ±0.21 / 11.36 ms │ +1.12x faster │
│ QQuery 45 │ 49.34 / 49.82 ±0.35 / 50.33 ms │ 48.66 / 49.08 ±0.30 / 49.52 ms │ no change │
│ QQuery 46 │ 8.88 / 9.01 ±0.10 / 9.17 ms │ 8.31 / 8.61 ±0.21 / 8.96 ms │ no change │
│ QQuery 47 │ 697.15 / 706.33 ±5.20 / 713.04 ms │ 690.80 / 694.20 ±2.11 / 696.14 ms │ no change │
│ QQuery 48 │ 273.24 / 276.13 ±3.59 / 282.54 ms │ 274.01 / 277.38 ±4.24 / 285.60 ms │ no change │
│ QQuery 49 │ 249.62 / 252.26 ±1.96 / 255.62 ms │ 248.46 / 249.64 ±1.14 / 251.69 ms │ no change │
│ QQuery 50 │ 204.41 / 209.26 ±5.31 / 219.61 ms │ 202.39 / 206.36 ±4.83 / 215.69 ms │ no change │
│ QQuery 51 │ 176.61 / 180.39 ±2.70 / 184.92 ms │ 175.29 / 178.76 ±1.96 / 181.37 ms │ no change │
│ QQuery 52 │ 107.84 / 108.54 ±0.65 / 109.75 ms │ 107.25 / 109.08 ±1.87 / 112.24 ms │ no change │
│ QQuery 53 │ 103.91 / 105.57 ±2.51 / 110.53 ms │ 101.84 / 102.33 ±0.52 / 103.30 ms │ no change │
│ QQuery 54 │ 145.48 / 148.43 ±2.42 / 152.88 ms │ 144.64 / 146.93 ±1.78 / 150.09 ms │ no change │
│ QQuery 55 │ 106.38 / 107.42 ±0.83 / 108.16 ms │ 105.60 / 106.30 ±0.62 / 107.31 ms │ no change │
│ QQuery 56 │ 139.93 / 141.76 ±1.08 / 142.84 ms │ 139.89 / 140.59 ±0.88 / 142.15 ms │ no change │
│ QQuery 57 │ 167.60 / 169.49 ±1.48 / 172.02 ms │ 164.04 / 166.52 ±1.94 / 169.70 ms │ no change │
│ QQuery 58 │ 267.63 / 270.05 ±2.13 / 273.93 ms │ 265.49 / 267.23 ±1.12 / 268.42 ms │ no change │
│ QQuery 59 │ 194.69 / 196.91 ±1.94 / 200.25 ms │ 193.20 / 194.71 ±1.87 / 198.34 ms │ no change │
│ QQuery 60 │ 141.95 / 143.73 ±1.68 / 146.93 ms │ 141.16 / 142.79 ±1.30 / 144.63 ms │ no change │
│ QQuery 61 │ 13.97 / 14.11 ±0.13 / 14.34 ms │ 13.58 / 13.74 ±0.14 / 13.98 ms │ no change │
│ QQuery 62 │ 865.93 / 871.87 ±4.50 / 877.29 ms │ 860.59 / 865.76 ±4.50 / 871.70 ms │ no change │
│ QQuery 63 │ 103.09 / 104.58 ±1.60 / 107.61 ms │ 102.59 / 104.20 ±2.00 / 107.76 ms │ no change │
│ QQuery 64 │ 660.95 / 666.67 ±5.01 / 675.67 ms │ 656.59 / 659.89 ±2.74 / 663.94 ms │ no change │
│ QQuery 65 │ 246.01 / 247.03 ±0.88 / 248.50 ms │ 242.74 / 247.38 ±4.36 / 255.49 ms │ no change │
│ QQuery 66 │ 216.89 / 222.75 ±6.45 / 232.60 ms │ 213.74 / 221.82 ±9.47 / 235.76 ms │ no change │
│ QQuery 67 │ 295.02 / 299.82 ±5.35 / 309.86 ms │ 293.56 / 299.67 ±6.04 / 310.97 ms │ no change │
│ QQuery 68 │ 8.79 / 9.09 ±0.25 / 9.47 ms │ 8.61 / 8.93 ±0.23 / 9.31 ms │ no change │
│ QQuery 69 │ 97.64 / 98.94 ±0.88 / 100.18 ms │ 97.39 / 98.13 ±0.49 / 98.91 ms │ no change │
│ QQuery 70 │ 312.08 / 319.54 ±5.43 / 327.08 ms │ 316.60 / 325.05 ±10.90 / 346.05 ms │ no change │
│ QQuery 71 │ 134.50 / 135.68 ±1.13 / 137.33 ms │ 133.17 / 135.42 ±2.43 / 140.13 ms │ no change │
│ QQuery 72 │ 579.07 / 591.05 ±7.67 / 601.63 ms │ 580.35 / 595.72 ±9.11 / 607.12 ms │ no change │
│ QQuery 73 │ 6.93 / 7.09 ±0.22 / 7.52 ms │ 6.60 / 6.76 ±0.25 / 7.26 ms │ no change │
│ QQuery 74 │ 574.77 / 576.46 ±0.96 / 577.71 ms │ 571.37 / 577.29 ±5.33 / 585.82 ms │ no change │
│ QQuery 75 │ 268.15 / 270.46 ±2.04 / 273.49 ms │ 268.35 / 269.83 ±1.19 / 271.29 ms │ no change │
│ QQuery 76 │ 130.74 / 132.43 ±1.91 / 135.09 ms │ 129.26 / 130.97 ±1.31 / 132.97 ms │ no change │
│ QQuery 77 │ 187.90 / 189.14 ±1.12 / 190.80 ms │ 187.06 / 187.62 ±0.52 / 188.53 ms │ no change │
│ QQuery 78 │ 332.01 / 333.94 ±1.83 / 336.83 ms │ 329.96 / 331.61 ±1.61 / 334.46 ms │ no change │
│ QQuery 79 │ 231.83 / 233.66 ±1.98 / 237.46 ms │ 228.97 / 231.16 ±1.82 / 234.33 ms │ no change │
│ QQuery 80 │ 317.42 / 321.96 ±2.40 / 324.48 ms │ 317.62 / 320.92 ±2.27 / 323.25 ms │ no change │
│ QQuery 81 │ 26.13 / 26.48 ±0.29 / 27.00 ms │ 25.79 / 26.27 ±0.52 / 27.23 ms │ no change │
│ QQuery 82 │ 39.82 / 40.20 ±0.37 / 40.81 ms │ 39.25 / 39.88 ±0.60 / 40.89 ms │ no change │
│ QQuery 83 │ 37.59 / 38.09 ±0.38 / 38.66 ms │ 37.10 / 37.29 ±0.11 / 37.44 ms │ no change │
│ QQuery 84 │ 46.61 / 47.96 ±1.58 / 51.05 ms │ 46.39 / 47.70 ±1.99 / 51.53 ms │ no change │
│ QQuery 85 │ 142.17 / 143.30 ±0.87 / 144.83 ms │ 141.02 / 142.16 ±1.49 / 145.04 ms │ no change │
│ QQuery 86 │ 37.95 / 38.87 ±1.16 / 41.06 ms │ 36.86 / 37.47 ±0.38 / 38.02 ms │ no change │
