Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix SM100 histogram tunings #3691

Merged
merged 6 commits into from
Feb 6, 2025

Conversation

bernhardmgruber
Copy link
Contributor

@bernhardmgruber bernhardmgruber commented Feb 5, 2025

The tuning data member names did not match the one used when selecting tunings, so all SM100 tunings were SFINAE-ed out.

  • SASS diff for SM100 on cub.bench.radix_sort.pairs.base is empty (no side effect)
  • SASS diff for SM100 on cub.bench.histogram.even.base contains instruction changes (tunings have effect)
  • Perf diff between before Add b200 tunings for histogram #3616 got merged and this PR.
  • Perf diff for multi histograms

Copy link
Contributor

github-actions bot commented Feb 5, 2025

🟩 CI finished in 1h 42m: Pass: 100%/90 | Total: 2d 16h | Avg: 43m 05s | Max: 1h 16m | Hits: 214%/13398
  • 🟩 cub: Pass: 100%/44 | Total: 1d 16h | Avg: 54m 44s | Max: 1h 16m | Hits: 216%/4168

    🟩 cpu
      🟩 amd64              Pass: 100%/42  | Total:  1d 14h | Avg: 54m 28s | Max:  1h 16m | Hits: 216%/4168  
      🟩 arm64              Pass: 100%/2   | Total:  2h 00m | Avg:  1h 00m | Max:  1h 01m
    🟩 ctk
      🟩 12.0               Pass: 100%/5   | Total:  4h 56m | Avg: 59m 21s | Max:  1h 02m | Hits: 217%/1042  
      🟩 12.5               Pass: 100%/2   | Total:  2h 15m | Avg:  1h 07m | Max:  1h 08m
      🟩 12.8               Pass: 100%/37  | Total:  1d 08h | Avg: 53m 25s | Max:  1h 16m | Hits: 216%/3126  
    🟩 cudacxx
      🟩 ClangCUDA18        Pass: 100%/2   | Total:  2h 01m | Avg:  1h 00m | Max:  1h 00m
      🟩 nvcc12.0           Pass: 100%/5   | Total:  4h 56m | Avg: 59m 21s | Max:  1h 02m | Hits: 217%/1042  
      🟩 nvcc12.5           Pass: 100%/2   | Total:  2h 15m | Avg:  1h 07m | Max:  1h 08m
      🟩 nvcc12.8           Pass: 100%/35  | Total:  1d 06h | Avg: 53m 00s | Max:  1h 16m | Hits: 216%/3126  
    🟩 cudacxx_family
      🟩 ClangCUDA          Pass: 100%/2   | Total:  2h 01m | Avg:  1h 00m | Max:  1h 00m
      🟩 nvcc               Pass: 100%/42  | Total:  1d 14h | Avg: 54m 27s | Max:  1h 16m | Hits: 216%/4168  
    🟩 cxx
      🟩 Clang14            Pass: 100%/4   | Total:  3h 57m | Avg: 59m 24s | Max:  1h 02m
      🟩 Clang15            Pass: 100%/2   | Total:  2h 00m | Avg:  1h 00m | Max:  1h 02m
      🟩 Clang16            Pass: 100%/2   | Total:  2h 04m | Avg:  1h 02m | Max:  1h 05m
      🟩 Clang17            Pass: 100%/2   | Total:  2h 08m | Avg:  1h 04m | Max:  1h 04m
      🟩 Clang18            Pass: 100%/7   | Total:  5h 44m | Avg: 49m 10s | Max:  1h 00m
      🟩 GCC7               Pass: 100%/2   | Total:  1h 56m | Avg: 58m 12s | Max: 58m 14s
      🟩 GCC8               Pass: 100%/1   | Total:  1h 00m | Avg:  1h 00m | Max:  1h 00m
      🟩 GCC9               Pass: 100%/2   | Total:  1h 58m | Avg: 59m 04s | Max: 59m 48s
      🟩 GCC10              Pass: 100%/2   | Total:  1h 58m | Avg: 59m 18s | Max: 59m 46s
      🟩 GCC11              Pass: 100%/2   | Total:  2h 01m | Avg:  1h 00m | Max:  1h 02m
      🟩 GCC12              Pass: 100%/2   | Total:  1h 56m | Avg: 58m 14s | Max: 59m 17s
      🟩 GCC13              Pass: 100%/10  | Total:  6h 24m | Avg: 38m 27s | Max:  1h 08m
      🟩 MSVC14.29          Pass: 100%/2   | Total:  2h 13m | Avg:  1h 06m | Max:  1h 10m | Hits: 217%/2084  
      🟩 MSVC14.39          Pass: 100%/2   | Total:  2h 28m | Avg:  1h 14m | Max:  1h 16m | Hits: 216%/2084  
      🟩 NVHPC24.7          Pass: 100%/2   | Total:  2h 15m | Avg:  1h 07m | Max:  1h 08m
    🟩 cxx_family
      🟩 Clang              Pass: 100%/17  | Total: 15h 55m | Avg: 56m 11s | Max:  1h 05m
      🟩 GCC                Pass: 100%/21  | Total: 17h 16m | Avg: 49m 20s | Max:  1h 08m
      🟩 MSVC               Pass: 100%/4   | Total:  4h 41m | Avg:  1h 10m | Max:  1h 16m | Hits: 216%/4168  
      🟩 NVHPC              Pass: 100%/2   | Total:  2h 15m | Avg:  1h 07m | Max:  1h 08m
    🟩 gpu
      🟩 h100               Pass: 100%/2   | Total: 54m 04s | Avg: 27m 02s | Max: 29m 24s
      🟩 rtx2080            Pass: 100%/34  | Total:  1d 11h | Avg:  1h 01m | Max:  1h 16m | Hits: 216%/4168  
      🟩 rtxa6000           Pass: 100%/8   | Total:  4h 08m | Avg: 31m 05s | Max:  1h 00m
    🟩 jobs
      🟩 Build              Pass: 100%/37  | Total:  1d 13h | Avg:  1h 00m | Max:  1h 16m | Hits: 216%/4168  
      🟩 DeviceLaunch       Pass: 100%/1   | Total: 23m 33s | Avg: 23m 33s | Max: 23m 33s
      🟩 GraphCapture       Pass: 100%/1   | Total: 17m 43s | Avg: 17m 43s | Max: 17m 43s
      🟩 HostLaunch         Pass: 100%/3   | Total:  1h 12m | Avg: 24m 18s | Max: 25m 13s
      🟩 TestGPU            Pass: 100%/2   | Total: 42m 09s | Avg: 21m 04s | Max: 22m 09s
    🟩 sm
      🟩 90                 Pass: 100%/2   | Total: 54m 04s | Avg: 27m 02s | Max: 29m 24s
      🟩 90;90a;100         Pass: 100%/1   | Total:  1h 08m | Avg:  1h 08m | Max:  1h 08m
    🟩 std
      🟩 17                 Pass: 100%/20  | Total: 20h 24m | Avg:  1h 01m | Max:  1h 11m | Hits: 217%/3126  
      🟩 20                 Pass: 100%/24  | Total: 19h 43m | Avg: 49m 19s | Max:  1h 16m | Hits: 216%/1042  
    
  • 🟩 thrust: Pass: 100%/43 | Total: 23h 57m | Avg: 33m 26s | Max: 1h 03m | Hits: 212%/9230

    🟩 cmake_options
      🟩 -DTHRUST_DISPATCH_TYPE=Force32bit Pass: 100%/2   | Total: 39m 41s | Avg: 19m 50s | Max: 28m 43s
    🟩 cpu
      🟩 amd64              Pass: 100%/41  | Total: 22h 58m | Avg: 33m 37s | Max:  1h 03m | Hits: 212%/9230  
      🟩 arm64              Pass: 100%/2   | Total: 59m 22s | Avg: 29m 41s | Max: 30m 48s
    🟩 ctk
      🟩 12.0               Pass: 100%/5   | Total:  3h 05m | Avg: 37m 02s | Max: 53m 19s | Hits: 174%/1846  
      🟩 12.5               Pass: 100%/2   | Total:  1h 53m | Avg: 56m 58s | Max: 59m 15s
      🟩 12.8               Pass: 100%/36  | Total: 18h 58m | Avg: 31m 37s | Max:  1h 03m | Hits: 222%/7384  
    🟩 cudacxx
      🟩 ClangCUDA18        Pass: 100%/2   | Total: 52m 18s | Avg: 26m 09s | Max: 26m 10s
      🟩 nvcc12.0           Pass: 100%/5   | Total:  3h 05m | Avg: 37m 02s | Max: 53m 19s | Hits: 174%/1846  
      🟩 nvcc12.5           Pass: 100%/2   | Total:  1h 53m | Avg: 56m 58s | Max: 59m 15s
      🟩 nvcc12.8           Pass: 100%/34  | Total: 18h 06m | Avg: 31m 57s | Max:  1h 03m | Hits: 222%/7384  
    🟩 cudacxx_family
      🟩 ClangCUDA          Pass: 100%/2   | Total: 52m 18s | Avg: 26m 09s | Max: 26m 10s
      🟩 nvcc               Pass: 100%/41  | Total: 23h 05m | Avg: 33m 47s | Max:  1h 03m | Hits: 212%/9230  
    🟩 cxx
      🟩 Clang14            Pass: 100%/4   | Total:  2h 07m | Avg: 31m 46s | Max: 32m 53s
      🟩 Clang15            Pass: 100%/2   | Total:  1h 03m | Avg: 31m 54s | Max: 32m 32s
      🟩 Clang16            Pass: 100%/2   | Total:  1h 03m | Avg: 31m 31s | Max: 33m 12s
      🟩 Clang17            Pass: 100%/2   | Total:  1h 05m | Avg: 32m 48s | Max: 33m 59s
      🟩 Clang18            Pass: 100%/7   | Total:  2h 43m | Avg: 23m 19s | Max: 33m 23s
      🟩 GCC7               Pass: 100%/2   | Total:  1h 06m | Avg: 33m 05s | Max: 34m 30s
      🟩 GCC8               Pass: 100%/1   | Total: 33m 40s | Avg: 33m 40s | Max: 33m 40s
      🟩 GCC9               Pass: 100%/2   | Total:  1h 06m | Avg: 33m 18s | Max: 34m 35s
      🟩 GCC10              Pass: 100%/2   | Total:  1h 05m | Avg: 32m 53s | Max: 35m 11s
      🟩 GCC11              Pass: 100%/2   | Total:  1h 09m | Avg: 34m 41s | Max: 34m 45s
      🟩 GCC12              Pass: 100%/2   | Total:  1h 13m | Avg: 36m 39s | Max: 37m 02s
      🟩 GCC13              Pass: 100%/8   | Total:  3h 21m | Avg: 25m 08s | Max: 38m 27s
      🟩 MSVC14.29          Pass: 100%/2   | Total:  1h 52m | Avg: 56m 18s | Max: 59m 18s | Hits: 174%/3692  
      🟩 MSVC14.39          Pass: 100%/3   | Total:  2h 32m | Avg: 50m 48s | Max:  1h 03m | Hits: 238%/5538  
      🟩 NVHPC24.7          Pass: 100%/2   | Total:  1h 53m | Avg: 56m 58s | Max: 59m 15s
    🟩 cxx_family
      🟩 Clang              Pass: 100%/17  | Total:  8h 02m | Avg: 28m 24s | Max: 33m 59s
      🟩 GCC                Pass: 100%/19  | Total:  9h 36m | Avg: 30m 19s | Max: 38m 27s
      🟩 MSVC               Pass: 100%/5   | Total:  4h 25m | Avg: 53m 00s | Max:  1h 03m | Hits: 212%/9230  
      🟩 NVHPC              Pass: 100%/2   | Total:  1h 53m | Avg: 56m 58s | Max: 59m 15s
    🟩 gpu
      🟩 rtx2080            Pass: 100%/33  | Total: 19h 51m | Avg: 36m 07s | Max: 59m 18s | Hits: 174%/5538  
      🟩 rtx4090            Pass: 100%/10  | Total:  4h 05m | Avg: 24m 35s | Max:  1h 03m | Hits: 270%/3692  
    🟩 jobs
      🟩 Build              Pass: 100%/37  | Total: 22h 35m | Avg: 36m 38s | Max:  1h 03m | Hits: 174%/7384  
      🟩 TestCPU            Pass: 100%/3   | Total: 49m 34s | Avg: 16m 31s | Max: 33m 17s | Hits: 365%/1846  
      🟩 TestGPU            Pass: 100%/3   | Total: 32m 37s | Avg: 10m 52s | Max: 11m 15s
    🟩 sm
      🟩 90;90a;100         Pass: 100%/1   | Total: 38m 27s | Avg: 38m 27s | Max: 38m 27s
    🟩 std
      🟩 17                 Pass: 100%/20  | Total: 12h 26m | Avg: 37m 19s | Max: 59m 18s | Hits: 174%/5538  
      🟩 20                 Pass: 100%/21  | Total: 10h 51m | Avg: 31m 02s | Max:  1h 03m | Hits: 270%/3692  
    
  • 🟩 cccl_c_parallel: Pass: 100%/2 | Total: 7m 32s | Avg: 3m 46s | Max: 5m 15s

    🟩 cpu
      🟩 amd64              Pass: 100%/2   | Total:  7m 32s | Avg:  3m 46s | Max:  5m 15s
    🟩 ctk
      🟩 12.8               Pass: 100%/2   | Total:  7m 32s | Avg:  3m 46s | Max:  5m 15s
    🟩 cudacxx
      🟩 nvcc12.8           Pass: 100%/2   | Total:  7m 32s | Avg:  3m 46s | Max:  5m 15s
    🟩 cudacxx_family
      🟩 nvcc               Pass: 100%/2   | Total:  7m 32s | Avg:  3m 46s | Max:  5m 15s
    🟩 cxx
      🟩 GCC13              Pass: 100%/2   | Total:  7m 32s | Avg:  3m 46s | Max:  5m 15s
    🟩 cxx_family
      🟩 GCC                Pass: 100%/2   | Total:  7m 32s | Avg:  3m 46s | Max:  5m 15s
    🟩 gpu
      🟩 rtx2080            Pass: 100%/2   | Total:  7m 32s | Avg:  3m 46s | Max:  5m 15s
    🟩 jobs
      🟩 Build              Pass: 100%/1   | Total:  2m 17s | Avg:  2m 17s | Max:  2m 17s
      🟩 Test               Pass: 100%/1   | Total:  5m 15s | Avg:  5m 15s | Max:  5m 15s
    
  • 🟩 python: Pass: 100%/1 | Total: 25m 05s | Avg: 25m 05s | Max: 25m 05s

    🟩 cpu
      🟩 amd64              Pass: 100%/1   | Total: 25m 05s | Avg: 25m 05s | Max: 25m 05s
    🟩 ctk
      🟩 12.8               Pass: 100%/1   | Total: 25m 05s | Avg: 25m 05s | Max: 25m 05s
    🟩 cudacxx
      🟩 nvcc12.8           Pass: 100%/1   | Total: 25m 05s | Avg: 25m 05s | Max: 25m 05s
    🟩 cudacxx_family
      🟩 nvcc               Pass: 100%/1   | Total: 25m 05s | Avg: 25m 05s | Max: 25m 05s
    🟩 cxx
      🟩 GCC13              Pass: 100%/1   | Total: 25m 05s | Avg: 25m 05s | Max: 25m 05s
    🟩 cxx_family
      🟩 GCC                Pass: 100%/1   | Total: 25m 05s | Avg: 25m 05s | Max: 25m 05s
    🟩 gpu
      🟩 rtx2080            Pass: 100%/1   | Total: 25m 05s | Avg: 25m 05s | Max: 25m 05s
    🟩 jobs
      🟩 Test               Pass: 100%/1   | Total: 25m 05s | Avg: 25m 05s | Max: 25m 05s
    

👃 Inspect Changes

Modifications in project?

Project
CCCL Infrastructure
libcu++
+/- CUB
Thrust
CUDA Experimental
python
CCCL C Parallel Library
Catch2Helper

Modifications in project or dependencies?

