Skip to content

feat: Support IEEE 754 negative zero semantics#22835

Open
comphead wants to merge 3 commits into
apache:mainfrom
comphead:array_nan
Open

feat: Support IEEE 754 negative zero semantics#22835
comphead wants to merge 3 commits into
apache:mainfrom
comphead:array_nan

Conversation

@comphead

@comphead comphead commented Jun 9, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

Which issue does this PR close?

Rationale for this change

What changes are included in this PR?

Are these changes tested?

Are there any user-facing changes?

@github-actions github-actions Bot added physical-expr Changes to the physical-expr crates sqllogictest SQL Logic Tests (.slt) common Related to common crate functions Changes to functions implementation physical-plan Changes to the physical-plan crate labels Jun 9, 2026
@comphead

comphead commented Jun 9, 2026

Copy link
Copy Markdown
Contributor Author

run benchmark tpch tpcds

@adriangbot

adriangbot commented Jun 9, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4655326814-495-cz9fl 6.12.68+ #1 SMP Sat May 2 07:49:07 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
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

Comparing array_nan (1a564ec) to 883c38e (merge-base) diff using: tpch
Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Jun 9, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4655326814-496-kmqln 6.12.68+ #1 SMP Sat May 2 07:49:07 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
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

Comparing array_nan (1a564ec) to 883c38e (merge-base) diff using: tpcds
Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Jun 9, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu) 
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 HEAD and array_nan
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃                      array_nan ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │ 38.72 / 41.03 ±2.85 / 46.43 ms │ 38.38 / 39.63 ±1.14 / 41.10 ms │     no change │
│ QQuery 2  │ 18.55 / 19.35 ±0.63 / 20.46 ms │ 18.58 / 19.20 ±0.73 / 20.60 ms │     no change │
│ QQuery 3  │ 31.16 / 34.84 ±3.23 / 39.72 ms │ 30.85 / 32.81 ±2.28 / 36.70 ms │ +1.06x faster │
│ QQuery 4  │ 17.52 / 17.93 ±0.58 / 19.08 ms │ 17.09 / 17.26 ±0.16 / 17.50 ms │     no change │
│ QQuery 5  │ 38.30 / 40.71 ±1.25 / 41.91 ms │ 39.53 / 40.19 ±0.44 / 40.78 ms │     no change │
│ QQuery 6  │ 16.47 / 16.56 ±0.10 / 16.71 ms │ 16.35 / 16.98 ±0.39 / 17.43 ms │     no change │
│ QQuery 7  │ 44.59 / 46.73 ±2.24 / 49.89 ms │ 43.45 / 45.16 ±1.39 / 46.89 ms │     no change │
│ QQuery 8  │ 43.35 / 43.72 ±0.32 / 44.23 ms │ 42.92 / 43.20 ±0.19 / 43.40 ms │     no change │
│ QQuery 9  │ 49.30 / 50.35 ±0.87 / 51.84 ms │ 49.12 / 50.19 ±0.91 / 51.75 ms │     no change │
│ QQuery 10 │ 42.29 / 42.74 ±0.31 / 43.15 ms │ 42.07 / 42.31 ±0.15 / 42.51 ms │     no change │
│ QQuery 11 │ 13.09 / 13.66 ±0.62 / 14.47 ms │ 13.24 / 13.51 ±0.32 / 14.02 ms │     no change │
│ QQuery 12 │ 23.56 / 24.29 ±0.39 / 24.70 ms │ 24.39 / 25.15 ±0.83 / 26.75 ms │     no change │
│ QQuery 13 │ 31.80 / 34.65 ±2.42 / 38.97 ms │ 31.94 / 33.95 ±1.63 / 36.88 ms │     no change │
│ QQuery 14 │ 23.80 / 23.92 ±0.10 / 24.08 ms │ 23.90 / 23.97 ±0.10 / 24.18 ms │     no change │
│ QQuery 15 │ 31.36 / 32.36 ±1.13 / 34.46 ms │ 31.46 / 31.65 ±0.19 / 31.96 ms │     no change │
│ QQuery 16 │ 13.90 / 14.13 ±0.16 / 14.27 ms │ 13.76 / 14.07 ±0.20 / 14.35 ms │     no change │
│ QQuery 17 │ 73.23 / 73.97 ±0.57 / 74.94 ms │ 73.18 / 74.28 ±0.99 / 75.87 ms │     no change │
│ QQuery 18 │ 57.16 / 59.53 ±1.50 / 61.70 ms │ 58.07 / 60.81 ±3.16 / 66.92 ms │     no change │
│ QQuery 19 │ 32.91 / 33.21 ±0.40 / 34.00 ms │ 32.88 / 33.21 ±0.34 / 33.76 ms │     no change │
│ QQuery 20 │ 32.14 / 32.83 ±0.46 / 33.58 ms │ 32.00 / 32.89 ±0.95 / 34.71 ms │     no change │
│ QQuery 21 │ 56.95 / 58.37 ±1.10 / 60.14 ms │ 55.91 / 58.41 ±2.28 / 61.88 ms │     no change │
│ QQuery 22 │ 14.07 / 14.20 ±0.10 / 14.37 ms │ 14.03 / 14.34 ±0.36 / 15.04 ms │     no change │
└───────────┴────────────────────────────────┴────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary        ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)        │ 769.08ms │
│ Total Time (array_nan)   │ 763.21ms │
│ Average Time (HEAD)      │  34.96ms │
│ Average Time (array_nan) │  34.69ms │
│ Queries Faster           │        1 │
│ Queries Slower           │        0 │
│ Queries with No Change   │       21 │
│ Queries with Failure     │        0 │
└──────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric Value
Wall time 5.0s
Peak memory 4.6 GiB
Avg memory 3.8 GiB
CPU user 30.1s
CPU sys 2.4s
Peak spill 0 B