│ QQuery 87 │ 3.65 / 3.77 ±0.18 / 4.13 ms │ 3.46 / 3.58 ±0.17 / 3.91 ms │ +1.05x faster │
│ QQuery 88 │ 100.01 / 104.96 ±8.93 / 122.81 ms │ 99.96 / 100.19 ±0.22 / 100.59 ms │ no change │
│ QQuery 89 │ 116.65 / 119.37 ±3.01 / 125.24 ms │ 115.54 / 118.75 ±4.97 / 128.61 ms │ no change │
│ QQuery 90 │ 22.63 / 23.02 ±0.44 / 23.79 ms │ 22.52 / 23.26 ±0.48 / 23.76 ms │ no change │
│ QQuery 91 │ 58.67 / 59.16 ±0.29 / 59.55 ms │ 58.20 / 58.55 ±0.26 / 58.98 ms │ no change │
│ QQuery 92 │ 56.86 / 57.28 ±0.31 / 57.74 ms │ 56.32 / 57.07 ±0.61 / 57.90 ms │ no change │
│ QQuery 93 │ 181.43 / 182.07 ±0.52 / 182.98 ms │ 182.06 / 183.51 ±1.53 / 186.40 ms │ no change │
│ QQuery 94 │ 61.39 / 62.09 ±0.54 / 63.00 ms │ 61.59 / 62.29 ±1.15 / 64.58 ms │ no change │
│ QQuery 95 │ 126.04 / 127.30 ±1.10 / 129.35 ms │ 126.40 / 127.86 ±1.06 / 129.14 ms │ no change │
│ QQuery 96 │ 68.77 / 70.78 ±2.06 / 74.64 ms │ 69.56 / 71.79 ±2.02 / 75.51 ms │ no change │
│ QQuery 97 │ 117.00 / 118.04 ±1.32 / 120.59 ms │ 119.16 / 120.41 ±1.47 / 123.26 ms │ no change │
│ QQuery 98 │ 150.67 / 152.36 ±1.83 / 155.93 ms │ 151.11 / 153.10 ±1.08 / 154.12 ms │ no change │
│ QQuery 99 │ 10737.94 / 10810.67 ±60.47 / 10876.58 ms │ 10832.69 / 11035.10 ±218.26 / 11459.10 ms │ no change │
└───────────┴──────────────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary ┃ ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD) │ 30571.50ms │
│ Total Time (feat_21784-has-true-has-false) │ 30655.25ms │
│ Average Time (HEAD) │ 308.80ms │
│ Average Time (feat_21784-has-true-has-false) │ 309.65ms │
│ Queries Faster │ 2 │
│ Queries Slower │ 0 │
│ Queries with No Change │ 97 │
│ Queries with Failure │ 0 │
└──────────────────────────────────────────────┴────────────┘
```
</p>
</details>
<details><summary>Resource Usage</summary>
**tpcds — base (merge-base)**
| Metric | Value |
|--------|-------|
| Wall time | 155.0s |
| Peak memory | 6.4 GiB |
| Avg memory | 5.7 GiB |
| CPU user | 243.2s |
| CPU sys | 7.7s |
| Peak spill | 0 B |
**tpcds — branch**
| Metric | Value |
|--------|-------|
| Wall time | 155.0s |
| Peak memory | 6.3 GiB |
| Avg memory | 5.7 GiB |
| CPU user | 242.0s |
| CPU sys | 8.7s |
| Peak spill | 0 B |
</details>
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4306006323
🤖 Benchmark running (GKE) | [trigger](https://github.com/apache/datafusion/pull/21806#issuecomment-4305985687)
**Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB) | `Linux bench-c4305985687-1779-7645x 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux`
<details><summary>CPU Details (lscpu)</summary>
```
Architecture: aarch64
CPU op-mode(s): 64-bit
Byte Order: Little Endian
CPU(s): 16
On-line CPU(s) list: 0-15
Vendor ID: ARM
Model name: Neoverse-V2
Model: 1
Thread(s) per core: 1
Core(s) per cluster: 16
Socket(s): -
Cluster(s): 1
Stepping: r0p1
BogoMIPS: 2000.00
Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache: 1 MiB (16 instances)
L1i cache: 1 MiB (16 instances)
L2 cache: 32 MiB (16 instances)
L3 cache: 80 MiB (1 instance)
NUMA node(s): 1
NUMA node0 CPU(s): 0-15
Vulnerability Gather data sampling: Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Not affected
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed: Not affected
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1: Mitigation; __user pointer sanitization
Vulnerability Spectre v2: Mitigation; CSV2, BHB
Vulnerability Srbds: Not affected
Vulnerability Tsa: Not affected
Vulnerability Tsx async abort: Not affected
Vulnerability Vmscape: Not affected
```
</details>
Comparing feat/21784-has-true-has-false (ca7e0a82865af30041a82609b5151266fc1f75d2) to 067ba4b (merge-base) [diff](https://github.com/apache/datafusion/compare/067ba4b02e180ccd3f7e43f087698463d53a566e..ca7e0a82865af30041a82609b5151266fc1f75d2) using: clickbench_partitioned
Results will be posted here when complete
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4352796936
🤖 Benchmark running (GKE) | [trigger](https://github.com/apache/datafusion/pull/21806#issuecomment-4352762538)
**Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB) | `Linux bench-c4352762538-1948-gvkd6 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux`
<details><summary>CPU Details (lscpu)</summary>
```
Architecture: aarch64
CPU op-mode(s): 64-bit
Byte Order: Little Endian
CPU(s): 16
On-line CPU(s) list: 0-15
Vendor ID: ARM
Model name: Neoverse-V2
Model: 1
Thread(s) per core: 1
Core(s) per cluster: 16
Socket(s): -
Cluster(s): 1
Stepping: r0p1
BogoMIPS: 2000.00
Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache: 1 MiB (16 instances)
L1i cache: 1 MiB (16 instances)
L2 cache: 32 MiB (16 instances)
L3 cache: 80 MiB (1 instance)
NUMA node(s): 1
NUMA node0 CPU(s): 0-15
Vulnerability Gather data sampling: Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Not affected
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed: Not affected
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1: Mitigation; __user pointer sanitization
Vulnerability Spectre v2: Mitigation; CSV2, BHB
Vulnerability Srbds: Not affected
Vulnerability Tsa: Not affected
Vulnerability Tsx async abort: Not affected
Vulnerability Vmscape: Not affected
```
</details>
Comparing feat/21784-has-true-has-false (8777c0987e264b56f007f310b4b494e6f0a39293) to 42cd2fa (merge-base) [diff](https://github.com/apache/datafusion/compare/42cd2fa90d18a8a54f433fc94917c20a4d1e7338..8777c0987e264b56f007f310b4b494e6f0a39293) using: tpch
Results will be posted here when complete
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "alamb (via GitHub)" <gi...@apache.org>.
alamb commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4355371528
> > Benchmark for [this request](https://github.com/apache/datafusion/pull/21806#issuecomment-4348058676) failed.
> > Last 20 lines of output:
> > Click to expand
>
> It looks like it can't load the lineitem table
maybe we should add a sanity check to the benchmark runner / better error message (so it fails with a message that says the files aren't there rather than failing on the expected output check) 🤔
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangb (via GitHub)" <gi...@apache.org>.
adriangb commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4348058676
run benchmarks
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangb (via GitHub)" <gi...@apache.org>.
adriangb commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4347924201
run benchmarks
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4352875201
🤖 Benchmark completed (GKE) | [trigger](https://github.com/apache/datafusion/pull/21806#issuecomment-4352762538)
**Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB)
<details><summary>CPU Details (lscpu)</summary>
```
Architecture: aarch64
CPU op-mode(s): 64-bit
Byte Order: Little Endian
CPU(s): 16
On-line CPU(s) list: 0-15
Vendor ID: ARM
Model name: Neoverse-V2
Model: 1
Thread(s) per core: 1
Core(s) per cluster: 16
Socket(s): -
Cluster(s): 1
Stepping: r0p1
BogoMIPS: 2000.00
Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache: 1 MiB (16 instances)
L1i cache: 1 MiB (16 instances)
L2 cache: 32 MiB (16 instances)
L3 cache: 80 MiB (1 instance)
NUMA node(s): 1
NUMA node0 CPU(s): 0-15
Vulnerability Gather data sampling: Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Not affected
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed: Not affected
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1: Mitigation; __user pointer sanitization
Vulnerability Spectre v2: Mitigation; CSV2, BHB
Vulnerability Srbds: Not affected
Vulnerability Tsa: Not affected
Vulnerability Tsx async abort: Not affected
Vulnerability Vmscape: Not affected
```
</details>
<details><summary>Details</summary>
<p>
```
Comparing HEAD and feat_21784-has-true-has-false
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query ┃ HEAD ┃ feat_21784-has-true-has-false ┃ Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1 │ 40.12 / 41.35 ±1.86 / 45.02 ms │ 40.86 / 42.00 ±0.79 / 43.12 ms │ no change │
│ QQuery 2 │ 21.23 / 21.38 ±0.18 / 21.73 ms │ 21.55 / 21.75 ±0.13 / 21.92 ms │ no change │
│ QQuery 3 │ 37.74 / 39.86 ±1.73 / 42.22 ms │ 38.69 / 41.09 ±1.89 / 43.30 ms │ no change │