Project
CCCL Infrastructure
libcu++
+/- CUB
+/- Thrust
CUDA Experimental
+/- python
+/- CCCL C Parallel Library
+/- Catch2Helper

🏃‍ Runner counts (total jobs: 90)

# Runner
65 linux-amd64-cpu16
9 windows-amd64-cpu16
6 linux-amd64-gpu-rtxa6000-latest-1
4 linux-arm64-cpu16
3 linux-amd64-gpu-rtx4090-latest-1
2 linux-amd64-gpu-rtx2080-latest-1
1 linux-amd64-gpu-h100-latest-1

@miscco miscco enabled auto-merge (squash) February 5, 2025 13:16
@bernhardmgruber
Copy link
Contributor Author

Please don't merge until we have a perf diff from @gonidelis

@gonidelis
Copy link
Member

histogram.even (tbh I consider it great and exactly what we expected. The sole provided tuning does great for 2^28 problem sizes). We should maybe take a second look on the scarce SLOWs in the float,double land.

|  SampleT{ct}  |  CounterT{ct}  |  OffsetT{ct}  |  Elements{io}  |  Bins   |  Entropy  |   Ref Time |   Ref Noise |   Cmp Time |   Cmp Noise |      Diff |   %Diff |  Status  |
|---------------|----------------|---------------|----------------|---------|-----------|------------|-------------|------------|-------------|-----------|---------|----------|
|      I8       |      I32       |      I32      |      2^16      |   32    |   0.201   |  11.636 us |      15.84% |  12.082 us |      12.16% |  0.446 us |   3.83% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      |   32    |   0.201   |  12.084 us |      11.31% |  12.798 us |       7.61% |  0.714 us |   5.91% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      |   32    |   0.201   |  18.929 us |       5.08% |  19.296 us |       3.13% |  0.367 us |   1.94% |   SAME   |
|      I8       |      I32       |      I32      |      2^28      |   32    |   0.201   | 129.858 us |       1.05% | 121.788 us |       0.97% | -8.070 us |  -6.21% |   FAST   |
|      I8       |      I32       |      I32      |      2^16      |   128   |   0.201   |  11.949 us |      13.36% |  12.368 us |       9.47% |  0.419 us |   3.51% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      |   128   |   0.201   |  12.232 us |      10.56% |  12.519 us |       8.23% |  0.287 us |   2.34% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      |   128   |   0.201   |  19.068 us |       4.43% |  19.258 us |       3.26% |  0.191 us |   1.00% |   SAME   |
|      I8       |      I32       |      I32      |      2^28      |   128   |   0.201   | 130.085 us |       1.01% | 122.038 us |       0.82% | -8.047 us |  -6.19% |   FAST   |
|      I8       |      I32       |      I32      |      2^16      |  2048   |   0.201   |  12.061 us |      11.92% |  12.281 us |       9.70% |  0.221 us |   1.83% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      |  2048   |   0.201   |  13.880 us |      12.19% |  14.044 us |      10.02% |  0.165 us |   1.19% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      |  2048   |   0.201   |  28.314 us |       5.43% |  27.527 us |       1.88% | -0.787 us |  -2.78% |   FAST   |
|      I8       |      I32       |      I32      |      2^28      |  2048   |   0.201   | 133.788 us |       0.64% | 126.094 us |       1.03% | -7.693 us |  -5.75% |   FAST   |
|      I8       |      I32       |      I32      |      2^16      | 2097152 |   0.201   |  16.439 us |       6.46% |  16.496 us |       6.19% |  0.058 us |   0.35% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      | 2097152 |   0.201   |  17.916 us |       9.54% |  17.208 us |       5.88% | -0.708 us |  -3.95% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      | 2097152 |   0.201   |  32.238 us |       2.90% |  31.763 us |       0.49% | -0.475 us |  -1.47% |   FAST   |
|      I8       |      I32       |      I32      |      2^28      | 2097152 |   0.201   | 138.509 us |       0.52% | 130.023 us |       0.79% | -8.485 us |  -6.13% |   FAST   |
|      I8       |      I32       |      I32      |      2^16      |   32    |     1     |  12.057 us |      11.67% |  12.255 us |      10.42% |  0.199 us |   1.65% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      |   32    |     1     |  12.460 us |      10.02% |  12.728 us |       7.84% |  0.268 us |   2.15% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      |   32    |     1     |  19.100 us |       4.41% |  19.185 us |       3.77% |  0.085 us |   0.45% |   SAME   |
|      I8       |      I32       |      I32      |      2^28      |   32    |     1     | 130.076 us |       0.98% | 121.835 us |       0.94% | -8.241 us |  -6.34% |   FAST   |
|      I8       |      I32       |      I32      |      2^16      |   128   |     1     |  11.983 us |      13.04% |  12.309 us |      10.00% |  0.326 us |   2.72% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      |   128   |     1     |  12.431 us |       9.16% |  12.612 us |       7.83% |  0.181 us |   1.45% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      |   128   |     1     |  19.073 us |       4.60% |  19.283 us |       3.21% |  0.210 us |   1.10% |   SAME   |
|      I8       |      I32       |      I32      |      2^28      |   128   |     1     | 130.289 us |       1.07% | 124.020 us |       0.98% | -6.269 us |  -4.81% |   FAST   |
|      I8       |      I32       |      I32      |      2^16      |  2048   |     1     |  12.220 us |      10.86% |  12.471 us |       9.46% |  0.250 us |   2.05% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      |  2048   |     1     |  20.918 us |       4.30% |  19.160 us |       3.79% | -1.758 us |  -8.40% |   FAST   |
|      I8       |      I32       |      I32      |      2^24      |  2048   |     1     |  41.867 us |       1.24% |  41.732 us |       1.70% | -0.135 us |  -0.32% |   SAME   |
|      I8       |      I32       |      I32      |      2^28      |  2048   |     1     | 141.692 us |       0.72% | 132.512 us |       1.10% | -9.180 us |  -6.48% |   FAST   |
|      I8       |      I32       |      I32      |      2^16      | 2097152 |     1     |  17.312 us |       3.08% |  17.423 us |       1.45% |  0.112 us |   0.64% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      | 2097152 |     1     |  25.026 us |       3.71% |  23.227 us |       3.33% | -1.799 us |  -7.19% |   FAST   |
|      I8       |      I32       |      I32      |      2^24      | 2097152 |     1     |  46.140 us |       0.42% |  45.881 us |       1.47% | -0.259 us |  -0.56% |   FAST   |
|      I8       |      I32       |      I32      |      2^28      | 2097152 |     1     | 145.672 us |       0.71% | 136.377 us |       0.49% | -9.296 us |  -6.38% |   FAST   |
|      I16      |      I32       |      I32      |      2^16      |   32    |   0.201   |  13.217 us |       4.78% |  13.296 us |       3.00% |  0.079 us |   0.60% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      |   32    |   0.201   |  13.324 us |      14.29% |  13.433 us |      14.97% |  0.108 us |   0.81% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      |   32    |   0.201   |  39.048 us |       2.54% |  39.237 us |       2.47% |  0.189 us |   0.48% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      |   32    |   0.201   | 427.039 us |       0.35% | 427.382 us |       0.35% |  0.343 us |   0.08% |   SAME   |
|      I16      |      I32       |      I32      |      2^16      |   128   |   0.201   |  14.176 us |       9.55% |  14.279 us |       8.26% |  0.103 us |   0.73% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      |   128   |   0.201   |  14.398 us |       9.36% |  14.474 us |       9.10% |  0.076 us |   0.53% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      |   128   |   0.201   |  39.662 us |       1.90% |  39.819 us |       1.35% |  0.157 us |   0.40% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      |   128   |   0.201   | 441.708 us |       0.29% | 441.846 us |       0.29% |  0.138 us |   0.03% |   SAME   |
|      I16      |      I32       |      I32      |      2^16      |  2048   |   0.201   |  17.415 us |       1.37% |  17.397 us |       1.69% | -0.018 us |  -0.10% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      |  2048   |   0.201   |  22.113 us |       7.76% |  22.320 us |       7.09% |  0.207 us |   0.94% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      |  2048   |   0.201   | 119.828 us |       0.67% | 119.854 us |       0.65% |  0.026 us |   0.02% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      |  2048   |   0.201   |   1.721 ms |       0.12% |   1.723 ms |       0.12% |  2.302 us |   0.13% |   SLOW   |
|      I16      |      I32       |      I32      |      2^16      | 2097152 |   0.201   | 547.951 us |       0.10% | 547.994 us |       0.10% |  0.043 us |   0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      | 2097152 |   0.201   |   1.158 ms |       0.16% |   1.158 ms |       0.17% |  0.048 us |   0.00% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      | 2097152 |   0.201   |   1.531 ms |       0.41% |   1.531 ms |       0.44% |  0.229 us |   0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      | 2097152 |   0.201   |   2.013 ms |       0.82% |   2.013 ms |       0.85% |  0.139 us |   0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^16      |   32    |     1     |  13.400 us |       4.85% |  13.434 us |       4.93% |  0.034 us |   0.25% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      |   32    |     1     |  13.790 us |      12.81% |  13.873 us |      12.54% |  0.083 us |   0.60% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      |   32    |     1     |  39.063 us |       2.58% |  39.282 us |       2.45% |  0.219 us |   0.56% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      |   32    |     1     | 440.844 us |       0.32% | 440.781 us |       0.30% | -0.063 us |  -0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^16      |   128   |     1     |  13.392 us |       4.15% |  13.337 us |       5.29% | -0.055 us |  -0.41% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      |   128   |     1     |  13.789 us |      13.11% |  13.708 us |      13.56% | -0.082 us |  -0.59% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      |   128   |     1     |  39.149 us |       2.73% |  39.050 us |       2.99% | -0.099 us |  -0.25% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      |   128   |     1     | 425.158 us |       0.23% | 425.219 us |       0.25% |  0.061 us |   0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^16      |  2048   |     1     |  19.459 us |       1.39% |  19.423 us |       2.03% | -0.036 us |  -0.18% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      |  2048   |     1     |  21.254 us |       3.30% |  21.304 us |       3.03% |  0.050 us |   0.23% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      |  2048   |     1     | 115.711 us |       0.84% | 115.526 us |       0.92% | -0.185 us |  -0.16% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      |  2048   |     1     |   1.639 ms |       0.31% |   1.638 ms |       0.31% | -0.595 us |  -0.04% |   SAME   |
|      I16      |      I32       |      I32      |      2^16      | 2097152 |     1     | 547.967 us |       0.09% | 548.047 us |       0.11% |  0.080 us |   0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      | 2097152 |     1     |   1.158 ms |       0.16% |   1.158 ms |       0.17% | -0.065 us |  -0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      | 2097152 |     1     |   1.531 ms |       0.42% |   1.531 ms |       0.44% |  0.048 us |   0.00% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      | 2097152 |     1     |   2.014 ms |       0.82% |   2.013 ms |       0.78% | -0.179 us |  -0.01% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      |   32    |   0.201   |  15.378 us |       1.04% |  15.320 us |       2.35% | -0.058 us |  -0.38% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      |   32    |   0.201   |  16.465 us |       6.31% |  16.446 us |       6.29% | -0.019 us |  -0.12% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      |   32    |   0.201   |  48.080 us |       1.16% |  47.941 us |       1.40% | -0.139 us |  -0.29% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      |   32    |   0.201   | 565.908 us |       0.16% | 565.755 us |       0.17% | -0.153 us |  -0.03% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      |   128   |   0.201   |  15.262 us |       3.24% |  15.080 us |       4.72% | -0.183 us |  -1.20% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      |   128   |   0.201   |  16.465 us |       6.33% |  16.551 us |       6.27% |  0.086 us |   0.52% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      |   128   |   0.201   |  47.964 us |       3.13% |  47.733 us |       3.06% | -0.231 us |  -0.48% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      |   128   |   0.201   | 566.748 us |       0.25% | 566.494 us |       0.24% | -0.254 us |  -0.04% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      |  2048   |   0.201   |  17.203 us |       3.69% |  17.041 us |       4.72% | -0.162 us |  -0.94% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      |  2048   |   0.201   |  21.308 us |       2.93% |  21.194 us |       3.63% | -0.115 us |  -0.54% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      |  2048   |   0.201   |  92.571 us |       1.30% |  92.636 us |       1.36% |  0.065 us |   0.07% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      |  2048   |   0.201   |   1.498 ms |       0.10% |   1.497 ms |       0.11% | -0.118 us |  -0.01% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      | 2097152 |   0.201   |   1.949 ms |       0.17% |   1.949 ms |       0.17% |  0.230 us |   0.01% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      | 2097152 |   0.201   |   2.799 ms |       0.11% |   2.799 ms |       0.11% | -0.073 us |  -0.00% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      | 2097152 |   0.201   |   3.425 ms |       0.26% |   3.425 ms |       0.26% |  0.446 us |   0.01% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      | 2097152 |   0.201   |   5.860 ms |       0.23% |   5.860 ms |       0.23% | -0.227 us |  -0.00% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      |   32    |     1     |  15.382 us |       1.20% |  15.336 us |       2.15% | -0.046 us |  -0.30% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      |   32    |     1     |  16.463 us |       6.24% |  16.512 us |       6.20% |  0.049 us |   0.30% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      |   32    |     1     |  50.037 us |       1.07% |  49.962 us |       1.36% | -0.074 us |  -0.15% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      |   32    |     1     | 582.436 us |       0.12% | 582.362 us |       0.13% | -0.074 us |  -0.01% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      |   128   |     1     |  15.323 us |       2.33% |  15.230 us |       3.58% | -0.093 us |  -0.61% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      |   128   |     1     |  16.477 us |       6.40% |  16.476 us |       6.24% | -0.001 us |  -0.00% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      |   128   |     1     |  47.317 us |       2.17% |  47.318 us |       2.10% |  0.001 us |   0.00% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      |   128   |     1     | 562.924 us |       0.23% | 562.901 us |       0.23% | -0.023 us |  -0.00% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      |  2048   |     1     |  18.754 us |       5.19% |  18.802 us |       5.19% |  0.048 us |   0.26% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      |  2048   |     1     |  23.362 us |       2.72% |  23.282 us |       3.06% | -0.080 us |  -0.34% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      |  2048   |     1     | 102.712 us |       1.41% | 102.674 us |       1.37% | -0.037 us |  -0.04% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      |  2048   |     1     |   1.567 ms |       2.20% |   1.570 ms |       2.10% |  3.357 us |   0.21% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      | 2097152 |     1     |   1.999 ms |       0.16% |   1.999 ms |       0.15% | -0.144 us |  -0.01% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      | 2097152 |     1     |   2.829 ms |       0.13% |   2.830 ms |       0.12% |  0.229 us |   0.01% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      | 2097152 |     1     |   3.994 ms |       0.25% |   3.995 ms |       0.24% |  0.625 us |   0.02% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      | 2097152 |     1     |  15.210 ms |       0.59% |  15.217 ms |       0.58% |  7.412 us |   0.05% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      |   32    |   0.201   |  13.152 us |       4.34% |  13.089 us |       4.89% | -0.063 us |  -0.48% |   SAME   |