tpch — branch

Metric Value
Wall time 5.0s
Peak memory 4.6 GiB
Avg memory 3.8 GiB
CPU user 30.1s
CPU sys 2.3s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

adriangbot commented Jun 9, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu) 
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 HEAD and array_nan
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃                             array_nan ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.84 / 6.32 ±0.81 / 7.94 ms │           5.75 / 6.24 ±0.89 / 8.02 ms │     no change │
│ QQuery 2  │        79.90 / 80.25 ±0.25 / 80.66 ms │        80.04 / 80.32 ±0.25 / 80.77 ms │     no change │
│ QQuery 3  │        28.89 / 28.97 ±0.11 / 29.19 ms │        28.50 / 28.67 ±0.14 / 28.89 ms │     no change │
│ QQuery 4  │     488.11 / 495.30 ±4.79 / 500.60 ms │    471.40 / 479.53 ±10.33 / 499.75 ms │     no change │
│ QQuery 5  │        51.40 / 51.84 ±0.25 / 52.09 ms │        51.22 / 52.12 ±0.59 / 53.03 ms │     no change │
│ QQuery 6  │        36.88 / 37.10 ±0.27 / 37.53 ms │        36.03 / 36.68 ±0.51 / 37.61 ms │     no change │
│ QQuery 7  │        94.63 / 95.43 ±0.52 / 96.23 ms │        95.55 / 96.16 ±0.44 / 96.82 ms │     no change │
│ QQuery 8  │        37.16 / 38.32 ±1.98 / 42.27 ms │        36.94 / 39.24 ±3.37 / 45.87 ms │     no change │
│ QQuery 9  │        53.62 / 55.87 ±2.30 / 60.14 ms │        52.03 / 53.47 ±1.12 / 55.36 ms │     no change │
│ QQuery 10 │        62.38 / 62.90 ±0.32 / 63.37 ms │        62.88 / 63.17 ±0.35 / 63.84 ms │     no change │
│ QQuery 11 │     297.92 / 301.68 ±3.81 / 308.17 ms │     288.80 / 296.23 ±7.71 / 307.70 ms │     no change │
│ QQuery 12 │        28.85 / 29.12 ±0.20 / 29.44 ms │        28.67 / 29.13 ±0.60 / 30.29 ms │     no change │
│ QQuery 13 │     118.95 / 119.47 ±0.44 / 120.16 ms │     119.00 / 120.23 ±0.86 / 121.55 ms │     no change │
│ QQuery 14 │     407.55 / 413.73 ±5.50 / 421.95 ms │     408.08 / 409.99 ±1.61 / 411.94 ms │     no change │
│ QQuery 15 │        58.66 / 59.07 ±0.34 / 59.59 ms │        57.88 / 58.76 ±0.56 / 59.64 ms │     no change │
│ QQuery 16 │           6.75 / 6.94 ±0.20 / 7.32 ms │           6.75 / 6.86 ±0.11 / 7.06 ms │     no change │
│ QQuery 17 │        80.98 / 81.41 ±0.32 / 81.90 ms │        80.70 / 81.72 ±1.26 / 84.17 ms │     no change │
│ QQuery 18 │     124.64 / 126.34 ±1.38 / 128.49 ms │     124.49 / 126.46 ±1.53 / 128.93 ms │     no change │
│ QQuery 19 │        41.68 / 41.84 ±0.10 / 41.99 ms │        41.36 / 41.51 ±0.20 / 41.89 ms │     no change │
│ QQuery 20 │        35.26 / 35.53 ±0.32 / 36.14 ms │        35.17 / 35.51 ±0.37 / 36.18 ms │     no change │
│ QQuery 21 │        16.85 / 17.13 ±0.26 / 17.52 ms │        16.72 / 17.09 ±0.29 / 17.54 ms │     no change │
│ QQuery 22 │        61.96 / 63.70 ±2.54 / 68.75 ms │        62.09 / 63.59 ±2.19 / 67.82 ms │     no change │
│ QQuery 23 │     319.24 / 321.29 ±1.56 / 324.06 ms │     314.88 / 317.74 ±1.74 / 319.82 ms │     no change │
│ QQuery 24 │     221.77 / 224.34 ±2.00 / 227.72 ms │     222.61 / 226.14 ±3.66 / 233.06 ms │     no change │
│ QQuery 25 │     110.10 / 111.51 ±1.38 / 114.09 ms │     111.02 / 114.34 ±4.83 / 123.96 ms │     no change │
│ QQuery 26 │        57.61 / 58.15 ±0.37 / 58.67 ms │        57.31 / 57.84 ±0.38 / 58.48 ms │     no change │
│ QQuery 27 │          6.60 / 7.63 ±1.76 / 11.13 ms │           6.60 / 6.76 ±0.13 / 6.98 ms │ +1.13x faster │
│ QQuery 28 │        56.31 / 61.16 ±3.70 / 67.39 ms │        56.43 / 59.76 ±2.33 / 61.92 ms │     no change │
│ QQuery 29 │        97.85 / 98.54 ±0.70 / 99.80 ms │       97.09 / 99.59 ±2.99 / 105.19 ms │     no change │
│ QQuery 30 │        32.22 / 32.84 ±0.45 / 33.34 ms │        32.10 / 32.47 ±0.25 / 32.76 ms │     no change │
│ QQuery 31 │     111.56 / 113.78 ±2.72 / 119.01 ms │     111.06 / 112.05 ±0.72 / 113.05 ms │     no change │
│ QQuery 32 │        20.47 / 20.68 ±0.15 / 20.80 ms │        19.84 / 20.24 ±0.39 / 20.82 ms │     no change │