│ QQuery 4 │ 18.37 / 18.78 ±0.53 / 19.79 ms │ 18.70 / 18.88 ±0.17 / 19.19 ms │ no change │
│ QQuery 5 │ 49.92 / 51.72 ±1.88 / 54.56 ms │ 48.23 / 51.51 ±2.82 / 56.11 ms │ no change │
│ QQuery 6 │ 17.73 / 18.67 ±0.70 / 19.55 ms │ 17.22 / 17.53 ±0.53 / 18.58 ms │ +1.07x faster │
│ QQuery 7 │ 55.37 / 58.03 ±1.53 / 59.41 ms │ 56.16 / 57.93 ±1.89 / 61.54 ms │ no change │
│ QQuery 8 │ 49.52 / 50.08 ±0.37 / 50.58 ms │ 48.64 / 49.24 ±0.38 / 49.76 ms │ no change │
│ QQuery 9 │ 53.76 / 54.49 ±0.58 / 55.41 ms │ 54.95 / 55.69 ±0.49 / 56.48 ms │ no change │
│ QQuery 10 │ 65.37 / 66.69 ±1.08 / 68.60 ms │ 67.03 / 68.45 ±1.21 / 69.95 ms │ no change │
│ QQuery 11 │ 14.87 / 15.19 ±0.17 / 15.37 ms │ 14.85 / 15.62 ±0.65 / 16.64 ms │ no change │
│ QQuery 12 │ 28.50 / 29.38 ±0.68 / 30.32 ms │ 28.59 / 29.47 ±0.99 / 30.82 ms │ no change │
│ QQuery 13 │ 38.37 / 39.08 ±0.78 / 40.60 ms │ 38.68 / 41.59 ±3.43 / 48.26 ms │ 1.06x slower │
│ QQuery 14 │ 27.91 / 28.42 ±0.58 / 29.54 ms │ 28.79 / 29.13 ±0.20 / 29.42 ms │ no change │
│ QQuery 15 │ 32.97 / 33.45 ±0.36 / 34.05 ms │ 34.50 / 35.03 ±0.45 / 35.83 ms │ no change │
│ QQuery 16 │ 15.42 / 15.67 ±0.22 / 16.04 ms │ 15.97 / 16.09 ±0.10 / 16.24 ms │ no change │
│ QQuery 17 │ 78.96 / 84.25 ±3.11 / 87.85 ms │ 85.86 / 86.36 ±0.43 / 87.10 ms │ no change │
│ QQuery 18 │ 75.84 / 77.70 ±1.93 / 80.89 ms │ 80.09 / 80.74 ±0.56 / 81.75 ms │ no change │
│ QQuery 19 │ 37.57 / 38.21 ±0.72 / 39.54 ms │ 39.13 / 39.61 ±0.32 / 40.04 ms │ no change │
│ QQuery 20 │ 40.34 / 43.29 ±3.77 / 50.60 ms │ 41.33 / 42.18 ±0.90 / 43.82 ms │ no change │
│ QQuery 21 │ 64.53 / 68.02 ±2.23 / 70.28 ms │ 65.62 / 67.72 ±1.68 / 70.39 ms │ no change │
│ QQuery 22 │ 25.09 / 25.25 ±0.16 / 25.54 ms │ 25.08 / 25.44 ±0.40 / 26.15 ms │ no change │
└───────────┴────────────────────────────────┴────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary ┃ ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD) │ 918.96ms │
│ Total Time (feat_21784-has-true-has-false) │ 933.03ms │
│ Average Time (HEAD) │ 41.77ms │
│ Average Time (feat_21784-has-true-has-false) │ 42.41ms │
│ Queries Faster │ 1 │
│ Queries Slower │ 1 │
│ Queries with No Change │ 20 │
│ Queries with Failure │ 0 │
└──────────────────────────────────────────────┴──────────┘
```
</p>
</details>
<details><summary>Resource Usage</summary>
**tpch — base (merge-base)**
| Metric | Value |
|--------|-------|
| Wall time | 5.0s |
| Peak memory | 5.4 GiB |
| Avg memory | 5.0 GiB |
| CPU user | 34.4s |
| CPU sys | 2.4s |
| Peak spill | 0 B |
**tpch — branch**
| Metric | Value |
|--------|-------|
| Wall time | 5.0s |
| Peak memory | 5.6 GiB |
| Avg memory | 5.0 GiB |
| CPU user | 34.9s |
| CPU sys | 2.4s |
| Peak spill | 0 B |
</details>
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangb (via GitHub)" <gi...@apache.org>.
adriangb commented on code in PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#discussion_r3170578419
##########
datafusion/physical-expr/src/expressions/case.rs:
##########
@@ -814,17 +814,15 @@ impl CaseBody {
}
}?;
- // `true_count` ignores `true` values where the validity bit is not set, so there's
- // no need to call `prep_null_mask_filter`.
- let when_true_count = when_value.true_count();
-
- // If the 'when' predicate did not match any rows, continue to the next branch immediately
- if when_true_count == 0 {
+ // If the 'when' predicate did not match any rows, continue to the next branch immediately.
+ // Like the old `true_count == 0` guard: `has_true()` only counts valid slots that are true
+ // (masked-null predicate slots are ignored), so no `prep_null_mask_filter` needed here.
Review Comment:
This seems like classic LLM mistake: I don't think we should have code comments referencing "the old implementation". IMO those should be comments on the PR to help the reviewer, but they are not helpful to future readers of the code. Code comments should stick to being targeted at future readers. So maybe something like:
```suggestion
// Only counts valid slots that are true (masked-null predicate slots are ignored),
// so no `prep_null_mask_filter` needed here.