|      I64      |      I32       |      I32      |      2^20      |   32    |   0.201   |  14.395 us |       7.39% |  14.535 us |       7.41% |  0.140 us |   0.97% |   SAME   |
|      I64      |      I32       |      I32      |      2^24      |   32    |   0.201   |  52.252 us |       0.43% |  52.184 us |       0.82% | -0.068 us |  -0.13% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      |   32    |   0.201   | 609.578 us |       0.25% | 609.583 us |       0.26% |  0.005 us |   0.00% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      |   128   |   0.201   |  13.323 us |       1.02% |  13.278 us |       1.99% | -0.044 us |  -0.33% |   SAME   |
|      I64      |      I32       |      I32      |      2^20      |   128   |   0.201   |  15.161 us |       4.19% |  15.104 us |       4.78% | -0.057 us |  -0.38% |   SAME   |
|      I64      |      I32       |      I32      |      2^24      |   128   |   0.201   |  52.149 us |       1.78% |  51.993 us |       2.07% | -0.156 us |  -0.30% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      |   128   |   0.201   | 616.253 us |       0.31% | 616.188 us |       0.30% | -0.065 us |  -0.01% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      |  2048   |   0.201   |  15.336 us |       1.61% |  15.250 us |       3.46% | -0.086 us |  -0.56% |   SAME   |
|      I64      |      I32       |      I32      |      2^20      |  2048   |   0.201   |  19.304 us |       3.17% |  19.163 us |       3.75% | -0.140 us |  -0.73% |   SAME   |
|      I64      |      I32       |      I32      |      2^24      |  2048   |   0.201   | 124.684 us |       1.01% | 124.552 us |       1.02% | -0.132 us |  -0.11% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      |  2048   |   0.201   |   1.776 ms |       0.10% |   1.777 ms |       0.09% |  0.227 us |   0.01% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      | 2097152 |   0.201   |   2.604 ms |       0.16% |   2.604 ms |       0.16% | -0.054 us |  -0.00% |   SAME   |
|      I64      |      I32       |      I32      |      2^20      | 2097152 |   0.201   |   3.097 ms |       0.18% |   3.097 ms |       0.18% |  0.086 us |   0.00% |   SAME   |
|      I64      |      I32       |      I32      |      2^24      | 2097152 |   0.201   |   3.452 ms |       0.25% |   3.451 ms |       0.25% | -1.101 us |  -0.03% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      | 2097152 |   0.201   |   5.995 ms |       0.22% |   5.995 ms |       0.24% | -0.096 us |  -0.00% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      |   32    |     1     |  13.123 us |       4.48% |  13.140 us |       4.42% |  0.016 us |   0.12% |   SAME   |
|      I64      |      I32       |      I32      |      2^20      |   32    |     1     |  14.616 us |       8.40% |  14.641 us |       8.13% |  0.025 us |   0.17% |   SAME   |
|      I64      |      I32       |      I32      |      2^24      |   32    |     1     |  53.649 us |       1.81% |  53.771 us |       1.76% |  0.122 us |   0.23% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      |   32    |     1     | 625.305 us |       0.18% | 625.247 us |       0.18% | -0.058 us |  -0.01% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      |   128   |     1     |  13.220 us |       3.66% |  13.162 us |       4.14% | -0.058 us |  -0.44% |   SAME   |
|      I64      |      I32       |      I32      |      2^20      |   128   |     1     |  14.767 us |       6.94% |  14.784 us |       6.65% |  0.018 us |   0.12% |   SAME   |
|      I64      |      I32       |      I32      |      2^24      |   128   |     1     |  52.004 us |       1.26% |  51.899 us |       1.41% | -0.105 us |  -0.20% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      |   128   |     1     | 605.876 us |       0.29% | 605.905 us |       0.29% |  0.029 us |   0.00% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      |  2048   |     1     |  15.534 us |       3.75% |  15.510 us |       3.52% | -0.024 us |  -0.16% |   SAME   |
|      I64      |      I32       |      I32      |      2^20      |  2048   |     1     |  21.221 us |       5.31% |  21.080 us |       5.16% | -0.141 us |  -0.66% |   SAME   |
|      I64      |      I32       |      I32      |      2^24      |  2048   |     1     | 121.883 us |       0.71% | 121.818 us |       0.81% | -0.065 us |  -0.05% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      |  2048   |     1     |   1.844 ms |       1.04% |   1.844 ms |       1.05% |  0.335 us |   0.02% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      | 2097152 |     1     |   2.589 ms |       0.11% |   2.589 ms |       0.11% | -0.094 us |  -0.00% |   SAME   |
|      I64      |      I32       |      I32      |      2^20      | 2097152 |     1     |   3.126 ms |       0.14% |   3.126 ms |       0.14% |  0.209 us |   0.01% |   SAME   |
|      I64      |      I32       |      I32      |      2^24      | 2097152 |     1     |   4.017 ms |       0.21% |   4.018 ms |       0.21% |  0.261 us |   0.01% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      | 2097152 |     1     |  15.380 ms |       0.26% |  15.382 ms |       0.28% |  2.827 us |   0.02% |   SAME   |
|      F32      |      I32       |      I32      |      2^16      |   32    |   0.201   |  12.797 us |       6.95% |  13.215 us |       3.79% |  0.418 us |   3.27% |   SAME   |
|      F32      |      I32       |      I32      |      2^20      |   32    |   0.201   |  13.963 us |       5.69% |  14.386 us |       7.02% |  0.423 us |   3.03% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      |   32    |   0.201   |  37.605 us |       2.43% |  37.925 us |       1.25% |  0.320 us |   0.85% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      |   32    |   0.201   | 418.240 us |       0.37% | 418.638 us |       0.31% |  0.398 us |   0.10% |   SAME   |
|      F32      |      I32       |      I32      |      2^16      |   128   |   0.201   |  12.950 us |       6.01% |  13.320 us |       1.78% |  0.370 us |   2.86% |   SLOW   |
|      F32      |      I32       |      I32      |      2^20      |   128   |   0.201   |  14.024 us |       6.15% |  14.545 us |       6.90% |  0.521 us |   3.71% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      |   128   |   0.201   |  37.512 us |       2.15% |  37.865 us |       0.87% |  0.353 us |   0.94% |   SLOW   |
|      F32      |      I32       |      I32      |      2^28      |   128   |   0.201   | 432.448 us |       0.36% | 432.605 us |       0.33% |  0.157 us |   0.04% |   SAME   |
|      F32      |      I32       |      I32      |      2^16      |  2048   |   0.201   |  15.041 us |       4.99% |  15.334 us |       2.08% |  0.292 us |   1.94% |   SAME   |
|      F32      |      I32       |      I32      |      2^20      |  2048   |   0.201   |  18.660 us |       5.52% |  18.918 us |       4.88% |  0.258 us |   1.38% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      |  2048   |   0.201   | 113.394 us |       1.01% | 113.965 us |       0.82% |  0.570 us |   0.50% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      |  2048   |   0.201   |   1.558 ms |       0.40% |   1.558 ms |       0.41% |  0.034 us |   0.00% |   SAME   |
|      F32      |      I32       |      I32      |      2^16      | 2097152 |   0.201   |   1.944 ms |       0.17% |   1.944 ms |       0.17% | -0.132 us |  -0.01% |   SAME   |
|      F32      |      I32       |      I32      |      2^20      | 2097152 |   0.201   |   2.796 ms |       0.10% |   2.796 ms |       0.10% |  0.060 us |   0.00% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      | 2097152 |   0.201   |   3.390 ms |       0.28% |   3.390 ms |       0.27% | -0.037 us |  -0.00% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      | 2097152 |   0.201   |   5.676 ms |       0.27% |   5.676 ms |       0.26% |  0.053 us |   0.00% |   SAME   |
|      F32      |      I32       |      I32      |      2^16      |   32    |     1     |  12.842 us |       7.02% |  13.227 us |       3.31% |  0.385 us |   3.00% |   SAME   |
|      F32      |      I32       |      I32      |      2^20      |   32    |     1     |  13.733 us |       5.08% |  14.176 us |       7.20% |  0.442 us |   3.22% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      |   32    |     1     |  27.278 us |       3.20% |  27.588 us |       1.64% |  0.310 us |   1.14% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      |   32    |     1     | 241.080 us |       0.53% | 241.565 us |       0.47% |  0.485 us |   0.20% |   SAME   |
|      F32      |      I32       |      I32      |      2^16      |   128   |     1     |  12.792 us |       7.22% |  13.272 us |       2.89% |  0.480 us |   3.75% |   SLOW   |
|      F32      |      I32       |      I32      |      2^20      |   128   |     1     |  13.545 us |      11.99% |  13.858 us |      12.41% |  0.313 us |   2.31% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      |   128   |     1     |  27.101 us |       3.59% |  27.560 us |       1.85% |  0.459 us |   1.70% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      |   128   |     1     | 244.274 us |       0.50% | 244.714 us |       0.40% |  0.440 us |   0.18% |   SAME   |
|      F32      |      I32       |      I32      |      2^16      |  2048   |     1     |  16.591 us |       7.00% |  16.803 us |       5.81% |  0.211 us |   1.27% |   SAME   |
|      F32      |      I32       |      I32      |      2^20      |  2048   |     1     |  20.609 us |       4.73% |  21.223 us |       3.40% |  0.614 us |   2.98% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      |  2048   |     1     | 106.823 us |       2.23% | 107.251 us |       1.99% |  0.427 us |   0.40% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      |  2048   |     1     |   1.503 ms |       1.08% |   1.503 ms |       1.14% |  0.425 us |   0.03% |   SAME   |
|      F32      |      I32       |      I32      |      2^16      | 2097152 |     1     |   1.995 ms |       0.15% |   1.995 ms |       0.15% | -0.343 us |  -0.02% |   SAME   |
|      F32      |      I32       |      I32      |      2^20      | 2097152 |     1     |   2.826 ms |       0.12% |   2.826 ms |       0.12% |  0.339 us |   0.01% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      | 2097152 |     1     |   3.970 ms |       0.18% |   3.970 ms |       0.16% |  0.226 us |   0.01% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      | 2097152 |     1     |  15.038 ms |       0.36% |  15.032 ms |       0.37% | -5.796 us |  -0.04% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      |   32    |   0.201   |  12.871 us |       6.52% |  13.266 us |       2.70% |  0.395 us |   3.07% |   SLOW   |
|      F64      |      I32       |      I32      |      2^20      |   32    |   0.201   |  14.004 us |       6.46% |  14.405 us |       7.18% |  0.400 us |   2.86% |   SAME   |
|      F64      |      I32       |      I32      |      2^24      |   32    |   0.201   |  45.519 us |       1.91% |  45.968 us |       1.09% |  0.448 us |   0.98% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      |   32    |   0.201   | 534.795 us |       0.18% | 535.352 us |       0.12% |  0.557 us |   0.10% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      |   128   |   0.201   |  12.582 us |       7.53% |  13.124 us |       4.49% |  0.542 us |   4.30% |   SAME   |
|      F64      |      I32       |      I32      |      2^20      |   128   |   0.201   |  14.515 us |       7.15% |  14.850 us |       6.21% |  0.336 us |   2.31% |   SAME   |
|      F64      |      I32       |      I32      |      2^24      |   128   |   0.201   |  45.403 us |       2.06% |  45.900 us |       1.30% |  0.498 us |   1.10% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      |   128   |   0.201   | 549.653 us |       0.12% | 549.877 us |       0.04% |  0.225 us |   0.04% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      |  2048   |   0.201   |  13.813 us |       7.81% |  14.005 us |       6.96% |  0.191 us |   1.38% |   SAME   |
|      F64      |      I32       |      I32      |      2^20      |  2048   |   0.201   |  18.591 us |       5.46% |  18.814 us |       5.14% |  0.223 us |   1.20% |   SAME   |
|      F64      |      I32       |      I32      |      2^24      |  2048   |   0.201   | 134.300 us |       1.09% | 134.615 us |       0.96% |  0.314 us |   0.23% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      |  2048   |   0.201   |   1.819 ms |       0.13% |   1.819 ms |       0.13% | -0.161 us |  -0.01% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      | 2097152 |   0.201   |   2.601 ms |       0.17% |   2.601 ms |       0.17% |  0.193 us |   0.01% |   SAME   |
|      F64      |      I32       |      I32      |      2^20      | 2097152 |   0.201   |   3.090 ms |       0.18% |   3.090 ms |       0.18% |  0.430 us |   0.01% |   SAME   |
|      F64      |      I32       |      I32      |      2^24      | 2097152 |   0.201   |   3.717 ms |       0.33% |   3.716 ms |       0.31% | -0.811 us |  -0.02% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      | 2097152 |   0.201   |   5.868 ms |       0.30% |   5.867 ms |       0.31% | -0.870 us |  -0.01% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      |   32    |     1     |  12.643 us |       9.08% |  12.889 us |       6.61% |  0.246 us |   1.95% |   SAME   |
|      F64      |      I32       |      I32      |      2^20      |   32    |     1     |  13.093 us |      10.52% |  14.579 us |       7.61% |  1.487 us |  11.36% |   SLOW   |
|      F64      |      I32       |      I32      |      2^24      |   32    |     1     |  37.917 us |       2.87% |  38.233 us |       2.03% |  0.316 us |   0.83% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      |   32    |     1     | 365.450 us |       1.33% | 362.784 us |       0.47% | -2.666 us |  -0.73% |   FAST   |
|      F64      |      I32       |      I32      |      2^16      |   128   |     1     |  12.318 us |       9.44% |  12.880 us |       6.48% |  0.562 us |   4.56% |   SAME   |
|      F64      |      I32       |      I32      |      2^20      |   128   |     1     |  13.821 us |      10.16% |  14.655 us |       6.67% |  0.834 us |   6.03% |   SAME   |
|      F64      |      I32       |      I32      |      2^24      |   128   |     1     |  37.743 us |       2.92% |  38.194 us |       2.00% |  0.451 us |   1.20% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      |   128   |     1     | 368.443 us |       0.71% | 368.424 us |       0.46% | -0.019 us |  -0.01% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      |  2048   |     1     |  15.213 us |       4.10% |  15.354 us |       1.19% |  0.141 us |   0.92% |   SAME   |
|      F64      |      I32       |      I32      |      2^20      |  2048   |     1     |  20.738 us |       4.86% |  21.579 us |       1.72% |  0.841 us |   4.06% |   SLOW   |
|      F64      |      I32       |      I32      |      2^24      |  2048   |     1     | 124.865 us |       1.11% | 124.963 us |       1.11% |  0.098 us |   0.08% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      |  2048   |     1     |   1.620 ms |       0.34% |   1.621 ms |       0.32% |  0.853 us |   0.05% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      | 2097152 |     1     |   2.587 ms |       0.12% |   2.587 ms |       0.11% |  0.321 us |   0.01% |   SAME   |
|      F64      |      I32       |      I32      |      2^20      | 2097152 |     1     |   3.122 ms |       0.15% |   3.123 ms |       0.14% |  0.702 us |   0.02% |   SAME   |
|      F64      |      I32       |      I32      |      2^24      | 2097152 |     1     |   4.316 ms |       0.18% |   4.317 ms |       0.17% |  0.303 us |   0.01% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      | 2097152 |     1     |  15.506 ms |       0.39% |  15.499 ms |       0.38% | -7.227 us |  -0.05% |   SAME   |