│ QQuery 33 │        38.38 / 38.80 ±0.22 / 39.02 ms │        38.50 / 38.89 ±0.38 / 39.61 ms │     no change │
│ QQuery 34 │           9.33 / 9.59 ±0.20 / 9.91 ms │          9.34 / 9.72 ±0.29 / 10.11 ms │     no change │
│ QQuery 35 │        70.88 / 72.01 ±0.93 / 73.26 ms │        70.85 / 71.31 ±0.27 / 71.64 ms │     no change │
│ QQuery 36 │           5.99 / 6.13 ±0.15 / 6.42 ms │           5.83 / 5.94 ±0.15 / 6.24 ms │     no change │
│ QQuery 37 │           6.82 / 7.18 ±0.20 / 7.42 ms │           6.62 / 6.75 ±0.09 / 6.84 ms │ +1.06x faster │
│ QQuery 38 │        63.62 / 63.99 ±0.30 / 64.38 ms │        62.37 / 63.49 ±1.02 / 65.40 ms │     no change │
│ QQuery 39 │     448.24 / 457.21 ±7.28 / 468.75 ms │     448.60 / 452.43 ±1.98 / 453.99 ms │     no change │
│ QQuery 40 │        22.62 / 23.80 ±1.80 / 27.38 ms │        22.95 / 23.03 ±0.09 / 23.21 ms │     no change │
│ QQuery 41 │        11.09 / 11.32 ±0.28 / 11.88 ms │        11.01 / 11.19 ±0.21 / 11.60 ms │     no change │
│ QQuery 42 │        23.50 / 23.91 ±0.23 / 24.15 ms │        23.28 / 24.27 ±1.58 / 27.40 ms │     no change │
│ QQuery 43 │          4.84 / 6.29 ±2.75 / 11.79 ms │           4.71 / 4.80 ±0.11 / 5.02 ms │ +1.31x faster │
│ QQuery 44 │           8.94 / 9.03 ±0.09 / 9.19 ms │           8.63 / 8.69 ±0.06 / 8.81 ms │     no change │
│ QQuery 45 │        38.01 / 39.34 ±1.04 / 40.65 ms │        37.41 / 39.67 ±2.60 / 44.67 ms │     no change │
│ QQuery 46 │        11.50 / 11.94 ±0.24 / 12.15 ms │        11.85 / 12.16 ±0.40 / 12.94 ms │     no change │
│ QQuery 47 │     224.77 / 229.22 ±2.79 / 233.36 ms │     228.77 / 230.40 ±1.03 / 231.80 ms │     no change │
│ QQuery 48 │        96.35 / 96.68 ±0.24 / 97.10 ms │        96.64 / 97.16 ±0.60 / 98.31 ms │     no change │
│ QQuery 49 │        76.97 / 77.99 ±0.68 / 78.66 ms │        77.53 / 80.73 ±3.80 / 87.85 ms │     no change │
│ QQuery 50 │        59.69 / 60.04 ±0.37 / 60.73 ms │        59.60 / 60.05 ±0.26 / 60.29 ms │     no change │
│ QQuery 51 │       90.97 / 95.53 ±4.44 / 103.55 ms │        91.69 / 93.75 ±2.72 / 98.90 ms │     no change │
│ QQuery 52 │        23.81 / 24.32 ±0.48 / 25.23 ms │        24.12 / 26.18 ±2.73 / 31.47 ms │  1.08x slower │
│ QQuery 53 │        29.58 / 29.88 ±0.30 / 30.44 ms │        29.01 / 29.62 ±0.53 / 30.23 ms │     no change │
│ QQuery 54 │        54.76 / 55.32 ±0.49 / 56.00 ms │        54.48 / 54.96 ±0.36 / 55.54 ms │     no change │
│ QQuery 55 │        23.17 / 23.39 ±0.20 / 23.64 ms │        23.01 / 23.40 ±0.29 / 23.73 ms │     no change │
│ QQuery 56 │        38.31 / 38.58 ±0.34 / 39.24 ms │        38.46 / 38.81 ±0.47 / 39.71 ms │     no change │
│ QQuery 57 │     174.19 / 177.33 ±3.71 / 184.55 ms │     174.09 / 177.25 ±2.92 / 182.69 ms │     no change │
│ QQuery 58 │     114.75 / 116.40 ±2.13 / 120.60 ms │     113.69 / 114.14 ±0.28 / 114.45 ms │     no change │
│ QQuery 59 │     116.08 / 117.48 ±1.48 / 120.31 ms │     116.31 / 117.31 ±0.80 / 118.42 ms │     no change │
│ QQuery 60 │        39.00 / 40.27 ±0.93 / 41.27 ms │        38.37 / 38.93 ±0.34 / 39.28 ms │     no change │
│ QQuery 61 │        12.94 / 13.07 ±0.21 / 13.49 ms │        13.22 / 13.36 ±0.21 / 13.77 ms │     no change │
│ QQuery 62 │        45.53 / 46.00 ±0.35 / 46.47 ms │        45.93 / 46.14 ±0.19 / 46.47 ms │     no change │
│ QQuery 63 │        29.54 / 29.84 ±0.29 / 30.37 ms │        29.02 / 29.54 ±0.41 / 30.07 ms │     no change │
│ QQuery 64 │     393.38 / 399.14 ±5.83 / 409.98 ms │     393.43 / 397.56 ±4.92 / 407.13 ms │     no change │
│ QQuery 65 │     147.56 / 151.22 ±3.11 / 155.55 ms │     144.85 / 150.67 ±4.97 / 159.38 ms │     no change │
│ QQuery 66 │        78.69 / 79.56 ±0.65 / 80.30 ms │        78.04 / 79.24 ±0.76 / 80.13 ms │     no change │
│ QQuery 67 │     246.97 / 252.99 ±3.57 / 256.67 ms │     241.84 / 245.79 ±4.07 / 253.05 ms │     no change │
│ QQuery 68 │        11.68 / 11.98 ±0.20 / 12.29 ms │        11.66 / 12.01 ±0.37 / 12.61 ms │     no change │