```
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4348151666
🤖 Benchmark running (GKE) | [trigger](https://github.com/apache/datafusion/pull/21806#issuecomment-4348058676)
**Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB) | `Linux bench-c4348058676-1946-lfkkj 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux`
<details><summary>CPU Details (lscpu)</summary>
```
Architecture: aarch64
CPU op-mode(s): 64-bit
Byte Order: Little Endian
CPU(s): 16
On-line CPU(s) list: 0-15
Vendor ID: ARM
Model name: Neoverse-V2
Model: 1
Thread(s) per core: 1
Core(s) per cluster: 16
Socket(s): -
Cluster(s): 1
Stepping: r0p1
BogoMIPS: 2000.00
Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache: 1 MiB (16 instances)
L1i cache: 1 MiB (16 instances)
L2 cache: 32 MiB (16 instances)
L3 cache: 80 MiB (1 instance)
NUMA node(s): 1
NUMA node0 CPU(s): 0-15
Vulnerability Gather data sampling: Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Not affected
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed: Not affected
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1: Mitigation; __user pointer sanitization
Vulnerability Spectre v2: Mitigation; CSV2, BHB
Vulnerability Srbds: Not affected
Vulnerability Tsa: Not affected
Vulnerability Tsx async abort: Not affected
Vulnerability Vmscape: Not affected
```
</details>
Comparing feat/21784-has-true-has-false (8777c0987e264b56f007f310b4b494e6f0a39293) to 42cd2fa (merge-base) [diff](https://github.com/apache/datafusion/compare/42cd2fa90d18a8a54f433fc94917c20a4d1e7338..8777c0987e264b56f007f310b4b494e6f0a39293) using: tpch
Results will be posted here when complete
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4306122377
🤖 Benchmark completed (GKE) | [trigger](https://github.com/apache/datafusion/pull/21806#issuecomment-4305985687)
**Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB)
<details><summary>CPU Details (lscpu)</summary>
```
Architecture: aarch64
CPU op-mode(s): 64-bit
Byte Order: Little Endian
CPU(s): 16
On-line CPU(s) list: 0-15
Vendor ID: ARM
Model name: Neoverse-V2
Model: 1
Thread(s) per core: 1
Core(s) per cluster: 16
Socket(s): -
Cluster(s): 1
Stepping: r0p1
BogoMIPS: 2000.00
Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache: 1 MiB (16 instances)
L1i cache: 1 MiB (16 instances)
L2 cache: 32 MiB (16 instances)
L3 cache: 80 MiB (1 instance)
NUMA node(s): 1
NUMA node0 CPU(s): 0-15
Vulnerability Gather data sampling: Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Not affected
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed: Not affected
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1: Mitigation; __user pointer sanitization
Vulnerability Spectre v2: Mitigation; CSV2, BHB
Vulnerability Srbds: Not affected
Vulnerability Tsa: Not affected
Vulnerability Tsx async abort: Not affected
Vulnerability Vmscape: Not affected
```
</details>
<details><summary>Details</summary>
<p>
```
Comparing HEAD and feat_21784-has-true-has-false
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query ┃ HEAD ┃ feat_21784-has-true-has-false ┃ Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0 │ 1.17 / 4.68 ±6.90 / 18.47 ms │ 1.16 / 4.65 ±6.83 / 18.31 ms │ no change │
│ QQuery 1 │ 12.40 / 12.78 ±0.22 / 13.06 ms │ 12.60 / 13.03 ±0.24 / 13.32 ms │ no change │
│ QQuery 2 │ 37.06 / 37.45 ±0.25 / 37.77 ms │ 37.45 / 38.85 ±1.35 / 41.15 ms │ no change │
│ QQuery 3 │ 31.69 / 32.87 ±1.40 / 35.24 ms │ 31.35 / 32.48 ±1.66 / 35.74 ms │ no change │
│ QQuery 4 │ 248.68 / 250.66 ±2.10 / 254.54 ms │ 239.74 / 243.91 ±2.85 / 248.51 ms │ no change │
│ QQuery 5 │ 285.62 / 286.56 ±0.88 / 288.11 ms │ 283.27 / 285.82 ±2.02 / 289.11 ms │ no change │
│ QQuery 6 │ 6.78 / 8.15 ±1.87 / 11.84 ms │ 6.77 / 7.29 ±0.44 / 7.85 ms │ +1.12x faster │
│ QQuery 7 │ 13.88 / 14.07 ±0.10 / 14.18 ms │ 14.49 / 14.64 ±0.19 / 15.01 ms │ no change │
│ QQuery 8 │ 330.53 / 333.94 ±2.12 / 337.07 ms │ 331.43 / 332.29 ±0.66 / 333.39 ms │ no change │
│ QQuery 9 │ 440.38 / 455.84 ±10.60 / 472.09 ms │ 454.44 / 462.78 ±10.69 / 483.71 ms │ no change │
│ QQuery 10 │ 73.20 / 74.28 ±0.78 / 75.58 ms │ 73.15 / 74.55 ±1.42 / 77.19 ms │ no change │