@gonidelis
Copy link
Member

preliminary histogram.range results. I don't like that only I8,I32 workloads are getting picked. Why are not the other tunings manifesting? Investigating...

|  SampleT{ct}  |  CounterT{ct}  |  OffsetT{ct}  |  Elements{io}  |  Bins   |  Entropy  |   Ref Time |   Ref Noise |   Cmp Time |   Cmp Noise |       Diff |   %Diff |  Status  |
|---------------|----------------|---------------|----------------|---------|-----------|------------|-------------|------------|-------------|------------|---------|----------|
|      I8       |      I32       |      I32      |      2^16      |   32    |   0.201   |  12.502 us |       8.65% |  12.053 us |      11.17% |  -0.449 us |  -3.59% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      |   32    |   0.201   |  12.391 us |       8.02% |  12.359 us |       7.96% |  -0.033 us |  -0.26% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      |   32    |   0.201   |  19.076 us |       5.49% |  18.299 us |       5.47% |  -0.777 us |  -4.07% |   SAME   |
|      I8       |      I32       |      I32      |      2^28      |   32    |   0.201   | 129.421 us |       0.88% | 122.701 us |       0.92% |  -6.720 us |  -5.19% |   FAST   |
|      I8       |      I32       |      I32      |      2^16      |   128   |   0.201   |  12.837 us |       6.74% |  12.304 us |       8.05% |  -0.533 us |  -4.15% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      |   128   |   0.201   |  12.367 us |       8.28% |  12.389 us |       8.28% |   0.022 us |   0.18% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      |   128   |   0.201   |  19.067 us |       5.60% |  18.478 us |       5.33% |  -0.589 us |  -3.09% |   SAME   |
|      I8       |      I32       |      I32      |      2^28      |   128   |   0.201   | 129.699 us |       0.81% | 122.967 us |       1.03% |  -6.732 us |  -5.19% |   FAST   |
|      I8       |      I32       |      I32      |      2^16      |  2048   |   0.201   |  13.709 us |      10.78% |  13.566 us |      10.93% |  -0.143 us |  -1.04% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      |  2048   |   0.201   |  13.824 us |      12.03% |  13.555 us |      12.01% |  -0.269 us |  -1.94% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      |  2048   |   0.201   |  20.785 us |       5.29% |  18.975 us |       4.45% |  -1.810 us |  -8.71% |   FAST   |
|      I8       |      I32       |      I32      |      2^28      |  2048   |   0.201   | 129.813 us |       0.89% | 123.830 us |       1.08% |  -5.983 us |  -4.61% |   FAST   |
|      I8       |      I32       |      I32      |      2^16      | 2097152 |   0.201   |  21.576 us |       1.78% |  21.520 us |       1.03% |  -0.056 us |  -0.26% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      | 2097152 |   0.201   |  20.901 us |       5.03% |  20.859 us |       4.30% |  -0.042 us |  -0.20% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      | 2097152 |   0.201   |  27.659 us |       1.99% |  26.955 us |       4.04% |  -0.704 us |  -2.55% |   FAST   |
|      I8       |      I32       |      I32      |      2^28      | 2097152 |   0.201   | 134.942 us |       0.75% | 129.629 us |       0.72% |  -5.313 us |  -3.94% |   FAST   |
|      I8       |      I32       |      I32      |      2^16      |   32    |     1     |  12.736 us |       7.25% |  12.365 us |       8.03% |  -0.371 us |  -2.91% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      |   32    |     1     |  12.446 us |       7.42% |  12.422 us |       7.45% |  -0.024 us |  -0.20% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      |   32    |     1     |  18.928 us |       5.95% |  18.405 us |       5.27% |  -0.524 us |  -2.77% |   SAME   |
|      I8       |      I32       |      I32      |      2^28      |   32    |     1     | 129.500 us |       0.81% | 122.836 us |       1.02% |  -6.664 us |  -5.15% |   FAST   |
|      I8       |      I32       |      I32      |      2^16      |   128   |     1     |  12.787 us |       6.89% |  12.378 us |       7.84% |  -0.410 us |  -3.21% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      |   128   |     1     |  12.513 us |       8.21% |  12.446 us |       8.42% |  -0.067 us |  -0.53% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      |   128   |     1     |  19.004 us |       6.05% |  18.429 us |       5.33% |  -0.575 us |  -3.02% |   SAME   |
|      I8       |      I32       |      I32      |      2^28      |   128   |     1     | 129.658 us |       0.88% | 122.985 us |       1.06% |  -6.672 us |  -5.15% |   FAST   |
|      I8       |      I32       |      I32      |      2^16      |  2048   |     1     |  13.516 us |      11.59% |  13.448 us |      11.02% |  -0.068 us |  -0.50% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      |  2048   |     1     |  13.762 us |      12.65% |  13.474 us |      12.46% |  -0.288 us |  -2.10% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      |  2048   |     1     |  20.838 us |       4.75% |  19.332 us |       2.56% |  -1.506 us |  -7.23% |   FAST   |
|      I8       |      I32       |      I32      |      2^28      |  2048   |     1     | 129.686 us |       1.02% | 124.065 us |       1.27% |  -5.621 us |  -4.33% |   FAST   |
|      I8       |      I32       |      I32      |      2^16      | 2097152 |     1     |  21.409 us |       2.76% |  21.351 us |       2.61% |  -0.059 us |  -0.27% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      | 2097152 |     1     |  20.828 us |       5.34% |  20.962 us |       4.68% |   0.134 us |   0.64% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      | 2097152 |     1     |  27.708 us |       0.94% |  27.280 us |       3.16% |  -0.428 us |  -1.55% |   FAST   |
|      I8       |      I32       |      I32      |      2^28      | 2097152 |     1     | 135.160 us |       0.78% | 129.744 us |       0.70% |  -5.416 us |  -4.01% |   FAST   |
|      I16      |      I32       |      I32      |      2^16      |   32    |   0.201   |  17.159 us |       4.05% |  17.149 us |       3.97% |  -0.010 us |  -0.06% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      |   32    |   0.201   |  16.328 us |       7.00% |  16.317 us |       6.97% |  -0.011 us |  -0.07% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      |   32    |   0.201   |  79.874 us |       1.31% |  79.750 us |       1.36% |  -0.124 us |  -0.16% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      |   32    |   0.201   |   1.068 ms |       0.13% |   1.068 ms |       0.12% |  -0.079 us |  -0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^16      |   128   |   0.201   |  19.014 us |       4.55% |  18.999 us |       4.73% |  -0.015 us |  -0.08% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      |   128   |   0.201   |  19.001 us |       7.77% |  18.937 us |       7.65% |  -0.064 us |  -0.34% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      |   128   |   0.201   |  97.009 us |       1.32% |  97.011 us |       1.36% |   0.002 us |   0.00% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      |   128   |   0.201   |   1.344 ms |       0.07% |   1.344 ms |       0.08% |  -0.041 us |  -0.00% |   SAME   |
|      I16      |      I32       |      I32      |      2^16      |  2048   |   0.201   |  27.167 us |       3.11% |  27.202 us |       3.03% |   0.035 us |   0.13% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      |  2048   |   0.201   |  30.799 us |       3.46% |  30.767 us |       3.37% |  -0.033 us |  -0.11% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      |  2048   |   0.201   | 158.406 us |       0.82% | 158.416 us |       0.82% |   0.010 us |   0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      |  2048   |   0.201   |   2.004 ms |       0.14% |   2.004 ms |       0.14% |   0.235 us |   0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^16      | 2097152 |   0.201   | 570.782 us |       0.16% | 570.925 us |       0.19% |   0.143 us |   0.03% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      | 2097152 |   0.201   |   1.176 ms |       0.14% |   1.176 ms |       0.15% |  -0.141 us |  -0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      | 2097152 |   0.201   |   1.629 ms |       0.26% |   1.629 ms |       0.26% |  -0.088 us |  -0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      | 2097152 |   0.201   |   4.499 ms |       0.08% |   4.499 ms |       0.09% |  -0.280 us |  -0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^16      |   32    |     1     |  17.022 us |       4.58% |  17.027 us |       4.53% |   0.005 us |   0.03% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      |   32    |     1     |  16.679 us |       7.67% |  16.579 us |       7.69% |  -0.100 us |  -0.60% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      |   32    |     1     |  79.939 us |       1.23% |  79.934 us |       1.23% |  -0.005 us |  -0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      |   32    |     1     |   1.071 ms |       0.12% |   1.071 ms |       0.12% |  -0.069 us |  -0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^16      |   128   |     1     |  18.942 us |       4.85% |  19.071 us |       4.25% |   0.128 us |   0.68% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      |   128   |     1     |  18.624 us |       5.54% |  18.549 us |       5.63% |  -0.075 us |  -0.40% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      |   128   |     1     |  92.583 us |       1.00% |  92.625 us |       0.98% |   0.043 us |   0.05% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      |   128   |     1     |   1.274 ms |       0.11% |   1.274 ms |       0.11% |  -0.014 us |  -0.00% |   SAME   |
|      I16      |      I32       |      I32      |      2^16      |  2048   |     1     |  33.240 us |       2.73% |  33.152 us |       2.82% |  -0.088 us |  -0.26% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      |  2048   |     1     |  35.080 us |       2.94% |  35.049 us |       3.01% |  -0.032 us |  -0.09% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      |  2048   |     1     | 242.108 us |       0.59% | 242.105 us |       0.57% |  -0.003 us |  -0.00% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      |  2048   |     1     |   3.376 ms |       0.05% |   3.376 ms |       0.06% |  -0.063 us |  -0.00% |   SAME   |
|      I16      |      I32       |      I32      |      2^16      | 2097152 |     1     | 570.924 us |       0.17% | 570.936 us |       0.16% |   0.012 us |   0.00% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      | 2097152 |     1     |   1.176 ms |       0.15% |   1.176 ms |       0.14% |   0.063 us |   0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      | 2097152 |     1     |   1.629 ms |       0.24% |   1.629 ms |       0.24% |  -0.159 us |  -0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      | 2097152 |     1     |   4.499 ms |       0.08% |   4.499 ms |       0.09% |   0.219 us |   0.00% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      |   32    |   0.201   |  19.039 us |       4.27% |  19.067 us |       4.11% |   0.028 us |   0.15% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      |   32    |   0.201   |  20.315 us |       4.76% |  20.222 us |       4.70% |  -0.094 us |  -0.46% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      |   32    |   0.201   |  84.417 us |       1.08% |  84.393 us |       1.11% |  -0.024 us |  -0.03% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      |   32    |   0.201   |   1.175 ms |       0.11% |   1.175 ms |       0.12% |  -0.137 us |  -0.01% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      |   128   |   0.201   |  20.902 us |       4.46% |  20.894 us |       4.41% |  -0.007 us |  -0.04% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      |   128   |   0.201   |  22.380 us |       4.50% |  22.344 us |       4.38% |  -0.036 us |  -0.16% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      |   128   |   0.201   | 104.910 us |       0.88% | 104.842 us |       0.89% |  -0.068 us |  -0.06% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      |   128   |   0.201   |   1.492 ms |       0.10% |   1.492 ms |       0.10% |  -0.431 us |  -0.03% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      |  2048   |   0.201   |  31.285 us |       2.76% |  30.828 us |       3.88% |  -0.458 us |  -1.46% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      |  2048   |   0.201   |  35.076 us |       3.45% |  34.575 us |       4.11% |  -0.502 us |  -1.43% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      |  2048   |   0.201   | 167.972 us |       0.69% | 167.523 us |       0.84% |  -0.449 us |  -0.27% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      |  2048   |   0.201   |   2.187 ms |       0.15% |   2.187 ms |       0.15% |  -0.268 us |  -0.01% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      | 2097152 |   0.201   |   2.065 ms |       0.25% |   2.065 ms |       0.22% |   0.161 us |   0.01% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      | 2097152 |   0.201   |   2.836 ms |       0.12% |   2.836 ms |       0.12% |   0.241 us |   0.01% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      | 2097152 |   0.201   |   3.720 ms |       0.19% |   3.720 ms |       0.18% |  -0.511 us |  -0.01% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      | 2097152 |   0.201   |   8.278 ms |       0.11% |   8.279 ms |       0.11% |   0.743 us |   0.01% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      |   32    |     1     |  19.197 us |       4.30% |  19.160 us |       4.29% |  -0.037 us |  -0.19% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      |   32    |     1     |  20.446 us |       4.86% |  20.427 us |       4.78% |  -0.018 us |  -0.09% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      |   32    |     1     |  84.640 us |       1.56% |  84.609 us |       1.43% |  -0.032 us |  -0.04% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      |   32    |     1     |   1.179 ms |       0.12% |   1.179 ms |       0.12% |  -0.182 us |  -0.02% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      |   128   |     1     |  21.012 us |       4.24% |  20.953 us |       4.25% |  -0.059 us |  -0.28% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      |   128   |     1     |  22.380 us |       4.29% |  22.376 us |       4.30% |  -0.004 us |  -0.02% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      |   128   |     1     | 100.486 us |       1.22% | 100.444 us |       1.24% |  -0.042 us |  -0.04% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      |   128   |     1     |   1.410 ms |       0.10% |   1.410 ms |       0.09% |   0.024 us |   0.00% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      |  2048   |     1     |  31.273 us |       2.80% |  31.223 us |       2.85% |  -0.050 us |  -0.16% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      |  2048   |     1     |  37.136 us |       2.49% |  37.150 us |       2.43% |   0.014 us |   0.04% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      |  2048   |     1     | 220.468 us |       0.65% | 220.612 us |       0.63% |   0.144 us |   0.07% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      |  2048   |     1     |   2.949 ms |       0.07% |   2.949 ms |       0.07% |   0.135 us |   0.00% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      | 2097152 |     1     |   2.371 ms |       0.17% |   2.371 ms |       0.17% |  -0.024 us |  -0.00% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      | 2097152 |     1     |   2.917 ms |       0.12% |   2.917 ms |       0.12% |  -0.294 us |  -0.01% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      | 2097152 |     1     |   4.529 ms |       0.79% |   4.530 ms |       0.77% |   0.636 us |   0.01% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      | 2097152 |     1     |  18.165 ms |       0.86% |  18.160 ms |       0.86% |  -5.504 us |  -0.03% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      |   32    |   0.201   |  14.892 us |       7.69% |  14.821 us |       7.50% |  -0.072 us |  -0.48% |   SAME   |
|      I64      |      I32       |      I32      |      2^20      |   32    |   0.201   |  18.903 us |       7.33% |  18.897 us |       6.97% |  -0.006 us |  -0.03% |   SAME   |
|      I64      |      I32       |      I32      |      2^24      |   32    |   0.201   |  98.532 us |       1.22% |  98.555 us |       1.16% |   0.023 us |   0.02% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      |   32    |   0.201   |   1.311 ms |       0.18% |   1.311 ms |       0.18% |   0.120 us |   0.01% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      |   128   |   0.201   |  17.918 us |       6.06% |  17.904 us |       5.74% |  -0.014 us |  -0.08% |   SAME   |
|      I64      |      I32       |      I32      |      2^20      |   128   |   0.201   |  21.294 us |       6.48% |  21.244 us |       5.77% |  -0.050 us |  -0.24% |   SAME   |
|      I64      |      I32       |      I32      |      2^24      |   128   |   0.201   | 124.638 us |       0.98% | 124.618 us |       1.06% |  -0.020 us |  -0.02% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      |   128   |   0.201   |   1.707 ms |       0.14% |   1.707 ms |       0.14% |   0.271 us |   0.02% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      |  2048   |   0.201   |  24.454 us |       5.50% |  24.377 us |       5.80% |  -0.076 us |  -0.31% |   SAME   |
|      I64      |      I32       |      I32      |      2^20      |  2048   |   0.201   |  29.976 us |       5.09% |  29.898 us |       4.94% |  -0.077 us |  -0.26% |   SAME   |
|      I64      |      I32       |      I32      |      2^24      |  2048   |   0.201   | 178.603 us |       0.66% | 178.440 us |       0.68% |  -0.162 us |  -0.09% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      |  2048   |   0.201   |   2.503 ms |       0.09% |   2.503 ms |       0.09% |  -0.064 us |  -0.00% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      | 2097152 |   0.201   |   2.637 ms |       0.17% |   2.637 ms |       0.17% |   0.017 us |   0.00% |   SAME   |
|      I64      |      I32       |      I32      |      2^20      | 2097152 |   0.201   |   3.097 ms |       0.14% |   3.096 ms |       0.13% |  -0.281 us |  -0.01% |   SAME   |
|      I64      |      I32       |      I32      |      2^24      | 2097152 |   0.201   |   3.555 ms |       0.15% |   3.555 ms |       0.17% |   0.200 us |   0.01% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      | 2097152 |   0.201   |   9.379 ms |       0.09% |   9.378 ms |       0.09% |  -1.149 us |  -0.01% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      |   32    |     1     |  16.290 us |       8.74% |  16.196 us |       8.98% |  -0.094 us |  -0.58% |   SAME   |
|      I64      |      I32       |      I32      |      2^20      |   32    |     1     |  18.810 us |       6.78% |  18.957 us |       7.00% |   0.147 us |   0.78% |   SAME   |
|      I64      |      I32       |      I32      |      2^24      |   32    |     1     |  95.562 us |       1.58% |  95.603 us |       1.56% |   0.041 us |   0.04% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      |   32    |     1     |   1.271 ms |       0.16% |   1.271 ms |       0.15% |  -0.138 us |  -0.01% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      |   128   |     1     |  16.876 us |       5.40% |  16.901 us |       5.53% |   0.025 us |   0.15% |   SAME   |
|      I64      |      I32       |      I32      |      2^20      |   128   |     1     |  20.880 us |       6.71% |  20.815 us |       6.84% |  -0.065 us |  -0.31% |   SAME   |
|      I64      |      I32       |      I32      |      2^24      |   128   |     1     | 121.726 us |       0.80% | 121.571 us |       0.82% |  -0.155 us |  -0.13% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      |   128   |     1     |   1.633 ms |       0.14% |   1.633 ms |       0.15% |   0.460 us |   0.03% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      |  2048   |     1     |  25.441 us |       2.27% |  25.457 us |       2.31% |   0.016 us |   0.06% |   SAME   |
|      I64      |      I32       |      I32      |      2^20      |  2048   |     1     |  37.061 us |       2.68% |  37.124 us |       2.73% |   0.062 us |   0.17% |   SAME   |
|      I64      |      I32       |      I32      |      2^24      |  2048   |     1     | 280.130 us |       0.37% | 280.034 us |       0.41% |  -0.096 us |  -0.03% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      |  2048   |     1     |   4.037 ms |       0.05% |   4.037 ms |       0.05% |  -0.142 us |  -0.00% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      | 2097152 |     1     |   2.646 ms |       0.12% |   2.646 ms |       0.12% |   0.369 us |   0.01% |   SAME   |
|      I64      |      I32       |      I32      |      2^20      | 2097152 |     1     |   3.182 ms |       0.22% |   3.182 ms |       0.21% |  -0.076 us |  -0.00% |   SAME   |
|      I64      |      I32       |      I32      |      2^24      | 2097152 |     1     |   4.351 ms |       0.63% |   4.349 ms |       0.59% |  -1.822 us |  -0.04% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      | 2097152 |     1     |  19.152 ms |       0.53% |  19.156 ms |       0.55% |   3.984 us |   0.02% |   SAME   |
|      F32      |      I32       |      I32      |      2^16      |   32    |   0.201   |  18.913 us |       4.71% |  18.941 us |       4.55% |   0.028 us |   0.15% |   SAME   |
|      F32      |      I32       |      I32      |      2^20      |   32    |   0.201   |  20.064 us |       4.05% |  20.031 us |       4.04% |  -0.033 us |  -0.16% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      |   32    |   0.201   |  86.525 us |       1.02% |  86.704 us |       0.84% |   0.178 us |   0.21% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      |   32    |   0.201   |   1.190 ms |       0.16% |   1.190 ms |       0.16% |   0.061 us |   0.01% |   SAME   |
|      F32      |      I32       |      I32      |      2^16      |   128   |   0.201   |  21.237 us |       3.33% |  21.226 us |       3.44% |  -0.011 us |  -0.05% |   SAME   |
|      F32      |      I32       |      I32      |      2^20      |   128   |   0.201   |  22.167 us |       4.03% |  22.239 us |       4.03% |   0.072 us |   0.33% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      |   128   |   0.201   | 105.136 us |       0.73% | 105.196 us |       0.70% |   0.060 us |   0.06% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      |   128   |   0.201   |   1.499 ms |       0.12% |   1.499 ms |       0.11% |   0.047 us |   0.00% |   SAME   |
|      F32      |      I32       |      I32      |      2^16      |  2048   |   0.201   |  30.131 us |       3.60% |  30.105 us |       3.62% |  -0.026 us |  -0.09% |   SAME   |
|      F32      |      I32       |      I32      |      2^20      |  2048   |   0.201   |  35.069 us |       2.99% |  35.053 us |       3.09% |  -0.015 us |  -0.04% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      |  2048   |   0.201   | 175.951 us |       1.01% | 175.508 us |       0.96% |  -0.443 us |  -0.25% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      |  2048   |   0.201   |   2.120 ms |       0.16% |   2.120 ms |       0.16% |  -0.290 us |  -0.01% |   SAME   |
|      F32      |      I32       |      I32      |      2^16      | 2097152 |   0.201   |   2.058 ms |       0.27% |   2.059 ms |       0.26% |   0.631 us |   0.03% |   SAME   |
|      F32      |      I32       |      I32      |      2^20      | 2097152 |   0.201   |   2.831 ms |       0.13% |   2.830 ms |       0.14% |  -0.332 us |  -0.01% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      | 2097152 |   0.201   |   3.729 ms |       0.22% |   3.729 ms |       0.21% |   0.233 us |   0.01% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      | 2097152 |   0.201   |   7.923 ms |       0.12% |   7.922 ms |       0.11% |  -0.882 us |  -0.01% |   SAME   |
|      F32      |      I32       |      I32      |      2^16      |   32    |     1     |  19.056 us |       4.38% |  19.077 us |       4.47% |   0.021 us |   0.11% |   SAME   |
|      F32      |      I32       |      I32      |      2^20      |   32    |     1     |  20.037 us |       4.00% |  19.840 us |       3.52% |  -0.196 us |  -0.98% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      |   32    |     1     |  84.558 us |       1.12% |  84.529 us |       1.10% |  -0.029 us |  -0.03% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      |   32    |     1     |   1.174 ms |       0.10% |   1.174 ms |       0.10% |  -0.077 us |  -0.01% |   SAME   |
|      F32      |      I32       |      I32      |      2^16      |   128   |     1     |  20.979 us |       4.38% |  20.917 us |       4.60% |  -0.063 us |  -0.30% |   SAME   |
|      F32      |      I32       |      I32      |      2^20      |   128   |     1     |  22.331 us |       4.40% |  22.346 us |       4.28% |   0.015 us |   0.07% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      |   128   |     1     | 100.883 us |       0.90% | 100.870 us |       0.93% |  -0.013 us |  -0.01% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      |   128   |     1     |   1.416 ms |       0.10% |   1.416 ms |       0.10% |   0.098 us |   0.01% |   SAME   |
|      F32      |      I32       |      I32      |      2^16      |  2048   |     1     |  31.199 us |       2.94% |  31.233 us |       3.00% |   0.034 us |   0.11% |   SAME   |
|      F32      |      I32       |      I32      |      2^20      |  2048   |     1     |  37.071 us |       2.72% |  37.007 us |       2.70% |  -0.064 us |  -0.17% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      |  2048   |     1     | 220.879 us |       0.67% | 220.882 us |       0.67% |   0.003 us |   0.00% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      |  2048   |     1     |   2.948 ms |       0.07% |   2.948 ms |       0.07% |   0.050 us |   0.00% |   SAME   |
|      F32      |      I32       |      I32      |      2^16      | 2097152 |     1     |   2.370 ms |       0.18% |   2.370 ms |       0.18% |  -0.131 us |  -0.01% |   SAME   |
|      F32      |      I32       |      I32      |      2^20      | 2097152 |     1     |   2.916 ms |       0.11% |   2.915 ms |       0.12% |  -0.355 us |  -0.01% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      | 2097152 |     1     |   4.529 ms |       0.79% |   4.528 ms |       0.78% |  -1.052 us |  -0.02% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      | 2097152 |     1     |  18.159 ms |       0.83% |  18.155 ms |       0.81% |  -4.142 us |  -0.02% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      |   32    |   0.201   |  14.944 us |       5.46% |  15.000 us |       5.15% |   0.056 us |   0.37% |   SAME   |
|      F64      |      I32       |      I32      |      2^20      |   32    |   0.201   |  18.154 us |       4.99% |  18.192 us |       5.19% |   0.038 us |   0.21% |   SAME   |
|      F64      |      I32       |      I32      |      2^24      |   32    |   0.201   |  92.915 us |       1.28% |  92.935 us |       1.25% |   0.020 us |   0.02% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      |   32    |   0.201   |   1.245 ms |       0.13% |   1.245 ms |       0.13% |  -0.148 us |  -0.01% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      |   128   |   0.201   |  16.718 us |       5.68% |  16.704 us |       5.62% |  -0.013 us |  -0.08% |   SAME   |
|      F64      |      I32       |      I32      |      2^20      |   128   |   0.201   |  20.311 us |       5.25% |  20.338 us |       5.29% |   0.027 us |   0.13% |   SAME   |
|      F64      |      I32       |      I32      |      2^24      |   128   |   0.201   | 115.367 us |       1.10% | 115.235 us |       1.03% |  -0.132 us |  -0.11% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      |   128   |   0.201   |   1.612 ms |       0.10% |   1.612 ms |       0.11% |   0.003 us |   0.00% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      |  2048   |   0.201   |  23.343 us |       3.40% |  23.380 us |       3.75% |   0.037 us |   0.16% |   SAME   |
|      F64      |      I32       |      I32      |      2^20      |  2048   |   0.201   |  30.697 us |       4.32% |  30.724 us |       4.13% |   0.027 us |   0.09% |   SAME   |
|      F64      |      I32       |      I32      |      2^24      |  2048   |   0.201   | 190.723 us |       0.59% | 190.683 us |       0.64% |  -0.040 us |  -0.02% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      |  2048   |   0.201   |   2.386 ms |       0.09% |   2.386 ms |       0.10% |   0.487 us |   0.02% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      | 2097152 |   0.201   |   2.633 ms |       0.17% |   2.633 ms |       0.17% |  -0.063 us |  -0.00% |   SAME   |
|      F64      |      I32       |      I32      |      2^20      | 2097152 |   0.201   |   3.091 ms |       0.17% |   3.092 ms |       0.15% |   0.560 us |   0.02% |   SAME   |
|      F64      |      I32       |      I32      |      2^24      | 2097152 |   0.201   |   3.811 ms |       0.19% |   3.810 ms |       0.20% |  -0.849 us |  -0.02% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      | 2097152 |   0.201   |   9.448 ms |       0.10% |   9.448 ms |       0.10% |  -0.188 us |  -0.00% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      |   32    |     1     |  14.880 us |       6.37% |  15.380 us |       1.97% |   0.500 us |   3.36% |   SLOW   |
|      F64      |      I32       |      I32      |      2^20      |   32    |     1     |  17.890 us |       4.19% |  18.589 us |       5.44% |   0.698 us |   3.90% |   SAME   |
|      F64      |      I32       |      I32      |      2^24      |   32    |     1     |  90.178 us |       1.20% |  90.798 us |       0.88% |   0.621 us |   0.69% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      |   32    |     1     |   1.205 ms |       0.14% |   1.206 ms |       0.12% |   0.431 us |   0.04% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      |   128   |     1     |  16.706 us |       5.51% |  17.423 us |       0.86% |   0.717 us |   4.29% |   SLOW   |
|      F64      |      I32       |      I32      |      2^20      |   128   |     1     |  20.016 us |       4.02% |  20.773 us |       4.81% |   0.756 us |   3.78% |   SAME   |
|      F64      |      I32       |      I32      |      2^24      |   128   |     1     | 111.360 us |       0.95% | 111.826 us |       0.57% |   0.466 us |   0.42% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      |   128   |     1     |   1.509 ms |       0.14% |   1.509 ms |       0.13% |   0.452 us |   0.03% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      |  2048   |     1     |  25.473 us |       2.08% |  25.623 us |       0.70% |   0.150 us |   0.59% |   SAME   |
|      F64      |      I32       |      I32      |      2^20      |  2048   |     1     |  37.132 us |       2.66% |  37.529 us |       2.10% |   0.397 us |   1.07% |   SAME   |
|      F64      |      I32       |      I32      |      2^24      |  2048   |     1     | 281.453 us |       0.30% | 281.748 us |       0.19% |   0.295 us |   0.10% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      |  2048   |     1     |   4.062 ms |       0.04% |   4.061 ms |       0.03% |  -0.276 us |  -0.01% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      | 2097152 |     1     |   2.644 ms |       0.13% |   2.644 ms |       0.11% |   0.065 us |   0.00% |   SAME   |
|      F64      |      I32       |      I32      |      2^20      | 2097152 |     1     |   3.179 ms |       0.22% |   3.180 ms |       0.22% |   0.591 us |   0.02% |   SAME   |
|      F64      |      I32       |      I32      |      2^24      | 2097152 |     1     |   4.693 ms |       0.59% |   4.692 ms |       0.60% |  -0.938 us |  -0.02% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      | 2097152 |     1     |  19.438 ms |       1.08% |  19.428 ms |       0.97% | -10.026 us |  -0.05% |   SAME   |