│ QQuery 69 │        56.94 / 57.43 ±0.38 / 58.02 ms │        57.04 / 59.26 ±3.58 / 66.40 ms │     no change │
│ QQuery 70 │     103.84 / 104.96 ±0.64 / 105.70 ms │     103.61 / 105.22 ±1.85 / 108.84 ms │     no change │
│ QQuery 71 │        36.08 / 38.92 ±4.08 / 46.96 ms │        35.25 / 36.52 ±1.74 / 39.93 ms │ +1.07x faster │
│ QQuery 72 │ 2098.49 / 2237.43 ±95.66 / 2396.76 ms │ 2069.10 / 2146.60 ±41.12 / 2186.24 ms │     no change │
│ QQuery 73 │           9.16 / 9.59 ±0.22 / 9.76 ms │           9.19 / 9.44 ±0.19 / 9.69 ms │     no change │
│ QQuery 74 │     169.59 / 174.92 ±5.75 / 185.96 ms │     168.32 / 175.00 ±6.54 / 184.02 ms │     no change │
│ QQuery 75 │     147.85 / 148.89 ±1.14 / 150.34 ms │     147.00 / 148.85 ±1.09 / 150.04 ms │     no change │
│ QQuery 76 │        35.42 / 40.16 ±7.61 / 55.31 ms │        34.91 / 35.24 ±0.26 / 35.60 ms │ +1.14x faster │
│ QQuery 77 │        61.20 / 61.70 ±0.44 / 62.43 ms │        60.16 / 61.11 ±0.51 / 61.58 ms │     no change │
│ QQuery 78 │     185.12 / 188.59 ±1.86 / 190.53 ms │     184.42 / 189.65 ±4.63 / 196.55 ms │     no change │
│ QQuery 79 │        66.86 / 67.46 ±0.51 / 68.18 ms │        66.54 / 69.14 ±3.61 / 76.08 ms │     no change │
│ QQuery 80 │     101.22 / 105.58 ±6.20 / 117.69 ms │     100.48 / 101.55 ±0.92 / 102.65 ms │     no change │
│ QQuery 81 │        25.29 / 25.55 ±0.21 / 25.80 ms │        25.10 / 25.55 ±0.30 / 26.04 ms │     no change │
│ QQuery 82 │        16.01 / 16.25 ±0.21 / 16.62 ms │        15.91 / 16.21 ±0.26 / 16.68 ms │     no change │
│ QQuery 83 │        38.70 / 39.08 ±0.34 / 39.60 ms │        38.94 / 41.26 ±3.86 / 48.97 ms │  1.06x slower │
│ QQuery 84 │        34.13 / 34.64 ±0.26 / 34.85 ms │        34.62 / 37.23 ±3.55 / 44.25 ms │  1.07x slower │
│ QQuery 85 │     106.23 / 111.53 ±4.61 / 117.75 ms │     108.05 / 108.52 ±0.38 / 109.07 ms │     no change │
│ QQuery 86 │        24.78 / 25.05 ±0.27 / 25.51 ms │        24.70 / 26.68 ±3.58 / 33.82 ms │  1.07x slower │
│ QQuery 87 │        63.66 / 64.08 ±0.22 / 64.26 ms │        63.93 / 66.22 ±1.97 / 69.33 ms │     no change │
│ QQuery 88 │        61.56 / 64.87 ±6.07 / 77.00 ms │        62.02 / 62.51 ±0.68 / 63.85 ms │     no change │
│ QQuery 89 │        35.80 / 36.10 ±0.28 / 36.60 ms │        35.50 / 35.98 ±0.42 / 36.74 ms │     no change │
│ QQuery 90 │        16.74 / 16.87 ±0.10 / 17.03 ms │        16.62 / 16.96 ±0.26 / 17.31 ms │     no change │
│ QQuery 91 │        44.38 / 44.97 ±0.37 / 45.43 ms │        44.17 / 44.56 ±0.34 / 45.11 ms │     no change │
│ QQuery 92 │        28.89 / 29.37 ±0.41 / 30.07 ms │        29.02 / 31.19 ±3.04 / 37.20 ms │  1.06x slower │
│ QQuery 93 │        50.04 / 51.68 ±1.60 / 54.59 ms │        49.87 / 51.21 ±1.05 / 53.00 ms │     no change │
│ QQuery 94 │        38.21 / 40.12 ±2.69 / 45.42 ms │        37.98 / 38.67 ±0.67 / 39.80 ms │     no change │
│ QQuery 95 │        83.19 / 85.12 ±1.37 / 86.91 ms │        83.58 / 85.24 ±2.17 / 89.51 ms │     no change │
│ QQuery 96 │        23.86 / 24.08 ±0.24 / 24.49 ms │        23.95 / 25.09 ±1.81 / 28.67 ms │     no change │
│ QQuery 97 │        45.61 / 46.53 ±1.01 / 48.41 ms │        46.06 / 46.48 ±0.44 / 47.20 ms │     no change │
│ QQuery 98 │        42.98 / 44.11 ±1.00 / 45.64 ms │        42.83 / 43.47 ±0.43 / 44.15 ms │     no change │
│ QQuery 99 │        68.80 / 69.31 ±0.31 / 69.71 ms │        69.36 / 70.96 ±2.65 / 76.23 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary        ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)        │ 10349.86ms │
│ Total Time (array_nan)   │ 10210.53ms │
│ Average Time (HEAD)      │   104.54ms │
│ Average Time (array_nan) │   103.14ms │
│ Queries Faster           │          5 │
│ Queries Slower           │          5 │
│ Queries with No Change   │         89 │
│ Queries with Failure     │          0 │
└──────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric Value
Wall time 55.0s
Peak memory 5.8 GiB
Avg memory 5.1 GiB
CPU user 237.7s
CPU sys 6.4s
Peak spill 0 B