│ QQuery 11 │ 83.55 / 84.73 ±0.70 / 85.49 ms │ 84.73 / 87.61 ±4.13 / 95.78 ms │ no change │
│ QQuery 12 │ 276.48 / 280.93 ±3.86 / 287.66 ms │ 276.37 / 279.94 ±3.14 / 285.21 ms │ no change │
│ QQuery 13 │ 395.89 / 403.36 ±10.09 / 423.20 ms │ 397.40 / 403.57 ±3.76 / 407.22 ms │ no change │
│ QQuery 14 │ 284.37 / 289.00 ±3.06 / 292.69 ms │ 289.08 / 294.83 ±3.20 / 297.86 ms │ no change │
│ QQuery 15 │ 288.65 / 290.19 ±1.52 / 292.93 ms │ 287.20 / 292.83 ±7.94 / 308.46 ms │ no change │
│ QQuery 16 │ 625.67 / 629.97 ±3.87 / 635.54 ms │ 628.69 / 635.48 ±4.04 / 639.04 ms │ no change │
│ QQuery 17 │ 634.51 / 637.93 ±2.29 / 640.79 ms │ 629.75 / 634.71 ±4.91 / 642.84 ms │ no change │
│ QQuery 18 │ 1271.87 / 1283.99 ±14.44 / 1309.76 ms │ 1267.37 / 1295.29 ±25.85 / 1343.88 ms │ no change │
│ QQuery 19 │ 28.68 / 28.96 ±0.21 / 29.20 ms │ 28.61 / 29.47 ±0.66 / 30.51 ms │ no change │
│ QQuery 20 │ 518.94 / 526.77 ±7.30 / 537.08 ms │ 516.53 / 522.42 ±3.18 / 525.83 ms │ no change │
│ QQuery 21 │ 597.19 / 606.36 ±10.97 / 627.55 ms │ 598.70 / 601.65 ±2.03 / 604.05 ms │ no change │
│ QQuery 22 │ 1057.94 / 1071.42 ±9.45 / 1087.50 ms │ 1064.79 / 1070.43 ±5.61 / 1080.21 ms │ no change │
│ QQuery 23 │ 3324.38 / 3355.32 ±18.56 / 3380.59 ms │ 3311.68 / 3345.99 ±25.37 / 3383.53 ms │ no change │
│ QQuery 24 │ 41.68 / 43.06 ±1.08 / 44.61 ms │ 41.47 / 43.49 ±1.77 / 46.53 ms │ no change │
│ QQuery 25 │ 113.54 / 116.20 ±2.87 / 120.59 ms │ 113.68 / 115.90 ±1.47 / 117.90 ms │ no change │
│ QQuery 26 │ 42.47 / 43.77 ±1.42 / 46.51 ms │ 42.49 / 44.25 ±1.55 / 47.15 ms │ no change │
│ QQuery 27 │ 671.37 / 677.83 ±5.17 / 687.10 ms │ 671.77 / 676.67 ±4.77 / 684.94 ms │ no change │
│ QQuery 28 │ 3017.20 / 3037.47 ±11.19 / 3051.51 ms │ 3015.63 / 3036.05 ±20.87 / 3072.43 ms │ no change │
│ QQuery 29 │ 42.74 / 49.40 ±8.99 / 65.93 ms │ 42.80 / 48.63 ±9.20 / 66.75 ms │ no change │
│ QQuery 30 │ 313.43 / 316.70 ±3.27 / 321.58 ms │ 312.35 / 318.99 ±6.82 / 329.19 ms │ no change │
│ QQuery 31 │ 310.41 / 317.18 ±5.84 / 327.78 ms │ 309.51 / 314.39 ±4.11 / 319.51 ms │ no change │
│ QQuery 32 │ 1019.35 / 1035.40 ±16.15 / 1065.60 ms │ 1019.33 / 1024.97 ±7.50 / 1039.66 ms │ no change │
│ QQuery 33 │ 1440.40 / 1457.11 ±9.07 / 1465.50 ms │ 1435.65 / 1461.02 ±15.89 / 1481.00 ms │ no change │
│ QQuery 34 │ 1458.14 / 1478.44 ±29.45 / 1536.76 ms │ 1463.76 / 1482.62 ±19.55 / 1519.68 ms │ no change │
│ QQuery 35 │ 296.70 / 321.76 ±28.89 / 377.32 ms │ 293.15 / 309.19 ±24.06 / 356.06 ms │ no change │
│ QQuery 36 │ 62.52 / 67.75 ±3.26 / 71.99 ms │ 63.23 / 68.19 ±4.25 / 75.76 ms │ no change │
│ QQuery 37 │ 35.76 / 38.67 ±4.59 / 47.68 ms │ 36.39 / 40.86 ±5.57 / 51.24 ms │ 1.06x slower │
│ QQuery 38 │ 40.80 / 44.97 ±3.37 / 49.52 ms │ 43.69 / 47.03 ±3.85 / 54.33 ms │ no change │
│ QQuery 39 │ 133.39 / 137.36 ±3.90 / 142.49 ms │ 124.34 / 133.36 ±5.30 / 140.19 ms │ no change │
│ QQuery 40 │ 14.49 / 15.00 ±0.34 / 15.54 ms │ 14.53 / 15.62 ±1.76 / 19.12 ms │ no change │
│ QQuery 41 │ 13.77 / 15.37 ±2.69 / 20.74 ms │ 13.91 / 14.20 ±0.24 / 14.61 ms │ +1.08x faster │
│ QQuery 42 │ 13.67 / 14.88 ±2.07 / 19.01 ms │ 13.55 / 13.73 ±0.18 / 14.05 ms │ +1.08x faster │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary ┃ ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD) │ 20228.58ms │
│ Total Time (feat_21784-has-true-has-false) │ 20213.69ms │
│ Average Time (HEAD) │ 470.43ms │
│ Average Time (feat_21784-has-true-has-false) │ 470.09ms │
│ Queries Faster │ 3 │
│ Queries Slower │ 1 │
│ Queries with No Change │ 39 │
│ Queries with Failure │ 0 │
└──────────────────────────────────────────────┴────────────┘
```
</p>
</details>
<details><summary>Resource Usage</summary>
**clickbench_partitioned — base (merge-base)**
| Metric | Value |
|--------|-------|
| Wall time | 105.0s |
| Peak memory | 31.1 GiB |
| Avg memory | 23.2 GiB |
| CPU user | 1068.8s |
| CPU sys | 64.1s |
| Peak spill | 0 B |
**clickbench_partitioned — branch**
| Metric | Value |
|--------|-------|
| Wall time | 105.0s |
| Peak memory | 29.1 GiB |
| Avg memory | 22.9 GiB |
| CPU user | 1068.4s |
| CPU sys | 65.9s |
| Peak spill | 0 B |
</details>
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangb (via GitHub)" <gi...@apache.org>.