@bernhardmgruber
Copy link
Contributor Author

histogram.even (tbh I consider it great and exactly what we expected. The sole provided tuning does great for 2^28 problem sizes). We should maybe take a second look on the scarce SLOWs in the float,double land.

From looking at the code, we only provide a tuning for I8 for even and those numbers look good. I consider the regressions on float and double noise, since we fall back to Policy900 for that.

@bernhardmgruber
Copy link
Contributor Author

preliminary histogram.range results. I don't like that only I8,I32 workloads are getting picked. Why are not the other tunings manifesting? Investigating...

From the benchmark results it seems that also only the I8 tuning has an impact. The other tunings barely change the outcome. I would remove all tunings but the one for I8.

@bernhardmgruber bernhardmgruber force-pushed the fix_histo_tunings branch 2 times, most recently from 7623d12 to 2699952 Compare February 6, 2025 09:13
Copy link
Contributor

github-actions bot commented Feb 6, 2025

🟩 CI finished in 1h 38m: Pass: 100%/90 | Total: 2d 19h | Avg: 44m 40s | Max: 1h 19m | Hits: 56%/132225
  • 🟩 cub: Pass: 100%/44 | Total: 1d 18h | Avg: 58m 00s | Max: 1h 19m | Hits: 29%/52320

    🟩 cpu
      🟩 amd64              Pass: 100%/42  | Total:  1d 16h | Avg: 57m 39s | Max:  1h 19m | Hits:  30%/49888 
      🟩 arm64              Pass: 100%/2   | Total:  2h 10m | Avg:  1h 05m | Max:  1h 06m | Hits:  16%/2432  
    🟩 ctk
      🟩 12.0               Pass: 100%/5   | Total:  5h 13m | Avg:  1h 02m | Max:  1h 07m | Hits:  15%/5914  
      🟩 12.5               Pass: 100%/2   | Total:  2h 26m | Avg:  1h 13m | Max:  1h 15m | Hits:  12%/2250  
      🟩 12.8               Pass: 100%/37  | Total:  1d 10h | Avg: 56m 33s | Max:  1h 19m | Hits:  32%/44156 
    🟩 cudacxx
      🟩 ClangCUDA18        Pass: 100%/2   | Total:  2h 08m | Avg:  1h 04m | Max:  1h 06m | Hits:  15%/2104  
      🟩 nvcc12.0           Pass: 100%/5   | Total:  5h 13m | Avg:  1h 02m | Max:  1h 07m | Hits:  15%/5914  
      🟩 nvcc12.5           Pass: 100%/2   | Total:  2h 26m | Avg:  1h 13m | Max:  1h 15m | Hits:  12%/2250  
      🟩 nvcc12.8           Pass: 100%/35  | Total:  1d 08h | Avg: 56m 07s | Max:  1h 19m | Hits:  33%/42052 
    🟩 cudacxx_family
      🟩 ClangCUDA          Pass: 100%/2   | Total:  2h 08m | Avg:  1h 04m | Max:  1h 06m | Hits:  15%/2104  
      🟩 nvcc               Pass: 100%/42  | Total:  1d 16h | Avg: 57m 42s | Max:  1h 19m | Hits:  30%/50216 
    🟩 cxx
      🟩 Clang14            Pass: 100%/4   | Total:  4h 03m | Avg:  1h 00m | Max:  1h 02m | Hits:  16%/4872  
      🟩 Clang15            Pass: 100%/2   | Total:  2h 10m | Avg:  1h 05m | Max:  1h 07m | Hits:  16%/2432  
      🟩 Clang16            Pass: 100%/2   | Total:  2h 06m | Avg:  1h 03m | Max:  1h 05m | Hits:  16%/2432  
      🟩 Clang17            Pass: 100%/2   | Total:  2h 07m | Avg:  1h 03m | Max:  1h 04m | Hits:  16%/2432  
      🟩 Clang18            Pass: 100%/7   | Total:  5h 55m | Avg: 50m 42s | Max:  1h 06m | Hits:  41%/8184  
      🟩 GCC7               Pass: 100%/2   | Total:  1h 57m | Avg: 58m 47s | Max: 58m 49s | Hits:  16%/2436  
      🟩 GCC8               Pass: 100%/1   | Total:  1h 02m | Avg:  1h 02m | Max:  1h 02m | Hits:  16%/1218  
      🟩 GCC9               Pass: 100%/2   | Total:  2h 12m | Avg:  1h 06m | Max:  1h 06m | Hits:  16%/2436  
      🟩 GCC10              Pass: 100%/2   | Total:  2h 03m | Avg:  1h 01m | Max:  1h 01m | Hits:  16%/2436  
      🟩 GCC11              Pass: 100%/2   | Total:  2h 13m | Avg:  1h 06m | Max:  1h 07m | Hits:  16%/2432  
      🟩 GCC12              Pass: 100%/2   | Total:  2h 14m | Avg:  1h 07m | Max:  1h 07m | Hits:  16%/2432  
      🟩 GCC13              Pass: 100%/10  | Total:  6h 58m | Avg: 41m 50s | Max:  1h 18m | Hits:  57%/12160 
      🟩 MSVC14.29          Pass: 100%/2   | Total:  2h 21m | Avg:  1h 10m | Max:  1h 14m | Hits:  12%/2084  
      🟩 MSVC14.42          Pass: 100%/2   | Total:  2h 38m | Avg:  1h 19m | Max:  1h 19m | Hits:  12%/2084  
      🟩 NVHPC24.7          Pass: 100%/2   | Total:  2h 26m | Avg:  1h 13m | Max:  1h 15m | Hits:  12%/2250  
    🟩 cxx_family
      🟩 Clang              Pass: 100%/17  | Total: 16h 22m | Avg: 57m 46s | Max:  1h 07m | Hits:  26%/20352 
      🟩 GCC                Pass: 100%/21  | Total: 18h 43m | Avg: 53m 29s | Max:  1h 18m | Hits:  36%/25550 
      🟩 MSVC               Pass: 100%/4   | Total:  5h 00m | Avg:  1h 15m | Max:  1h 19m | Hits:  12%/4168  
      🟩 NVHPC              Pass: 100%/2   | Total:  2h 26m | Avg:  1h 13m | Max:  1h 15m | Hits:  12%/2250  
    🟩 gpu
      🟩 h100               Pass: 100%/2   | Total: 56m 12s | Avg: 28m 06s | Max: 30m 18s | Hits:  57%/2432  
      🟩 rtx2080            Pass: 100%/34  | Total:  1d 13h | Avg:  1h 06m | Max:  1h 19m | Hits:  15%/40160 
      🟩 rtxa6000           Pass: 100%/8   | Total:  4h 12m | Avg: 31m 30s | Max:  1h 05m | Hits:  78%/9728  
    🟩 jobs
      🟩 Build              Pass: 100%/37  | Total:  1d 16h | Avg:  1h 04m | Max:  1h 19m | Hits:  15%/43808 
      🟩 DeviceLaunch       Pass: 100%/1   | Total: 21m 04s | Avg: 21m 04s | Max: 21m 04s | Hits:  99%/1216  
      🟩 GraphCapture       Pass: 100%/1   | Total: 16m 44s | Avg: 16m 44s | Max: 16m 44s | Hits:  99%/1216  
      🟩 HostLaunch         Pass: 100%/3   | Total:  1h 12m | Avg: 24m 19s | Max: 25m 54s | Hits:  99%/3648  
      🟩 TestGPU            Pass: 100%/2   | Total: 41m 15s | Avg: 20m 37s | Max: 22m 02s | Hits:  99%/2432  
    🟩 sm
      🟩 90                 Pass: 100%/2   | Total: 56m 12s | Avg: 28m 06s | Max: 30m 18s | Hits:  57%/2432  
      🟩 90;90a;100         Pass: 100%/1   | Total:  1h 18m | Avg:  1h 18m | Max:  1h 18m | Hits:  16%/1216  
    🟩 std
      🟩 17                 Pass: 100%/20  | Total: 21h 52m | Avg:  1h 05m | Max:  1h 19m | Hits:  15%/23559 
      🟩 20                 Pass: 100%/24  | Total: 20h 40m | Avg: 51m 40s | Max:  1h 18m | Hits:  40%/28761 
    
  • 🟩 thrust: Pass: 100%/43 | Total: 23h 47m | Avg: 33m 11s | Max: 1h 02m | Hits: 74%/79625