tpcds — branch

Metric Value
Wall time 55.0s
Peak memory 5.7 GiB
Avg memory 5.0 GiB
CPU user 234.9s
CPU sys 6.3s
Peak spill 0 B

File an issue against this benchmark runner

@comphead comphead changed the title feat: Support IEEE 754 for SQL ops feat: Support IEEE 754 negative zero semantics Jun 9, 2026
@comphead

comphead commented Jun 9, 2026

Copy link
Copy Markdown
Contributor Author

run benchmark tpch tpcds

@adriangbot

adriangbot commented Jun 9, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4655762895-497-bbtkh 6.12.68+ #1 SMP Sat May 2 07:49:07 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
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

Comparing array_nan (beb8750) to 883c38e (merge-base) diff using: tpch
Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Jun 9, 2026

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4655762895-498-4xr8g 6.12.68+ #1 SMP Sat May 2 07:49:07 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu) 
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

Comparing array_nan (beb8750) to 883c38e (merge-base) diff using: tpcds
Results will be posted here when complete

File an issue against this benchmark runner

@adriangbot

adriangbot commented Jun 9, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu) 
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 HEAD and array_nan
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃                      array_nan ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │ 38.43 / 39.66 ±1.01 / 41.21 ms │ 38.32 / 39.16 ±1.08 / 41.20 ms │ no change │
│ QQuery 2  │ 18.48 / 18.78 ±0.16 / 18.94 ms │ 18.34 / 18.72 ±0.29 / 19.15 ms │ no change │
│ QQuery 3  │ 31.38 / 32.89 ±1.21 / 34.58 ms │ 30.72 / 34.40 ±2.29 / 37.02 ms │ no change │
│ QQuery 4  │ 16.96 / 17.34 ±0.45 / 18.21 ms │ 16.82 / 17.25 ±0.33 / 17.84 ms │ no change │
│ QQuery 5  │ 37.37 / 38.84 ±1.50 / 40.86 ms │ 37.77 / 39.92 ±1.19 / 40.99 ms │ no change │
│ QQuery 6  │ 16.21 / 16.42 ±0.15 / 16.61 ms │ 16.14 / 16.27 ±0.08 / 16.39 ms │ no change │
│ QQuery 7  │ 42.90 / 45.18 ±2.95 / 50.90 ms │ 44.95 / 46.12 ±1.16 / 47.94 ms │ no change │
│ QQuery 8  │ 42.79 / 43.18 ±0.28 / 43.57 ms │ 42.47 / 43.10 ±0.44 / 43.78 ms │ no change │
│ QQuery 9  │ 49.21 / 49.90 ±0.69 / 51.07 ms │ 48.42 / 49.79 ±0.90 / 50.80 ms │ no change │
│ QQuery 10 │ 42.15 / 43.00 ±1.06 / 45.07 ms │ 41.66 / 42.26 ±0.82 / 43.88 ms │ no change │
│ QQuery 11 │ 12.76 / 13.10 ±0.21 / 13.38 ms │ 12.95 / 13.03 ±0.10 / 13.24 ms │ no change │
│ QQuery 12 │ 24.17 / 24.48 ±0.21 / 24.77 ms │ 23.62 / 23.96 ±0.28 / 24.32 ms │ no change │
│ QQuery 13 │ 32.16 / 34.18 ±1.82 / 37.28 ms │ 32.98 / 34.63 ±2.86 / 40.32 ms │ no change │
│ QQuery 14 │ 23.97 / 24.07 ±0.09 / 24.19 ms │ 23.58 / 23.95 ±0.30 / 24.46 ms │ no change │
│ QQuery 15 │ 31.52 / 32.32 ±0.80 / 33.43 ms │ 31.16 / 31.72 ±0.74 / 33.15 ms │ no change │
│ QQuery 16 │ 13.99 / 14.08 ±0.10 / 14.28 ms │ 13.18 / 13.84 ±0.55 / 14.86 ms │ no change │
│ QQuery 17 │ 73.38 / 74.38 ±0.87 / 75.86 ms │ 72.74 / 73.80 ±0.75 / 74.62 ms │ no change │
│ QQuery 18 │ 57.65 / 59.00 ±0.85 / 60.06 ms │ 57.66 / 58.46 ±0.75 / 59.82 ms │ no change │
│ QQuery 19 │ 32.89 / 33.33 ±0.38 / 33.87 ms │ 32.66 / 32.99 ±0.35 / 33.64 ms │ no change │
│ QQuery 20 │ 31.81 / 32.28 ±0.26 / 32.50 ms │ 31.49 / 32.31 ±0.44 / 32.78 ms │ no change │
│ QQuery 21 │ 55.44 / 57.67 ±1.60 / 59.89 ms │ 55.43 / 57.74 ±1.62 / 59.73 ms │ no change │
│ QQuery 22 │ 13.63 / 13.96 ±0.23 / 14.24 ms │ 13.55 / 14.03 ±0.72 / 15.46 ms │ no change │
└───────────┴────────────────────────────────┴────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary        ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)        │ 758.05ms │
│ Total Time (array_nan)   │ 757.45ms │
│ Average Time (HEAD)      │  34.46ms │
│ Average Time (array_nan) │  34.43ms │
│ Queries Faster           │        0 │
│ Queries Slower           │        0 │
│ Queries with No Change   │       22 │
│ Queries with Failure     │        0 │
└──────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric Value
Wall time 5.0s
Peak memory 4.6 GiB
Avg memory 3.8 GiB
CPU user 30.0s
CPU sys 2.4s
Peak spill 0 B

tpch — branch

Metric Value
Wall time 5.0s
Peak memory 4.6 GiB
Avg memory 3.8 GiB
CPU user 29.9s
CPU sys 2.3s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