adriangb merged PR #21806:
URL: https://github.com/apache/datafusion/pull/21806
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangb (via GitHub)" <gi...@apache.org>.
adriangb commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4352762538
run benchmark tpch
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "raushanprabhakar1 (via GitHub)" <gi...@apache.org>.
raushanprabhakar1 commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4324342599
Hi @adriangb , requesting you to review. Thanks
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangb (via GitHub)" <gi...@apache.org>.
adriangb commented on code in PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#discussion_r3164650235
##########
datafusion/physical-expr/src/expressions/case.rs:
##########
@@ -814,17 +814,13 @@ impl CaseBody {
}
}?;
- // `true_count` ignores `true` values where the validity bit is not set, so there's
- // no need to call `prep_null_mask_filter`.
Review Comment:
Worth keeping this comment?
##########
datafusion/functions-nested/src/replace.rs:
##########
@@ -347,7 +347,7 @@ fn general_replace<O: OffsetSizeTrait>(
let mut counter = 0;
// All elements are false, no need to replace, just copy original data
- if eq_array.false_count() == eq_array.len() {
+ if !eq_array.has_true() {
Review Comment:
We used to not take this path if there were any nulls but now we do. The implementation seems to allow this so it's okay, but I wanted to flag it for you to confirm.
##########
datafusion/physical-expr/src/expressions/case.rs:
##########
@@ -903,17 +899,13 @@ impl CaseBody {
internal_datafusion_err!("WHEN expression did not return a BooleanArray")
})?;
- // `true_count` ignores `true` values where the validity bit is not set, so there's
- // no need to call `prep_null_mask_filter`.
Review Comment:
Same here
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "Omega359 (via GitHub)" <gi...@apache.org>.
Omega359 commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4348479700
> Benchmark for [this request](https://github.com/apache/datafusion/pull/21806#issuecomment-4348058676) failed.
>
> Last 20 lines of output:
> Click to expand
It looks like it can't load the lineitem table
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangb (via GitHub)" <gi...@apache.org>.
adriangb commented on code in PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#discussion_r3164615682
##########
datafusion/physical-plan/src/topk/mod.rs:
##########
@@ -254,11 +254,11 @@ impl TopK {
let num_rows = batch.num_rows();
let array = filtered.into_array(num_rows)?;
let mut filter = array.as_boolean().clone();
- let true_count = filter.true_count();
- if true_count == 0 {
+ if !filter.has_true() {
// nothing to filter, so no need to update
return Ok(());
}
+ let true_count = filter.true_count();
// only update the keys / rows if the filter does not match all rows
if true_count < num_rows {
Review Comment:
Maybe:
```suggestion
// only update the keys / rows if the filter does not match all rows
if filter.null_count() > 0 || filter.has_false() {
```
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangb (via GitHub)" <gi...@apache.org>.
adriangb commented on code in PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#discussion_r3164615682
##########
datafusion/physical-plan/src/topk/mod.rs:
##########
@@ -254,11 +254,11 @@ impl TopK {
let num_rows = batch.num_rows();
let array = filtered.into_array(num_rows)?;
let mut filter = array.as_boolean().clone();
- let true_count = filter.true_count();
- if true_count == 0 {
+ if !filter.has_true() {
// nothing to filter, so no need to update
return Ok(());
}
+ let true_count = filter.true_count();
// only update the keys / rows if the filter does not match all rows
if true_count < num_rows {
Review Comment:
Maybe:
```suggestion
// only update the keys / rows if the filter does not match all rows
if filter.has_false() {
```
##########
datafusion/spark/src/function/array/array_contains.rs:
##########
@@ -84,7 +84,7 @@ fn apply_spark_null_semantics(
haystack_arg: &ColumnarValue,
) -> Result<BooleanArray> {
// happy path
- if result.false_count() == 0 || haystack_arg.data_type() == DataType::Null {
+ if !result.has_false() || haystack_arg.data_type() == DataType::Null {
Review Comment:
Consider flipping this conditional? The second check should be cheaper.
```suggestion
if haystack_arg.data_type() == DataType::Null || !result.has_false() {
```
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangb (via GitHub)" <gi...@apache.org>.
adriangb commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4356670584
Thanks @raushanprabhakar1 !