    🟩 cmake_options
      🟩 -DTHRUST_DISPATCH_TYPE=Force32bit Pass: 100%/2   | Total: 41m 07s | Avg: 20m 33s | Max: 29m 42s | Hits:  87%/3706  
    🟩 cpu
      🟩 amd64              Pass: 100%/41  | Total: 22h 42m | Avg: 33m 14s | Max:  1h 02m | Hits:  74%/75920 
      🟩 arm64              Pass: 100%/2   | Total:  1h 04m | Avg: 32m 07s | Max: 34m 46s | Hits:  74%/3705  
    🟩 ctk
      🟩 12.0               Pass: 100%/5   | Total:  3h 00m | Avg: 36m 04s | Max: 53m 52s | Hits:  70%/9256  
      🟩 12.5               Pass: 100%/2   | Total:  1h 53m | Avg: 56m 46s | Max: 58m 16s | Hits:  56%/3704  
      🟩 12.8               Pass: 100%/36  | Total: 18h 53m | Avg: 31m 28s | Max:  1h 02m | Hits:  76%/66665 
    🟩 cudacxx
      🟩 ClangCUDA18        Pass: 100%/2   | Total: 56m 25s | Avg: 28m 12s | Max: 28m 30s | Hits:  74%/3704  
      🟩 nvcc12.0           Pass: 100%/5   | Total:  3h 00m | Avg: 36m 04s | Max: 53m 52s | Hits:  70%/9256  
      🟩 nvcc12.5           Pass: 100%/2   | Total:  1h 53m | Avg: 56m 46s | Max: 58m 16s | Hits:  56%/3704  
      🟩 nvcc12.8           Pass: 100%/34  | Total: 17h 56m | Avg: 31m 40s | Max:  1h 02m | Hits:  76%/62961 
    🟩 cudacxx_family
      🟩 ClangCUDA          Pass: 100%/2   | Total: 56m 25s | Avg: 28m 12s | Max: 28m 30s | Hits:  74%/3704  
      🟩 nvcc               Pass: 100%/41  | Total: 22h 50m | Avg: 33m 26s | Max:  1h 02m | Hits:  74%/75921 
    🟩 cxx
      🟩 Clang14            Pass: 100%/4   | Total:  2h 03m | Avg: 30m 45s | Max: 31m 46s | Hits:  74%/7408  
      🟩 Clang15            Pass: 100%/2   | Total:  1h 05m | Avg: 32m 40s | Max: 33m 38s | Hits:  74%/3704  
      🟩 Clang16            Pass: 100%/2   | Total:  1h 06m | Avg: 33m 21s | Max: 34m 47s | Hits:  74%/3704  
      🟩 Clang17            Pass: 100%/2   | Total:  1h 05m | Avg: 32m 34s | Max: 33m 26s | Hits:  74%/3704  
      🟩 Clang18            Pass: 100%/7   | Total:  2h 46m | Avg: 23m 45s | Max: 33m 05s | Hits:  81%/12964 
      🟩 GCC7               Pass: 100%/2   | Total:  1h 06m | Avg: 33m 08s | Max: 33m 45s | Hits:  74%/3706  
      🟩 GCC8               Pass: 100%/1   | Total: 31m 46s | Avg: 31m 46s | Max: 31m 46s | Hits:  74%/1853  
      🟩 GCC9               Pass: 100%/2   | Total:  1h 05m | Avg: 32m 49s | Max: 32m 59s | Hits:  74%/3706  
      🟩 GCC10              Pass: 100%/2   | Total:  1h 08m | Avg: 34m 18s | Max: 35m 45s | Hits:  74%/3706  
      🟩 GCC11              Pass: 100%/2   | Total:  1h 07m | Avg: 33m 42s | Max: 36m 12s | Hits:  74%/3706  
      🟩 GCC12              Pass: 100%/2   | Total:  1h 07m | Avg: 33m 58s | Max: 35m 54s | Hits:  74%/3706  
      🟩 GCC13              Pass: 100%/8   | Total:  3h 17m | Avg: 24m 43s | Max: 35m 01s | Hits:  83%/14824 
      🟩 MSVC14.29          Pass: 100%/2   | Total:  1h 47m | Avg: 53m 41s | Max: 53m 52s | Hits:  53%/3692  
      🟩 MSVC14.42          Pass: 100%/3   | Total:  2h 34m | Avg: 51m 28s | Max:  1h 02m | Hits:  58%/5538  
      🟩 NVHPC24.7          Pass: 100%/2   | Total:  1h 53m | Avg: 56m 46s | Max: 58m 16s | Hits:  56%/3704  
    🟩 cxx_family
      🟩 Clang              Pass: 100%/17  | Total:  8h 06m | Avg: 28m 37s | Max: 34m 47s | Hits:  77%/31484 
      🟩 GCC                Pass: 100%/19  | Total:  9h 25m | Avg: 29m 45s | Max: 36m 12s | Hits:  78%/35207 
      🟩 MSVC               Pass: 100%/5   | Total:  4h 21m | Avg: 52m 21s | Max:  1h 02m | Hits:  56%/9230  
      🟩 NVHPC              Pass: 100%/2   | Total:  1h 53m | Avg: 56m 46s | Max: 58m 16s | Hits:  56%/3704  
    🟩 gpu
      🟩 rtx2080            Pass: 100%/33  | Total: 19h 46m | Avg: 35m 57s | Max:  1h 00m | Hits:  71%/61112 
      🟩 rtx4090            Pass: 100%/10  | Total:  4h 00m | Avg: 24m 04s | Max:  1h 02m | Hits:  84%/18513 
    🟩 jobs
      🟩 Build              Pass: 100%/37  | Total: 22h 26m | Avg: 36m 24s | Max:  1h 02m | Hits:  71%/68516 
      🟩 TestCPU            Pass: 100%/3   | Total: 46m 44s | Avg: 15m 34s | Max: 30m 55s | Hits:  89%/5551  
      🟩 TestGPU            Pass: 100%/3   | Total: 33m 38s | Avg: 11m 12s | Max: 11m 44s | Hits:  99%/5558  
    🟩 sm
      🟩 90;90a;100         Pass: 100%/1   | Total: 35m 01s | Avg: 35m 01s | Max: 35m 01s | Hits:  74%/1853  
    🟩 std
      🟩 17                 Pass: 100%/20  | Total: 12h 27m | Avg: 37m 21s | Max:  1h 00m | Hits:  70%/37031 
      🟩 20                 Pass: 100%/21  | Total: 10h 38m | Avg: 30m 25s | Max:  1h 02m | Hits:  77%/38888 
    
  • 🟩 cccl_c_parallel: Pass: 100%/2 | Total: 8m 13s | Avg: 4m 06s | Max: 5m 43s | Hits: 97%/280

    🟩 cpu
      🟩 amd64              Pass: 100%/2   | Total:  8m 13s | Avg:  4m 06s | Max:  5m 43s | Hits:  97%/280   
    🟩 ctk
      🟩 12.8               Pass: 100%/2   | Total:  8m 13s | Avg:  4m 06s | Max:  5m 43s | Hits:  97%/280   
    🟩 cudacxx
      🟩 nvcc12.8           Pass: 100%/2   | Total:  8m 13s | Avg:  4m 06s | Max:  5m 43s | Hits:  97%/280   
    🟩 cudacxx_family
      🟩 nvcc               Pass: 100%/2   | Total:  8m 13s | Avg:  4m 06s | Max:  5m 43s | Hits:  97%/280   
    🟩 cxx
      🟩 GCC13              Pass: 100%/2   | Total:  8m 13s | Avg:  4m 06s | Max:  5m 43s | Hits:  97%/280   
    🟩 cxx_family
      🟩 GCC                Pass: 100%/2   | Total:  8m 13s | Avg:  4m 06s | Max:  5m 43s | Hits:  97%/280   
    🟩 gpu
      🟩 rtx2080            Pass: 100%/2   | Total:  8m 13s | Avg:  4m 06s | Max:  5m 43s | Hits:  97%/280   
    🟩 jobs
      🟩 Build              Pass: 100%/1   | Total:  2m 30s | Avg:  2m 30s | Max:  2m 30s | Hits:  97%/140   
      🟩 Test               Pass: 100%/1   | Total:  5m 43s | Avg:  5m 43s | Max:  5m 43s | Hits:  98%/140   
    
  • 🟩 python: Pass: 100%/1 | Total: 32m 24s | Avg: 32m 24s | Max: 32m 24s

    🟩 cpu
      🟩 amd64              Pass: 100%/1   | Total: 32m 24s | Avg: 32m 24s | Max: 32m 24s
    🟩 ctk
      🟩 12.8               Pass: 100%/1   | Total: 32m 24s | Avg: 32m 24s | Max: 32m 24s
    🟩 cudacxx
      🟩 nvcc12.8           Pass: 100%/1   | Total: 32m 24s | Avg: 32m 24s | Max: 32m 24s
    🟩 cudacxx_family
      🟩 nvcc               Pass: 100%/1   | Total: 32m 24s | Avg: 32m 24s | Max: 32m 24s
    🟩 cxx
      🟩 GCC13              Pass: 100%/1   | Total: 32m 24s | Avg: 32m 24s | Max: 32m 24s
    🟩 cxx_family
      🟩 GCC                Pass: 100%/1   | Total: 32m 24s | Avg: 32m 24s | Max: 32m 24s
    🟩 gpu
      🟩 rtx2080            Pass: 100%/1   | Total: 32m 24s | Avg: 32m 24s | Max: 32m 24s
    🟩 jobs
      🟩 Test               Pass: 100%/1   | Total: 32m 24s | Avg: 32m 24s | Max: 32m 24s
    

👃 Inspect Changes

Modifications in project?

Project
CCCL Infrastructure
libcu++
+/- CUB
Thrust
CUDA Experimental
python
CCCL C Parallel Library
Catch2Helper

Modifications in project or dependencies?

Project
CCCL Infrastructure
libcu++
+/- CUB
+/- Thrust
CUDA Experimental
+/- python
+/- CCCL C Parallel Library
+/- Catch2Helper

🏃‍ Runner counts (total jobs: 90)

# Runner
65 linux-amd64-cpu16
9 windows-amd64-cpu16
6 linux-amd64-gpu-rtxa6000-latest-1
4 linux-arm64-cpu16
3 linux-amd64-gpu-rtx4090-latest-1
2 linux-amd64-gpu-rtx2080-latest-1
1 linux-amd64-gpu-h100-latest-1

@bernhardmgruber
Copy link
Contributor Author

@gonidelis please provide multi histogram benchmarks as well. Thx!

Copy link
Contributor

github-actions bot commented Feb 6, 2025

🟩 CI finished in 1h 10m: Pass: 100%/90 | Total: 1d 16h | Avg: 27m 18s | Max: 1h 04m | Hits: 90%/132225
  • 🟩 cub: Pass: 100%/44 | Total: 1d 05h | Avg: 40m 15s | Max: 1h 04m | Hits: 88%/52320

    🟩 cpu
      🟩 amd64              Pass: 100%/42  | Total:  1d 03h | Avg: 39m 56s | Max:  1h 04m | Hits:  88%/49888 
      🟩 arm64              Pass: 100%/2   | Total:  1h 34m | Avg: 47m 10s | Max: 47m 53s | Hits:  94%/2432  
    🟩 ctk
      🟩 12.0               Pass: 100%/5   | Total:  3h 25m | Avg: 41m 04s | Max: 43m 56s | Hits:  80%/5914  
      🟩 12.5               Pass: 100%/2   | Total:  1h 36m | Avg: 48m 20s | Max: 51m 37s | Hits:  93%/2250  
      🟩 12.8               Pass: 100%/37  | Total:  1d 00h | Avg: 39m 42s | Max:  1h 04m | Hits:  89%/44156 
    🟩 cudacxx
      🟩 ClangCUDA18        Pass: 100%/2   | Total:  1h 50m | Avg: 55m 00s | Max: 57m 00s | Hits:  95%/2104  
      🟩 nvcc12.0           Pass: 100%/5   | Total:  3h 25m | Avg: 41m 04s | Max: 43m 56s | Hits:  80%/5914  
      🟩 nvcc12.5           Pass: 100%/2   | Total:  1h 36m | Avg: 48m 20s | Max: 51m 37s | Hits:  93%/2250  
      🟩 nvcc12.8           Pass: 100%/35  | Total: 22h 39m | Avg: 38m 50s | Max:  1h 04m | Hits:  88%/42052 
    🟩 cudacxx_family
      🟩 ClangCUDA          Pass: 100%/2   | Total:  1h 50m | Avg: 55m 00s | Max: 57m 00s | Hits:  95%/2104  
      🟩 nvcc               Pass: 100%/42  | Total:  1d 03h | Avg: 39m 33s | Max:  1h 04m | Hits:  88%/50216 
    🟩 cxx
      🟩 Clang14            Pass: 100%/4   | Total:  2h 49m | Avg: 42m 19s | Max: 43m 56s | Hits:  94%/4872  
      🟩 Clang15            Pass: 100%/2   | Total:  1h 28m | Avg: 44m 06s | Max: 45m 27s | Hits:  94%/2432  
      🟩 Clang16            Pass: 100%/2   | Total:  1h 21m | Avg: 40m 30s | Max: 41m 17s | Hits:  94%/2432  
      🟩 Clang17            Pass: 100%/2   | Total:  1h 24m | Avg: 42m 09s | Max: 43m 49s | Hits:  94%/2432  
      🟩 Clang18            Pass: 100%/7   | Total:  4h 41m | Avg: 40m 16s | Max: 57m 00s | Hits:  96%/8184  
      🟩 GCC7               Pass: 100%/2   | Total:  1h 45m | Avg: 52m 42s | Max:  1h 04m | Hits:  81%/2436  
      🟩 GCC8               Pass: 100%/1   | Total: 42m 25s | Avg: 42m 25s | Max: 42m 25s | Hits:  94%/1218  
      🟩 GCC9               Pass: 100%/2   | Total:  1h 27m | Avg: 43m 52s | Max: 44m 08s | Hits:  94%/2436  
      🟩 GCC10              Pass: 100%/2   | Total:  1h 23m | Avg: 41m 30s | Max: 42m 42s | Hits:  94%/2436  
      🟩 GCC11              Pass: 100%/2   | Total:  1h 28m | Avg: 44m 20s | Max: 44m 58s | Hits:  94%/2432  
      🟩 GCC12              Pass: 100%/2   | Total:  1h 28m | Avg: 44m 22s | Max: 45m 32s | Hits:  94%/2432  
      🟩 GCC13              Pass: 100%/10  | Total:  5h 14m | Avg: 31m 24s | Max: 54m 34s | Hits:  96%/12160 
      🟩 MSVC14.29          Pass: 100%/2   | Total:  1h 15m | Avg: 37m 55s | Max: 41m 48s | Hits:  15%/2084  
      🟩 MSVC14.42          Pass: 100%/2   | Total:  1h 24m | Avg: 42m 06s | Max: 42m 36s | Hits:  15%/2084  
      🟩 NVHPC24.7          Pass: 100%/2   | Total:  1h 36m | Avg: 48m 20s | Max: 51m 37s | Hits:  93%/2250  
    🟩 cxx_family
      🟩 Clang              Pass: 100%/17  | Total: 11h 44m | Avg: 41m 27s | Max: 57m 00s | Hits:  95%/20352 
      🟩 GCC                Pass: 100%/21  | Total: 13h 30m | Avg: 38m 34s | Max:  1h 04m | Hits:  94%/25550 
      🟩 MSVC               Pass: 100%/4   | Total:  2h 40m | Avg: 40m 00s | Max: 42m 36s | Hits:  15%/4168  
      🟩 NVHPC              Pass: 100%/2   | Total:  1h 36m | Avg: 48m 20s | Max: 51m 37s | Hits:  93%/2250  
    🟩 gpu
      🟩 h100               Pass: 100%/2   | Total: 44m 02s | Avg: 22m 01s | Max: 25m 26s | Hits:  96%/2432  
      🟩 rtx2080            Pass: 100%/34  | Total:  1d 01h | Avg: 44m 38s | Max:  1h 04m | Hits:  85%/40160 
      🟩 rtxa6000           Pass: 100%/8   | Total:  3h 29m | Avg: 26m 14s | Max: 43m 46s | Hits:  98%/9728  
    🟩 jobs
      🟩 Build              Pass: 100%/37  | Total:  1d 03h | Avg: 43m 47s | Max:  1h 04m | Hits:  86%/43808 
      🟩 DeviceLaunch       Pass: 100%/1   | Total: 21m 14s | Avg: 21m 14s | Max: 21m 14s | Hits:  99%/1216  
      🟩 GraphCapture       Pass: 100%/1   | Total: 16m 48s | Avg: 16m 48s | Max: 16m 48s | Hits:  99%/1216  
      🟩 HostLaunch         Pass: 100%/3   | Total:  1h 13m | Avg: 24m 21s | Max: 25m 26s | Hits:  99%/3648  
      🟩 TestGPU            Pass: 100%/2   | Total: 40m 18s | Avg: 20m 09s | Max: 21m 04s | Hits:  99%/2432  
    🟩 sm
      🟩 90                 Pass: 100%/2   | Total: 44m 02s | Avg: 22m 01s | Max: 25m 26s | Hits:  96%/2432  
      🟩 90;90a;100         Pass: 100%/1   | Total: 54m 34s | Avg: 54m 34s | Max: 54m 34s | Hits:  94%/1216  
    🟩 std
      🟩 17                 Pass: 100%/20  | Total: 14h 44m | Avg: 44m 13s | Max:  1h 04m | Hits:  82%/23559 
      🟩 20                 Pass: 100%/24  | Total: 14h 47m | Avg: 36m 57s | Max: 54m 34s | Hits:  93%/28761 
    
  • 🟩 thrust: Pass: 100%/43 | Total: 10h 48m | Avg: 15m 04s | Max: 30m 51s | Hits: 92%/79625