adriangbot commented Jun 9, 2026

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu) 
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 HEAD and array_nan
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃                             array_nan ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.70 / 6.24 ±0.84 / 7.91 ms │           5.83 / 6.35 ±0.89 / 8.13 ms │     no change │
│ QQuery 2  │        79.86 / 80.09 ±0.16 / 80.31 ms │        80.73 / 81.21 ±0.34 / 81.78 ms │     no change │
│ QQuery 3  │        28.68 / 28.84 ±0.15 / 29.11 ms │        29.06 / 29.20 ±0.12 / 29.41 ms │     no change │
│ QQuery 4  │     471.72 / 479.12 ±9.36 / 497.27 ms │     476.94 / 479.72 ±3.38 / 486.18 ms │     no change │
│ QQuery 5  │        51.11 / 52.58 ±1.00 / 54.21 ms │        51.34 / 51.80 ±0.56 / 52.87 ms │     no change │
│ QQuery 6  │        36.04 / 36.28 ±0.19 / 36.57 ms │        35.78 / 36.43 ±0.51 / 37.35 ms │     no change │
│ QQuery 7  │        94.82 / 95.08 ±0.17 / 95.30 ms │        95.08 / 95.33 ±0.30 / 95.89 ms │     no change │
│ QQuery 8  │        36.29 / 37.61 ±1.93 / 41.42 ms │        36.74 / 38.15 ±1.78 / 41.57 ms │     no change │
│ QQuery 9  │        51.69 / 54.13 ±2.89 / 59.77 ms │        51.12 / 53.44 ±1.19 / 54.27 ms │     no change │
│ QQuery 10 │        62.14 / 62.41 ±0.24 / 62.77 ms │        62.28 / 62.85 ±0.43 / 63.40 ms │     no change │
│ QQuery 11 │     290.59 / 293.90 ±2.89 / 298.78 ms │     286.15 / 289.95 ±2.80 / 294.29 ms │     no change │
│ QQuery 12 │        28.03 / 28.47 ±0.35 / 29.07 ms │        27.95 / 28.51 ±0.36 / 29.07 ms │     no change │
│ QQuery 13 │     118.26 / 119.32 ±0.57 / 119.95 ms │     118.79 / 119.45 ±0.59 / 120.54 ms │     no change │
│ QQuery 14 │     407.88 / 409.57 ±1.37 / 411.51 ms │     407.39 / 409.55 ±1.93 / 412.97 ms │     no change │
│ QQuery 15 │        57.98 / 59.80 ±2.93 / 65.64 ms │        57.70 / 59.42 ±2.77 / 64.90 ms │     no change │
│ QQuery 16 │           6.67 / 6.87 ±0.29 / 7.43 ms │           6.73 / 6.83 ±0.14 / 7.11 ms │     no change │
│ QQuery 17 │        80.27 / 81.17 ±0.90 / 82.83 ms │        80.55 / 81.54 ±1.26 / 84.01 ms │     no change │
│ QQuery 18 │     123.83 / 126.61 ±3.49 / 132.28 ms │     125.80 / 127.61 ±2.97 / 133.53 ms │     no change │
│ QQuery 19 │        41.48 / 41.77 ±0.27 / 42.14 ms │        41.47 / 41.78 ±0.46 / 42.68 ms │     no change │
│ QQuery 20 │        34.59 / 35.16 ±0.68 / 36.46 ms │        34.73 / 35.18 ±0.40 / 35.80 ms │     no change │
│ QQuery 21 │        16.52 / 17.47 ±0.92 / 19.08 ms │        16.78 / 16.92 ±0.19 / 17.28 ms │     no change │
│ QQuery 22 │        61.56 / 62.20 ±0.52 / 62.86 ms │        61.69 / 62.34 ±0.56 / 63.05 ms │     no change │
│ QQuery 23 │     314.59 / 317.15 ±2.99 / 322.57 ms │     314.43 / 317.66 ±2.81 / 322.89 ms │     no change │
│ QQuery 24 │     223.22 / 225.86 ±2.85 / 230.23 ms │     222.85 / 225.41 ±2.63 / 229.70 ms │     no change │
│ QQuery 25 │     110.59 / 112.29 ±2.02 / 116.17 ms │     110.74 / 113.96 ±3.86 / 118.99 ms │     no change │
│ QQuery 26 │        57.56 / 57.72 ±0.17 / 58.00 ms │        57.62 / 57.95 ±0.31 / 58.54 ms │     no change │
│ QQuery 27 │           6.77 / 6.84 ±0.12 / 7.08 ms │           6.63 / 6.73 ±0.10 / 6.93 ms │     no change │
│ QQuery 28 │        56.40 / 60.32 ±1.97 / 61.72 ms │        60.92 / 62.28 ±2.01 / 66.25 ms │     no change │
│ QQuery 29 │      97.95 / 100.07 ±2.28 / 104.44 ms │       97.13 / 99.04 ±1.78 / 101.96 ms │     no change │
│ QQuery 30 │        31.71 / 32.05 ±0.23 / 32.39 ms │        31.86 / 32.30 ±0.37 / 32.84 ms │     no change │
│ QQuery 31 │     110.33 / 112.13 ±1.76 / 115.18 ms │     110.58 / 112.53 ±2.28 / 116.17 ms │     no change │
│ QQuery 32 │        19.80 / 20.05 ±0.27 / 20.50 ms │        19.70 / 20.08 ±0.35 / 20.72 ms │     no change │
│ QQuery 33 │        38.16 / 38.51 ±0.23 / 38.73 ms │        38.30 / 38.65 ±0.32 / 39.18 ms │     no change │
│ QQuery 34 │           9.29 / 9.51 ±0.25 / 9.97 ms │          9.30 / 9.79 ±0.46 / 10.52 ms │     no change │
│ QQuery 35 │        70.45 / 71.53 ±0.94 / 73.25 ms │        70.56 / 71.47 ±0.53 / 72.05 ms │     no change │