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4306007615
🤖 Benchmark running (GKE) | [trigger](https://github.com/apache/datafusion/pull/21806#issuecomment-4305985687)
**Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB) | `Linux bench-c4305985687-1781-wplz4 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux`
<details><summary>CPU Details (lscpu)</summary>
```
Architecture: aarch64
CPU op-mode(s): 64-bit
Byte Order: Little Endian
CPU(s): 16
On-line CPU(s) list: 0-15
Vendor ID: ARM
Model name: Neoverse-V2
Model: 1
Thread(s) per core: 1
Core(s) per cluster: 16
Socket(s): -
Cluster(s): 1
Stepping: r0p1
BogoMIPS: 2000.00
Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache: 1 MiB (16 instances)
L1i cache: 1 MiB (16 instances)
L2 cache: 32 MiB (16 instances)
L3 cache: 80 MiB (1 instance)
NUMA node(s): 1
NUMA node0 CPU(s): 0-15
Vulnerability Gather data sampling: Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Not affected
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed: Not affected
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1: Mitigation; __user pointer sanitization
Vulnerability Spectre v2: Mitigation; CSV2, BHB
Vulnerability Srbds: Not affected
Vulnerability Tsa: Not affected
Vulnerability Tsx async abort: Not affected
Vulnerability Vmscape: Not affected
```
</details>
Comparing feat/21784-has-true-has-false (ca7e0a82865af30041a82609b5151266fc1f75d2) to 067ba4b (merge-base) [diff](https://github.com/apache/datafusion/compare/067ba4b02e180ccd3f7e43f087698463d53a566e..ca7e0a82865af30041a82609b5151266fc1f75d2) using: tpch
Results will be posted here when complete
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4306004149
Benchmark for [this request](https://github.com/apache/datafusion/pull/21806#issuecomment-4305985687) failed.
Last 20 lines of output:
<details><summary>Click to expand</summary>
```
Cloning into '/workspace/datafusion-branch'...
error: RPC failed; HTTP 500 curl 22 The requested URL returned error: 500
fatal: expected 'acknowledgments'
```
</details>
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangb (via GitHub)" <gi...@apache.org>.
adriangb commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4353048433
Benches look good. @raushanprabhakar1 if you can address my feedback we can move to merge this
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangb (via GitHub)" <gi...@apache.org>.
adriangb commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4355429631
> > > Benchmark for [this request](https://github.com/apache/datafusion/pull/21806#issuecomment-4348058676) failed.
> > > Last 20 lines of output:
> > > Click to expand
> >
> >
> > It looks like it can't load the lineitem table
>
> maybe we should add a sanity check to the benchmark runner / better error message (so it fails with a message that says the files aren't there rather than failing on the expected output check) 🤔
Yes, I’ve run into this issue before. Even with just DataFusion CLI if you pointed to an empty directory, it will give you a table with no columns instead of some sort of error that says there’s no data here. I can’t think of any use case where you’d want to infer a schema with no columns instead of en error saying there’s no data.
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4348148973
🤖 Benchmark running (GKE) | [trigger](https://github.com/apache/datafusion/pull/21806#issuecomment-4348058676)
**Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB) | `Linux bench-c4348058676-1944-6cnsd 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux`
<details><summary>CPU Details (lscpu)</summary>
```
Architecture: aarch64
CPU op-mode(s): 64-bit
Byte Order: Little Endian
CPU(s): 16
On-line CPU(s) list: 0-15
Vendor ID: ARM
Model name: Neoverse-V2
Model: 1
Thread(s) per core: 1
Core(s) per cluster: 16
Socket(s): -
Cluster(s): 1
Stepping: r0p1
BogoMIPS: 2000.00
Flags: fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache: 1 MiB (16 instances)
L1i cache: 1 MiB (16 instances)
L2 cache: 32 MiB (16 instances)
L3 cache: 80 MiB (1 instance)
NUMA node(s): 1
NUMA node0 CPU(s): 0-15
Vulnerability Gather data sampling: Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Not affected
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed: Not affected
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1: Mitigation; __user pointer sanitization
Vulnerability Spectre v2: Mitigation; CSV2, BHB
Vulnerability Srbds: Not affected
Vulnerability Tsa: Not affected
Vulnerability Tsx async abort: Not affected
Vulnerability Vmscape: Not affected
```
</details>
Comparing feat/21784-has-true-has-false (8777c0987e264b56f007f310b4b494e6f0a39293) to 42cd2fa (merge-base) [diff](https://github.com/apache/datafusion/compare/42cd2fa90d18a8a54f433fc94917c20a4d1e7338..8777c0987e264b56f007f310b4b494e6f0a39293) using: clickbench_partitioned
Results will be posted here when complete
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Adding Use of arrow's has_true() / has_false() [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #21806:
URL: https://github.com/apache/datafusion/pull/21806#issuecomment-4348240399
Benchmark for [this request](https://github.com/apache/datafusion/pull/21806#issuecomment-4348058676) failed.
Last 20 lines of output:
<details><summary>Click to expand</summary>
```
Downloading crates ...
Downloaded anyhow v1.0.102
Downloaded prost-derive v0.14.3
Downloaded prost v0.14.3
Compiling anyhow v1.0.102
Compiling either v1.15.0
Compiling datafusion-benchmarks v53.1.0 (/workspace/datafusion-base/benchmarks)
Compiling itertools v0.14.0
Compiling prost-derive v0.14.3
Compiling prost v0.14.3
Compiling datafusion-proto-common v53.1.0 (/workspace/datafusion-base/datafusion/proto-common)
Compiling datafusion-proto v53.1.0 (/workspace/datafusion-base/datafusion/proto)
Finished `bench` profile [optimized + debuginfo] target(s) in 8m 51s
Running benches/sql.rs (/workspace/datafusion-base/target/release/deps/sql-2c9f51cbfcaf27f8)
Gnuplot not found, using plotters backend
thread 'main' (21293) panicked at benchmarks/benches/sql.rs:131:46:
assertion failed: Execution("Error in result on row 1, column 1 running query \"\nSELECT COUNT(*) > 0 from lineitem;\": expected value \"true\" but got value \"false\" in row: [\"false\"]")
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
error: bench failed, to rerun pass `--bench sql`
```
</details>
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
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: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org