    🟩 cmake_options
      🟩 -DTHRUST_DISPATCH_TYPE=Force32bit Pass: 100%/2   | Total: 23m 03s | Avg: 11m 31s | Max: 11m 41s | Hits:  97%/3706  
    🟩 cpu
      🟩 amd64              Pass: 100%/41  | Total: 10h 24m | Avg: 15m 13s | Max: 30m 51s | Hits:  92%/75920 
      🟩 arm64              Pass: 100%/2   | Total: 23m 57s | Avg: 11m 58s | Max: 12m 36s | Hits:  94%/3705  
    🟩 ctk
      🟩 12.0               Pass: 100%/5   | Total:  1h 16m | Avg: 15m 12s | Max: 23m 37s | Hits:  89%/9256  
      🟩 12.5               Pass: 100%/2   | Total: 52m 30s | Avg: 26m 15s | Max: 26m 38s | Hits:  93%/3704  
      🟩 12.8               Pass: 100%/36  | Total:  8h 39m | Avg: 14m 26s | Max: 30m 51s | Hits:  92%/66665 
    🟩 cudacxx
      🟩 ClangCUDA18        Pass: 100%/2   | Total: 25m 50s | Avg: 12m 55s | Max: 13m 02s | Hits:  95%/3704  
      🟩 nvcc12.0           Pass: 100%/5   | Total:  1h 16m | Avg: 15m 12s | Max: 23m 37s | Hits:  89%/9256  
      🟩 nvcc12.5           Pass: 100%/2   | Total: 52m 30s | Avg: 26m 15s | Max: 26m 38s | Hits:  93%/3704  
      🟩 nvcc12.8           Pass: 100%/34  | Total:  8h 13m | Avg: 14m 31s | Max: 30m 51s | Hits:  92%/62961 
    🟩 cudacxx_family
      🟩 ClangCUDA          Pass: 100%/2   | Total: 25m 50s | Avg: 12m 55s | Max: 13m 02s | Hits:  95%/3704  
      🟩 nvcc               Pass: 100%/41  | Total: 10h 22m | Avg: 15m 10s | Max: 30m 51s | Hits:  92%/75921 
    🟩 cxx
      🟩 Clang14            Pass: 100%/4   | Total: 51m 08s | Avg: 12m 47s | Max: 13m 50s | Hits:  95%/7408  
      🟩 Clang15            Pass: 100%/2   | Total: 25m 59s | Avg: 12m 59s | Max: 13m 24s | Hits:  95%/3704  
      🟩 Clang16            Pass: 100%/2   | Total: 25m 48s | Avg: 12m 54s | Max: 12m 59s | Hits:  95%/3704  
      🟩 Clang17            Pass: 100%/2   | Total: 26m 06s | Avg: 13m 03s | Max: 13m 57s | Hits:  95%/3704  
      🟩 Clang18            Pass: 100%/7   | Total:  1h 21m | Avg: 11m 42s | Max: 13m 40s | Hits:  96%/12964 
      🟩 GCC7               Pass: 100%/2   | Total: 27m 25s | Avg: 13m 42s | Max: 13m 43s | Hits:  94%/3706  
      🟩 GCC8               Pass: 100%/1   | Total: 12m 46s | Avg: 12m 46s | Max: 12m 46s | Hits:  94%/1853  
      🟩 GCC9               Pass: 100%/2   | Total: 27m 32s | Avg: 13m 46s | Max: 13m 51s | Hits:  94%/3706  
      🟩 GCC10              Pass: 100%/2   | Total: 28m 56s | Avg: 14m 28s | Max: 14m 39s | Hits:  94%/3706  
      🟩 GCC11              Pass: 100%/2   | Total: 28m 26s | Avg: 14m 13s | Max: 14m 34s | Hits:  94%/3706  
      🟩 GCC12              Pass: 100%/2   | Total: 29m 42s | Avg: 14m 51s | Max: 15m 19s | Hits:  94%/3706  
      🟩 GCC13              Pass: 100%/8   | Total:  1h 39m | Avg: 12m 26s | Max: 15m 28s | Hits:  96%/14824 
      🟩 MSVC14.29          Pass: 100%/2   | Total: 48m 48s | Avg: 24m 24s | Max: 25m 11s | Hits:  69%/3692  
      🟩 MSVC14.42          Pass: 100%/3   | Total:  1h 21m | Avg: 27m 13s | Max: 30m 51s | Hits:  69%/5538  
      🟩 NVHPC24.7          Pass: 100%/2   | Total: 52m 30s | Avg: 26m 15s | Max: 26m 38s | Hits:  93%/3704  
    🟩 cxx_family
      🟩 Clang              Pass: 100%/17  | Total:  3h 30m | Avg: 12m 24s | Max: 13m 57s | Hits:  95%/31484 
      🟩 GCC                Pass: 100%/19  | Total:  4h 14m | Avg: 13m 23s | Max: 15m 28s | Hits:  95%/35207 
      🟩 MSVC               Pass: 100%/5   | Total:  2h 10m | Avg: 26m 05s | Max: 30m 51s | Hits:  69%/9230  
      🟩 NVHPC              Pass: 100%/2   | Total: 52m 30s | Avg: 26m 15s | Max: 26m 38s | Hits:  93%/3704  
    🟩 gpu
      🟩 rtx2080            Pass: 100%/33  | Total:  8h 20m | Avg: 15m 09s | Max: 26m 38s | Hits:  92%/61112 
      🟩 rtx4090            Pass: 100%/10  | Total:  2h 27m | Avg: 14m 47s | Max: 30m 51s | Hits:  92%/18513 
    🟩 jobs
      🟩 Build              Pass: 100%/37  | Total:  9h 27m | Avg: 15m 20s | Max: 26m 38s | Hits:  92%/68516 
      🟩 TestCPU            Pass: 100%/3   | Total: 47m 00s | Avg: 15m 40s | Max: 30m 51s | Hits:  89%/5551  
      🟩 TestGPU            Pass: 100%/3   | Total: 33m 32s | Avg: 11m 10s | Max: 11m 42s | Hits:  99%/5558  
    🟩 sm
      🟩 90;90a;100         Pass: 100%/1   | Total: 13m 16s | Avg: 13m 16s | Max: 13m 16s | Hits:  94%/1853  
    🟩 std
      🟩 17                 Pass: 100%/20  | Total:  5h 14m | Avg: 15m 44s | Max: 25m 52s | Hits:  91%/37031 
      🟩 20                 Pass: 100%/21  | Total:  5h 10m | Avg: 14m 46s | Max: 30m 51s | Hits:  93%/38888 
    
  • 🟩 cccl_c_parallel: Pass: 100%/2 | Total: 8m 00s | Avg: 4m 00s | Max: 5m 37s | Hits: 98%/280

    🟩 cpu
      🟩 amd64              Pass: 100%/2   | Total:  8m 00s | Avg:  4m 00s | Max:  5m 37s | Hits:  98%/280   
    🟩 ctk
      🟩 12.8               Pass: 100%/2   | Total:  8m 00s | Avg:  4m 00s | Max:  5m 37s | Hits:  98%/280   
    🟩 cudacxx
      🟩 nvcc12.8           Pass: 100%/2   | Total:  8m 00s | Avg:  4m 00s | Max:  5m 37s | Hits:  98%/280   
    🟩 cudacxx_family
      🟩 nvcc               Pass: 100%/2   | Total:  8m 00s | Avg:  4m 00s | Max:  5m 37s | Hits:  98%/280   
    🟩 cxx
      🟩 GCC13              Pass: 100%/2   | Total:  8m 00s | Avg:  4m 00s | Max:  5m 37s | Hits:  98%/280   
    🟩 cxx_family
      🟩 GCC                Pass: 100%/2   | Total:  8m 00s | Avg:  4m 00s | Max:  5m 37s | Hits:  98%/280   
    🟩 gpu
      🟩 rtx2080            Pass: 100%/2   | Total:  8m 00s | Avg:  4m 00s | Max:  5m 37s | Hits:  98%/280   
    🟩 jobs
      🟩 Build              Pass: 100%/1   | Total:  2m 23s | Avg:  2m 23s | Max:  2m 23s | Hits:  98%/140   
      🟩 Test               Pass: 100%/1   | Total:  5m 37s | Avg:  5m 37s | Max:  5m 37s | Hits:  98%/140   
    
  • 🟩 python: Pass: 100%/1 | Total: 29m 17s | Avg: 29m 17s | Max: 29m 17s

    🟩 cpu
      🟩 amd64              Pass: 100%/1   | Total: 29m 17s | Avg: 29m 17s | Max: 29m 17s
    🟩 ctk
      🟩 12.8               Pass: 100%/1   | Total: 29m 17s | Avg: 29m 17s | Max: 29m 17s
    🟩 cudacxx
      🟩 nvcc12.8           Pass: 100%/1   | Total: 29m 17s | Avg: 29m 17s | Max: 29m 17s
    🟩 cudacxx_family
      🟩 nvcc               Pass: 100%/1   | Total: 29m 17s | Avg: 29m 17s | Max: 29m 17s
    🟩 cxx
      🟩 GCC13              Pass: 100%/1   | Total: 29m 17s | Avg: 29m 17s | Max: 29m 17s
    🟩 cxx_family
      🟩 GCC                Pass: 100%/1   | Total: 29m 17s | Avg: 29m 17s | Max: 29m 17s
    🟩 gpu
      🟩 rtx2080            Pass: 100%/1   | Total: 29m 17s | Avg: 29m 17s | Max: 29m 17s
    🟩 jobs
      🟩 Test               Pass: 100%/1   | Total: 29m 17s | Avg: 29m 17s | Max: 29m 17s
    

👃 Inspect Changes

Modifications in project?

Project
CCCL Infrastructure
libcu++
+/- CUB
Thrust
CUDA Experimental
python
CCCL C Parallel Library
Catch2Helper

Modifications in project or dependencies?

Project
CCCL Infrastructure
libcu++
+/- CUB
+/- Thrust
CUDA Experimental
+/- python
+/- CCCL C Parallel Library
+/- Catch2Helper

🏃‍ Runner counts (total jobs: 90)

# Runner
65 linux-amd64-cpu16
9 windows-amd64-cpu16
6 linux-amd64-gpu-rtxa6000-latest-1
4 linux-arm64-cpu16
3 linux-amd64-gpu-rtx4090-latest-1
2 linux-amd64-gpu-rtx2080-latest-1
1 linux-amd64-gpu-h100-latest-1

@bernhardmgruber bernhardmgruber merged commit e7aae03 into NVIDIA:main Feb 6, 2025
102 of 104 checks passed
Copy link
Contributor

github-actions bot commented Feb 6, 2025

Backport failed for branch/2.8.x, because it was unable to cherry-pick the commit(s).

Please cherry-pick the changes locally and resolve any conflicts.

git fetch origin branch/2.8.x
git worktree add -d .worktree/backport-3691-to-branch/2.8.x origin/branch/2.8.x
cd .worktree/backport-3691-to-branch/2.8.x
git switch --create backport-3691-to-branch/2.8.x
git cherry-pick -x e7aae03124f20d8a4783d3e1668307d4a9e3bb8b

@bernhardmgruber bernhardmgruber deleted the fix_histo_tunings branch February 6, 2025 22:44
bernhardmgruber added a commit to bernhardmgruber/cccl that referenced this pull request Feb 6, 2025
The tuning data member names did not match the one used when selecting
tunings, so all SM100 tunings were SFINAE-ed out.

Also drop tunings with no benefit.
elstehle pushed a commit to bernhardmgruber/cccl that referenced this pull request Feb 7, 2025
The tuning data member names did not match the one used when selecting
tunings, so all SM100 tunings were SFINAE-ed out.

Also drop tunings with no benefit.
elstehle pushed a commit that referenced this pull request Feb 7, 2025
* Add b200 tunings for histogram (#3616)

Co-authored-by: Giannis Gonidelis <[email protected]>

* Fix SM100 histogram tunings (#3691)

The tuning data member names did not match the one used when selecting
tunings, so all SM100 tunings were SFINAE-ed out.

Also drop tunings with no benefit.

---------

Co-authored-by: Giannis Gonidelis <[email protected]>
@gonidelis
Copy link
Member