│ QQuery 36 │           5.81 / 5.95 ±0.16 / 6.25 ms │           5.75 / 5.91 ±0.22 / 6.35 ms │     no change │
│ QQuery 37 │           6.71 / 6.80 ±0.10 / 6.96 ms │           6.72 / 6.83 ±0.07 / 6.93 ms │     no change │
│ QQuery 38 │        62.27 / 63.44 ±1.83 / 67.08 ms │        62.12 / 63.18 ±0.67 / 63.93 ms │     no change │
│ QQuery 39 │     443.42 / 452.66 ±6.02 / 460.43 ms │     445.31 / 450.97 ±3.76 / 456.64 ms │     no change │
│ QQuery 40 │        22.73 / 22.98 ±0.23 / 23.39 ms │        22.62 / 22.89 ±0.24 / 23.33 ms │     no change │
│ QQuery 41 │        11.20 / 11.33 ±0.20 / 11.72 ms │        11.01 / 11.14 ±0.12 / 11.34 ms │     no change │
│ QQuery 42 │        23.33 / 24.32 ±1.25 / 26.75 ms │        23.36 / 24.82 ±2.02 / 28.82 ms │     no change │
│ QQuery 43 │           4.85 / 4.93 ±0.06 / 5.04 ms │           4.71 / 4.79 ±0.11 / 4.99 ms │     no change │
│ QQuery 44 │           9.02 / 9.18 ±0.09 / 9.30 ms │           8.58 / 8.71 ±0.12 / 8.91 ms │ +1.05x faster │
│ QQuery 45 │        37.80 / 39.79 ±2.98 / 45.74 ms │        37.48 / 40.17 ±4.30 / 48.74 ms │     no change │
│ QQuery 46 │        11.57 / 11.96 ±0.38 / 12.69 ms │        11.25 / 11.96 ±0.40 / 12.50 ms │     no change │
│ QQuery 47 │     223.81 / 227.78 ±2.26 / 230.63 ms │     224.39 / 228.24 ±3.10 / 232.32 ms │     no change │
│ QQuery 48 │        95.88 / 96.34 ±0.25 / 96.63 ms │        96.16 / 96.68 ±0.41 / 97.14 ms │     no change │
│ QQuery 49 │        76.54 / 79.96 ±5.70 / 91.34 ms │        77.06 / 78.64 ±1.43 / 81.05 ms │     no change │
│ QQuery 50 │        59.67 / 59.85 ±0.15 / 60.03 ms │        59.09 / 59.81 ±0.68 / 60.87 ms │     no change │
│ QQuery 51 │        92.04 / 94.27 ±2.76 / 99.09 ms │        90.38 / 92.62 ±1.25 / 93.95 ms │     no change │
│ QQuery 52 │        23.80 / 24.04 ±0.24 / 24.37 ms │        23.47 / 23.90 ±0.23 / 24.12 ms │     no change │
│ QQuery 53 │        29.32 / 31.84 ±4.46 / 40.74 ms │        29.13 / 32.03 ±4.47 / 40.94 ms │     no change │
│ QQuery 54 │        55.15 / 55.72 ±0.37 / 56.14 ms │        54.46 / 55.02 ±0.49 / 55.90 ms │     no change │
│ QQuery 55 │        23.00 / 23.41 ±0.28 / 23.76 ms │        23.17 / 23.61 ±0.61 / 24.81 ms │     no change │
│ QQuery 56 │        38.38 / 39.33 ±1.20 / 41.68 ms │        38.29 / 38.79 ±0.38 / 39.40 ms │     no change │
│ QQuery 57 │     172.96 / 175.18 ±2.19 / 178.10 ms │     173.47 / 175.19 ±1.54 / 178.00 ms │     no change │
│ QQuery 58 │     113.21 / 114.25 ±1.33 / 116.84 ms │     113.23 / 115.54 ±2.75 / 120.85 ms │     no change │
│ QQuery 59 │     115.58 / 116.43 ±1.20 / 118.79 ms │     116.04 / 116.54 ±0.32 / 116.88 ms │     no change │
│ QQuery 60 │        39.02 / 39.64 ±0.51 / 40.51 ms │        38.62 / 39.27 ±0.41 / 39.71 ms │     no change │
│ QQuery 61 │        12.90 / 13.07 ±0.17 / 13.38 ms │        12.53 / 14.10 ±2.60 / 19.28 ms │  1.08x slower │
│ QQuery 62 │        45.07 / 47.49 ±3.71 / 54.83 ms │        45.57 / 46.06 ±0.32 / 46.36 ms │     no change │
│ QQuery 63 │        28.97 / 29.54 ±0.36 / 30.05 ms │        29.84 / 30.11 ±0.22 / 30.42 ms │     no change │
│ QQuery 64 │     394.02 / 405.58 ±8.23 / 416.92 ms │     393.93 / 404.78 ±8.85 / 415.30 ms │     no change │
│ QQuery 65 │     145.69 / 148.71 ±3.88 / 156.07 ms │     147.20 / 152.07 ±6.83 / 165.55 ms │     no change │
│ QQuery 66 │        78.64 / 79.79 ±0.84 / 81.20 ms │        78.94 / 79.59 ±0.44 / 80.22 ms │     no change │
│ QQuery 67 │     241.67 / 245.91 ±3.61 / 251.85 ms │     240.80 / 246.98 ±7.05 / 259.68 ms │     no change │
│ QQuery 68 │        11.63 / 11.96 ±0.21 / 12.22 ms │        11.55 / 11.75 ±0.22 / 12.17 ms │     no change │
│ QQuery 69 │        56.86 / 59.21 ±3.71 / 66.60 ms │        57.28 / 59.95 ±4.25 / 68.35 ms │     no change │
│ QQuery 70 │     104.16 / 104.94 ±0.89 / 106.63 ms │     103.50 / 107.13 ±5.79 / 118.66 ms │     no change │
│ QQuery 71 │        34.95 / 37.48 ±3.80 / 45.05 ms │        35.27 / 35.65 ±0.34 / 36.16 ms │     no change │
│ QQuery 72 │ 2088.19 / 2232.94 ±85.35 / 2354.36 ms │ 2116.43 / 2160.53 ±46.37 / 2227.43 ms │     no change │
│ QQuery 73 │          9.26 / 9.82 ±0.47 / 10.66 ms │          9.20 / 9.42 ±0.30 / 10.00 ms │     no change │
│ QQuery 74 │     168.11 / 173.50 ±6.14 / 184.20 ms │     167.05 / 169.92 ±2.65 / 174.76 ms │     no change │
│ QQuery 75 │     147.58 / 150.85 ±4.64 / 160.00 ms │     146.21 / 150.67 ±6.12 / 162.72 ms │     no change │
│ QQuery 76 │        35.19 / 35.48 ±0.21 / 35.68 ms │        34.90 / 35.49 ±0.48 / 36.27 ms │     no change │
│ QQuery 77 │        60.54 / 63.85 ±4.07 / 71.86 ms │        60.26 / 60.63 ±0.28 / 61.07 ms │ +1.05x faster │
│ QQuery 78 │     181.44 / 187.53 ±4.19 / 194.06 ms │     182.72 / 186.98 ±5.71 / 198.18 ms │     no change │
│ QQuery 79 │        66.84 / 67.71 ±0.56 / 68.56 ms │        66.46 / 68.57 ±2.60 / 72.74 ms │     no change │
│ QQuery 80 │     101.33 / 102.06 ±0.38 / 102.42 ms │      99.26 / 100.82 ±2.06 / 104.88 ms │     no change │
│ QQuery 81 │        25.50 / 25.79 ±0.23 / 26.15 ms │        25.14 / 25.38 ±0.19 / 25.56 ms │     no change │
│ QQuery 82 │        16.44 / 19.16 ±4.93 / 29.02 ms │        16.00 / 16.13 ±0.10 / 16.28 ms │ +1.19x faster │
│ QQuery 83 │        39.68 / 40.30 ±0.45 / 41.02 ms │        38.69 / 38.93 ±0.20 / 39.22 ms │     no change │
│ QQuery 84 │        34.51 / 34.82 ±0.17 / 35.01 ms │        35.06 / 38.14 ±4.35 / 46.52 ms │  1.10x slower │
│ QQuery 85 │     108.25 / 113.02 ±5.47 / 121.91 ms │     106.99 / 108.47 ±0.93 / 109.84 ms │     no change │
│ QQuery 86 │        25.00 / 25.43 ±0.36 / 25.88 ms │        24.45 / 24.69 ±0.22 / 25.06 ms │     no change │
│ QQuery 87 │        63.75 / 64.88 ±0.88 / 66.13 ms │        63.03 / 64.31 ±1.36 / 66.77 ms │     no change │
│ QQuery 88 │        61.71 / 63.19 ±1.57 / 66.13 ms │        61.39 / 61.80 ±0.44 / 62.63 ms │     no change │
│ QQuery 89 │        35.46 / 36.36 ±0.60 / 37.16 ms │        35.23 / 35.79 ±0.51 / 36.71 ms │     no change │
│ QQuery 90 │        16.97 / 18.58 ±3.02 / 24.61 ms │        16.80 / 16.94 ±0.13 / 17.18 ms │ +1.10x faster │
│ QQuery 91 │        44.53 / 45.73 ±0.68 / 46.49 ms │        44.09 / 44.44 ±0.22 / 44.75 ms │     no change │
│ QQuery 92 │        28.83 / 30.05 ±0.89 / 31.10 ms │        28.61 / 32.45 ±3.86 / 38.44 ms │  1.08x slower │
│ QQuery 93 │        50.97 / 52.26 ±1.37 / 54.89 ms │        50.06 / 50.89 ±0.79 / 52.30 ms │     no change │
│ QQuery 94 │        38.56 / 39.16 ±0.49 / 39.74 ms │        37.96 / 38.36 ±0.30 / 38.76 ms │     no change │
│ QQuery 95 │        84.95 / 86.32 ±1.41 / 88.94 ms │        82.76 / 83.57 ±0.55 / 84.28 ms │     no change │
│ QQuery 96 │        24.15 / 24.36 ±0.23 / 24.80 ms │        24.04 / 25.48 ±2.37 / 30.20 ms │     no change │
│ QQuery 97 │        45.67 / 46.41 ±0.38 / 46.75 ms │        45.67 / 46.87 ±1.68 / 50.12 ms │     no change │
│ QQuery 98 │        42.05 / 43.54 ±0.83 / 44.42 ms │        42.06 / 42.68 ±0.38 / 43.26 ms │     no change │
│ QQuery 99 │        70.04 / 71.87 ±2.58 / 76.86 ms │        68.76 / 69.40 ±0.55 / 70.35 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary        ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)        │ 10298.76ms │
│ Total Time (array_nan)   │ 10204.60ms │
│ Average Time (HEAD)      │   104.03ms │
│ Average Time (array_nan) │   103.08ms │
│ Queries Faster           │          4 │
│ Queries Slower           │          3 │
│ Queries with No Change   │         92 │
│ Queries with Failure     │          0 │
└──────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric Value
Wall time 55.0s
Peak memory 5.6 GiB
Avg memory 5.0 GiB
CPU user 236.2s
CPU sys 6.7s
Peak spill 0 B

tpcds — branch

Metric Value
Wall time 55.0s
Peak memory 5.4 GiB
Avg memory 4.8 GiB
CPU user 235.5s
CPU sys 6.3s
Peak spill 0 B

File an issue against this benchmark runner

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

common Related to common crate functions Changes to functions implementation physical-expr Changes to the physical-expr crates physical-plan Changes to the physical-plan crate sqllogictest SQL Logic Tests (.slt)

Projects

None yet

Milestone

No milestone

2 participants

@comphead @adriangbot