histogram.multi.even

|  SampleT{ct}  |  CounterT{ct}  |  OffsetT{ct}  |  Elements{io}  |  Bins   |  Entropy  |   Ref Time |   Ref Noise |   Cmp Time |   Cmp Noise |       Diff |   %Diff |  Status  |
|---------------|----------------|---------------|----------------|---------|-----------|------------|-------------|------------|-------------|------------|---------|----------|
|      I8       |      I32       |      I32      |      2^16      |   32    |   0.201   |  12.636 us |       7.67% |  12.918 us |       6.15% |   0.281 us |   2.23% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      |   32    |   0.201   |  15.666 us |      10.71% |  16.049 us |       9.49% |   0.383 us |   2.44% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      |   32    |   0.201   |  61.911 us |       1.53% |  62.337 us |       0.91% |   0.425 us |   0.69% |   SAME   |
|      I8       |      I32       |      I32      |      2^28      |   32    |   0.201   | 793.974 us |       0.19% | 794.522 us |       0.17% |   0.547 us |   0.07% |   SAME   |
|      I8       |      I32       |      I32      |      2^16      |   128   |   0.201   |  12.708 us |       7.32% |  13.220 us |       3.41% |   0.512 us |   4.03% |   SLOW   |
|      I8       |      I32       |      I32      |      2^20      |   128   |   0.201   |  16.492 us |       6.21% |  16.666 us |       5.86% |   0.174 us |   1.06% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      |   128   |   0.201   |  72.124 us |       1.28% |  72.582 us |       0.76% |   0.458 us |   0.64% |   SAME   |
|      I8       |      I32       |      I32      |      2^28      |   128   |   0.201   | 953.612 us |       0.71% | 960.688 us |       0.12% |   7.076 us |   0.74% |   SLOW   |
|      I8       |      I32       |      I32      |      2^16      |  2048   |   0.201   |  12.848 us |       6.68% |  13.198 us |       3.51% |   0.350 us |   2.72% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      |  2048   |   0.201   |  20.896 us |       4.93% |  21.455 us |       4.10% |   0.559 us |   2.68% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      |  2048   |   0.201   |  83.926 us |       1.67% |  84.857 us |       0.62% |   0.932 us |   1.11% |   SLOW   |
|      I8       |      I32       |      I32      |      2^28      |  2048   |   0.201   | 956.802 us |       0.18% | 962.421 us |       0.14% |   5.619 us |   0.59% |   SLOW   |
|      I8       |      I32       |      I32      |      2^16      | 2097152 |   0.201   |  17.468 us |       1.15% |  17.426 us |       0.58% |  -0.043 us |  -0.24% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      | 2097152 |   0.201   |  25.945 us |       5.03% |  26.337 us |       4.06% |   0.392 us |   1.51% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      | 2097152 |   0.201   |  90.583 us |       1.09% |  90.843 us |       0.83% |   0.260 us |   0.29% |   SAME   |
|      I8       |      I32       |      I32      |      2^28      | 2097152 |   0.201   | 960.641 us |       0.14% | 966.982 us |       0.12% |   6.341 us |   0.66% |   SLOW   |
|      I8       |      I32       |      I32      |      2^16      |   32    |     1     |  12.712 us |       7.29% |  13.114 us |       4.56% |   0.403 us |   3.17% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      |   32    |     1     |  14.506 us |       6.40% |  14.990 us |       5.21% |   0.483 us |   3.33% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      |   32    |     1     |  45.262 us |       2.73% |  45.866 us |       1.40% |   0.604 us |   1.33% |   SAME   |
|      I8       |      I32       |      I32      |      2^28      |   32    |     1     | 507.051 us |       0.71% | 509.597 us |       0.75% |   2.547 us |   0.50% |   SAME   |
|      I8       |      I32       |      I32      |      2^16      |   128   |     1     |  12.633 us |       7.54% |  13.018 us |       5.53% |   0.385 us |   3.04% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      |   128   |     1     |  14.503 us |       7.09% |  14.758 us |       6.46% |   0.255 us |   1.76% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      |   128   |     1     |  44.156 us |       3.45% |  44.575 us |       2.75% |   0.420 us |   0.95% |   SAME   |
|      I8       |      I32       |      I32      |      2^28      |   128   |     1     | 495.109 us |       0.75% | 495.508 us |       0.73% |   0.399 us |   0.08% |   SAME   |
|      I8       |      I32       |      I32      |      2^16      |  2048   |     1     |  14.816 us |       6.12% |  15.123 us |       4.29% |   0.308 us |   2.08% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      |  2048   |     1     |  53.750 us |       2.28% |  54.329 us |       2.04% |   0.579 us |   1.08% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      |  2048   |     1     |  78.723 us |       0.84% |  78.897 us |       0.65% |   0.174 us |   0.22% |   SAME   |
|      I8       |      I32       |      I32      |      2^28      |  2048   |     1     | 499.107 us |       0.66% | 499.523 us |       0.80% |   0.416 us |   0.08% |   SAME   |
|      I8       |      I32       |      I32      |      2^16      | 2097152 |     1     |  19.515 us |       0.99% |  19.477 us |       0.46% |  -0.039 us |  -0.20% |   SAME   |
|      I8       |      I32       |      I32      |      2^20      | 2097152 |     1     |  67.249 us |       1.94% |  67.273 us |       1.60% |   0.025 us |   0.04% |   SAME   |
|      I8       |      I32       |      I32      |      2^24      | 2097152 |     1     |  87.458 us |       0.93% |  87.365 us |       0.91% |  -0.093 us |  -0.11% |   SAME   |
|      I8       |      I32       |      I32      |      2^28      | 2097152 |     1     | 504.073 us |       0.64% | 504.380 us |       0.77% |   0.307 us |   0.06% |   SAME   |
|      I16      |      I32       |      I32      |      2^16      |   32    |   0.201   |  15.049 us |       5.03% |  15.351 us |       1.51% |   0.302 us |   2.00% |   SLOW   |
|      I16      |      I32       |      I32      |      2^20      |   32    |   0.201   |  18.595 us |       5.46% |  18.848 us |       5.03% |   0.253 us |   1.36% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      |   32    |   0.201   |  97.510 us |       1.72% |  97.986 us |       1.30% |   0.475 us |   0.49% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      |   32    |   0.201   |   1.340 ms |       0.12% |   1.341 ms |       0.11% |   0.455 us |   0.03% |   SAME   |
|      I16      |      I32       |      I32      |      2^16      |   128   |   0.201   |  14.414 us |       6.96% |  14.709 us |       6.41% |   0.295 us |   2.05% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      |   128   |   0.201   |  18.309 us |       5.36% |  18.593 us |       5.49% |   0.284 us |   1.55% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      |   128   |   0.201   |  97.570 us |       1.87% |  97.869 us |       1.70% |   0.299 us |   0.31% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      |   128   |   0.201   |   1.344 ms |       0.12% |   1.345 ms |       0.11% |   0.679 us |   0.05% |   SAME   |
|      I16      |      I32       |      I32      |      2^16      |  2048   |   0.201   |  19.295 us |       3.25% |  19.449 us |       1.21% |   0.154 us |   0.80% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      |  2048   |   0.201   |  32.061 us |       5.25% |  32.450 us |       4.69% |   0.389 us |   1.21% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      |  2048   |   0.201   | 287.051 us |       1.07% | 287.077 us |       1.06% |   0.026 us |   0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      |  2048   |   0.201   |   3.868 ms |       0.36% |   3.869 ms |       0.37% |   0.372 us |   0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^16      | 2097152 |   0.201   |   7.819 ms |       0.11% |   7.819 ms |       0.11% |  -0.759 us |  -0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      | 2097152 |   0.201   |  10.481 ms |       0.16% |  10.481 ms |       0.17% |  -0.889 us |  -0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      | 2097152 |   0.201   |  10.818 ms |       0.22% |  10.816 ms |       0.20% |  -1.358 us |  -0.01% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      | 2097152 |   0.201   |  12.259 ms |       0.36% |  12.263 ms |       0.35% |   3.878 us |   0.03% |   SAME   |
|      I16      |      I32       |      I32      |      2^16      |   32    |     1     |  15.214 us |       5.69% |  15.477 us |       3.89% |   0.263 us |   1.73% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      |   32    |     1     |  18.538 us |       5.54% |  18.715 us |       5.38% |   0.177 us |   0.96% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      |   32    |     1     | 100.754 us |       1.02% | 101.174 us |       0.59% |   0.420 us |   0.42% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      |   32    |     1     |   1.387 ms |       0.11% |   1.388 ms |       0.09% |   0.457 us |   0.03% |   SAME   |
|      I16      |      I32       |      I32      |      2^16      |   128   |     1     |  14.409 us |       6.84% |  14.663 us |       6.56% |   0.254 us |   1.76% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      |   128   |     1     |  18.389 us |       5.47% |  18.539 us |       5.54% |   0.149 us |   0.81% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      |   128   |     1     |  96.490 us |       1.21% |  96.850 us |       1.18% |   0.360 us |   0.37% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      |   128   |     1     |   1.332 ms |       0.13% |   1.333 ms |       0.13% |   0.715 us |   0.05% |   SAME   |
|      I16      |      I32       |      I32      |      2^16      |  2048   |     1     |  20.846 us |       4.93% |  21.046 us |       4.07% |   0.199 us |   0.96% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      |  2048   |     1     |  38.012 us |       3.93% |  38.647 us |       3.19% |   0.634 us |   1.67% |   SAME   |
|      I16      |      I32       |      I32      |      2^24      |  2048   |     1     | 280.428 us |       1.43% | 280.963 us |       1.43% |   0.535 us |   0.19% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      |  2048   |     1     |   4.243 ms |       0.40% |   4.253 ms |       0.53% |  10.314 us |   0.24% |   SAME   |
|      I16      |      I32       |      I32      |      2^16      | 2097152 |     1     |   7.821 ms |       0.11% |   7.823 ms |       0.12% |   2.025 us |   0.03% |   SAME   |
|      I16      |      I32       |      I32      |      2^20      | 2097152 |     1     |  10.481 ms |       0.16% |  10.497 ms |       0.15% |  16.394 us |   0.16% |   SLOW   |
|      I16      |      I32       |      I32      |      2^24      | 2097152 |     1     |  10.817 ms |       0.22% |  10.839 ms |       0.22% |  22.290 us |   0.21% |   SAME   |
|      I16      |      I32       |      I32      |      2^28      | 2097152 |     1     |  12.261 ms |       0.34% |  12.273 ms |       0.36% |  11.527 us |   0.09% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      |   32    |   0.201   |  15.092 us |       5.66% |  15.434 us |       3.25% |   0.342 us |   2.27% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      |   32    |   0.201   |  20.434 us |       5.17% |  20.618 us |       5.27% |   0.183 us |   0.90% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      |   32    |   0.201   | 137.563 us |       0.94% | 137.778 us |       0.84% |   0.214 us |   0.16% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      |   32    |   0.201   |   1.946 ms |       0.15% |   1.946 ms |       0.14% |   0.688 us |   0.04% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      |   128   |   0.201   |  14.442 us |       6.94% |  14.703 us |       6.43% |   0.261 us |   1.81% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      |   128   |   0.201   |  20.463 us |       5.17% |  20.660 us |       5.48% |   0.198 us |   0.97% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      |   128   |   0.201   | 137.816 us |       1.18% | 138.068 us |       1.01% |   0.252 us |   0.18% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      |   128   |   0.201   |   1.949 ms |       0.15% |   1.949 ms |       0.14% |   0.539 us |   0.03% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      |  2048   |   0.201   |  18.701 us |       5.16% |  19.033 us |       4.38% |   0.332 us |   1.78% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      |  2048   |   0.201   |  34.386 us |       4.25% |  34.913 us |       3.54% |   0.526 us |   1.53% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      |  2048   |   0.201   | 304.389 us |       1.08% | 304.582 us |       1.12% |   0.193 us |   0.06% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      |  2048   |   0.201   |   4.430 ms |       0.21% |   4.431 ms |       0.22% |   0.966 us |   0.02% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      | 2097152 |   0.201   |   7.814 ms |       0.12% |   7.818 ms |       0.11% |   3.955 us |   0.05% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      | 2097152 |   0.201   |   9.754 ms |       0.21% |   9.775 ms |       0.24% |  21.148 us |   0.22% |   SLOW   |
|      I32      |      I32       |      I32      |      2^24      | 2097152 |   0.201   |  10.245 ms |       0.14% |  10.252 ms |       0.15% |   7.711 us |   0.08% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      | 2097152 |   0.201   |  16.234 ms |       0.23% |  16.227 ms |       0.23% |  -6.886 us |  -0.04% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      |   32    |     1     |  15.149 us |       5.56% |  15.599 us |       3.23% |   0.449 us |   2.97% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      |   32    |     1     |  21.620 us |       8.83% |  22.379 us |       6.18% |   0.759 us |   3.51% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      |   32    |     1     | 143.429 us |       0.77% | 144.409 us |       0.28% |   0.981 us |   0.68% |   SLOW   |
|      I32      |      I32       |      I32      |      2^28      |   32    |     1     |   2.039 ms |       0.15% |   2.040 ms |       0.13% |   1.176 us |   0.06% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      |   128   |     1     |  14.529 us |       6.94% |  15.348 us |       1.80% |   0.819 us |   5.64% |   SLOW   |
|      I32      |      I32       |      I32      |      2^20      |   128   |     1     |  20.430 us |       5.14% |  21.235 us |       3.84% |   0.805 us |   3.94% |   SLOW   |
|      I32      |      I32       |      I32      |      2^24      |   128   |     1     | 137.307 us |       0.75% | 138.239 us |       0.18% |   0.931 us |   0.68% |   SLOW   |
|      I32      |      I32       |      I32      |      2^28      |   128   |     1     |   1.936 ms |       0.16% |   1.938 ms |       0.15% |   1.159 us |   0.06% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      |  2048   |     1     |  20.990 us |       4.32% |  21.518 us |       1.17% |   0.528 us |   2.51% |   SLOW   |
|      I32      |      I32       |      I32      |      2^20      |  2048   |     1     |  37.668 us |       3.92% |  38.472 us |       2.44% |   0.805 us |   2.14% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      |  2048   |     1     | 290.055 us |       2.64% | 290.313 us |       2.56% |   0.259 us |   0.09% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      |  2048   |     1     |   4.614 ms |       0.21% |   4.616 ms |       0.26% |   1.527 us |   0.03% |   SAME   |
|      I32      |      I32       |      I32      |      2^16      | 2097152 |     1     |   7.808 ms |       0.07% |   7.805 ms |       0.07% |  -2.747 us |  -0.04% |   SAME   |
|      I32      |      I32       |      I32      |      2^20      | 2097152 |     1     |   9.916 ms |       0.16% |   9.914 ms |       0.12% |  -2.772 us |  -0.03% |   SAME   |
|      I32      |      I32       |      I32      |      2^24      | 2097152 |     1     |  12.236 ms |       0.26% |  12.251 ms |       0.26% |  15.203 us |   0.12% |   SAME   |
|      I32      |      I32       |      I32      |      2^28      | 2097152 |     1     |  47.390 ms |       0.20% |  47.417 ms |       0.21% |  27.017 us |   0.06% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      |   32    |   0.201   |  12.974 us |       8.70% |  13.553 us |       3.97% |   0.578 us |   4.46% |   SLOW   |
|      I64      |      I32       |      I32      |      2^20      |   32    |   0.201   |  22.527 us |       3.88% |  23.525 us |       0.70% |   0.998 us |   4.43% |   SLOW   |
|      I64      |      I32       |      I32      |      2^24      |   32    |   0.201   | 156.853 us |       0.84% | 157.491 us |       0.64% |   0.638 us |   0.41% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      |   32    |   0.201   |   2.301 ms |       0.09% |   2.282 ms |       0.57% | -19.024 us |  -0.83% |   FAST   |
|      I64      |      I32       |      I32      |      2^16      |   128   |   0.201   |  13.300 us |       5.60% |  13.513 us |       4.05% |   0.213 us |   1.60% |   SAME   |
|      I64      |      I32       |      I32      |      2^20      |   128   |   0.201   |  22.018 us |       3.50% |  23.362 us |       2.50% |   1.344 us |   6.10% |   SLOW   |
|      I64      |      I32       |      I32      |      2^24      |   128   |   0.201   | 157.480 us |       0.61% | 157.641 us |       0.67% |   0.161 us |   0.10% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      |   128   |   0.201   |   2.327 ms |       0.09% |   2.302 ms |       0.07% | -25.093 us |  -1.08% |   FAST   |
|      I64      |      I32       |      I32      |      2^16      |  2048   |   0.201   |  17.310 us |       6.99% |  17.644 us |       3.51% |   0.334 us |   1.93% |   SAME   |
|      I64      |      I32       |      I32      |      2^20      |  2048   |   0.201   |  39.839 us |       3.51% |  40.795 us |       2.55% |   0.956 us |   2.40% |   SAME   |
|      I64      |      I32       |      I32      |      2^24      |  2048   |   0.201   | 393.915 us |       0.76% | 394.278 us |       0.64% |   0.364 us |   0.09% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      |  2048   |   0.201   |   5.983 ms |       0.16% |   5.980 ms |       0.16% |  -2.970 us |  -0.05% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      | 2097152 |   0.201   |   7.776 ms |       0.06% |   7.746 ms |       0.05% | -30.465 us |  -0.39% |   FAST   |
|      I64      |      I32       |      I32      |      2^20      | 2097152 |   0.201   |  10.791 ms |       0.15% |  10.755 ms |       0.11% | -36.228 us |  -0.34% |   FAST   |
|      I64      |      I32       |      I32      |      2^24      | 2097152 |   0.201   |  11.283 ms |       0.16% |  11.272 ms |       0.16% | -10.902 us |  -0.10% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      | 2097152 |   0.201   |  17.723 ms |       0.18% |  17.721 ms |       0.16% |  -2.672 us |  -0.02% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      |   32    |     1     |  13.175 us |       8.93% |  13.524 us |       5.87% |   0.349 us |   2.65% |   SAME   |
|      I64      |      I32       |      I32      |      2^20      |   32    |     1     |  20.383 us |       4.42% |  20.772 us |       4.67% |   0.389 us |   1.91% |   SAME   |
|      I64      |      I32       |      I32      |      2^24      |   32    |     1     | 131.393 us |       0.82% | 131.352 us |       0.82% |  -0.041 us |  -0.03% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      |   32    |     1     |   1.825 ms |       0.65% |   1.821 ms |       0.33% |  -4.446 us |  -0.24% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      |   128   |     1     |  13.011 us |       8.12% |  13.254 us |       5.94% |   0.244 us |   1.87% |   SAME   |
|      I64      |      I32       |      I32      |      2^20      |   128   |     1     |  20.105 us |       4.11% |  20.451 us |       4.88% |   0.347 us |   1.72% |   SAME   |
|      I64      |      I32       |      I32      |      2^24      |   128   |     1     | 127.011 us |       0.85% | 128.065 us |       0.83% |   1.054 us |   0.83% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      |   128   |     1     |   1.773 ms |       0.31% |   1.774 ms |       0.33% |   1.774 us |   0.10% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      |  2048   |     1     |  19.590 us |       2.59% |  19.649 us |       2.28% |   0.059 us |   0.30% |   SAME   |
|      I64      |      I32       |      I32      |      2^20      |  2048   |     1     |  40.909 us |       3.60% |  41.069 us |       3.41% |   0.160 us |   0.39% |   SAME   |
|      I64      |      I32       |      I32      |      2^24      |  2048   |     1     | 333.305 us |       1.10% | 334.033 us |       1.19% |   0.728 us |   0.22% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      |  2048   |     1     |   4.830 ms |       0.27% |   4.831 ms |       0.28% |   0.714 us |   0.01% |   SAME   |
|      I64      |      I32       |      I32      |      2^16      | 2097152 |     1     |   7.773 ms |       0.18% |   7.757 ms |       0.05% | -16.275 us |  -0.21% |   FAST   |
|      I64      |      I32       |      I32      |      2^20      | 2097152 |     1     |  10.959 ms |       0.16% |  10.937 ms |       0.10% | -22.613 us |  -0.21% |   FAST   |
|      I64      |      I32       |      I32      |      2^24      | 2097152 |     1     |  13.226 ms |       0.10% |  13.216 ms |       0.12% |  -9.905 us |  -0.07% |   SAME   |
|      I64      |      I32       |      I32      |      2^28      | 2097152 |     1     |  49.094 ms |       0.14% |  49.104 ms |       0.15% |  10.293 us |   0.02% |   SAME   |
|      F32      |      I32       |      I32      |      2^16      |   32    |   0.201   |  13.142 us |       7.96% |  13.312 us |       6.70% |   0.170 us |   1.30% |   SAME   |
|      F32      |      I32       |      I32      |      2^20      |   32    |   0.201   |  17.949 us |       8.88% |  18.303 us |       7.92% |   0.354 us |   1.97% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      |   32    |   0.201   |  82.553 us |       0.99% |  81.590 us |       1.23% |  -0.962 us |  -1.17% |   FAST   |
|      F32      |      I32       |      I32      |      2^28      |   32    |   0.201   |   1.114 ms |       0.15% |   1.102 ms |       0.21% | -11.573 us |  -1.04% |   FAST   |
|      F32      |      I32       |      I32      |      2^16      |   128   |   0.201   |  12.428 us |       7.89% |  13.011 us |       5.58% |   0.583 us |   4.69% |   SAME   |
|      F32      |      I32       |      I32      |      2^20      |   128   |   0.201   |  18.007 us |       4.76% |  18.515 us |       5.60% |   0.509 us |   2.82% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      |   128   |   0.201   |  81.053 us |       1.64% |  80.855 us |       0.39% |  -0.199 us |  -0.25% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      |   128   |   0.201   |   1.155 ms |       0.14% |   1.142 ms |       0.20% | -12.283 us |  -1.06% |   FAST   |
|      F32      |      I32       |      I32      |      2^16      |  2048   |   0.201   |  19.342 us |       2.98% |  19.448 us |       0.86% |   0.106 us |   0.55% |   SAME   |
|      F32      |      I32       |      I32      |      2^20      |  2048   |   0.201   |  33.639 us |       4.49% |  33.470 us |       3.34% |  -0.168 us |  -0.50% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      |  2048   |   0.201   | 318.562 us |       1.22% | 318.161 us |       1.18% |  -0.402 us |  -0.13% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      |  2048   |   0.201   |   4.688 ms |       0.24% |   4.658 ms |       0.25% | -30.016 us |  -0.64% |   FAST   |
|      F32      |      I32       |      I32      |      2^16      | 2097152 |   0.201   |   7.815 ms |       0.11% |   7.801 ms |       0.22% | -14.245 us |  -0.18% |   FAST   |
|      F32      |      I32       |      I32      |      2^20      | 2097152 |   0.201   |   9.783 ms |       0.18% |   9.784 ms |       0.18% |   0.806 us |   0.01% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      | 2097152 |   0.201   |  10.187 ms |       0.14% |  10.185 ms |       0.14% |  -1.340 us |  -0.01% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      | 2097152 |   0.201   |  15.906 ms |       0.45% |  15.906 ms |       0.45% |  -0.266 us |  -0.00% |   SAME   |
|      F32      |      I32       |      I32      |      2^16      |   32    |     1     |  13.080 us |       7.39% |  13.345 us |       4.38% |   0.264 us |   2.02% |   SAME   |
|      F32      |      I32       |      I32      |      2^20      |   32    |     1     |  16.377 us |       5.81% |  17.058 us |       4.48% |   0.681 us |   4.16% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      |   32    |     1     |  66.071 us |       1.76% |  66.494 us |       1.02% |   0.424 us |   0.64% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      |   32    |     1     | 708.371 us |       1.46% | 707.886 us |       1.52% |  -0.485 us |  -0.07% |   SAME   |
|      F32      |      I32       |      I32      |      2^16      |   128   |     1     |  12.576 us |       8.13% |  12.995 us |       6.03% |   0.419 us |   3.33% |   SAME   |
|      F32      |      I32       |      I32      |      2^20      |   128   |     1     |  16.298 us |       6.73% |  16.641 us |       7.13% |   0.344 us |   2.11% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      |   128   |     1     |  65.100 us |       2.00% |  65.604 us |       2.06% |   0.504 us |   0.77% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      |   128   |     1     | 722.775 us |       3.20% | 703.315 us |       1.72% | -19.460 us |  -2.69% |   FAST   |
|      F32      |      I32       |      I32      |      2^16      |  2048   |     1     |  21.533 us |       2.40% |  21.564 us |       1.10% |   0.031 us |   0.15% |   SAME   |
|      F32      |      I32       |      I32      |      2^20      |  2048   |     1     |  38.792 us |       3.75% |  39.165 us |       3.09% |   0.374 us |   0.96% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      |  2048   |     1     | 287.144 us |       1.51% | 287.686 us |       1.58% |   0.542 us |   0.19% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      |  2048   |     1     |   4.561 ms |       0.19% |   4.561 ms |       0.19% |  -0.042 us |  -0.00% |   SAME   |
|      F32      |      I32       |      I32      |      2^16      | 2097152 |     1     |   7.803 ms |       0.16% |   7.800 ms |       0.17% |  -2.775 us |  -0.04% |   SAME   |
|      F32      |      I32       |      I32      |      2^20      | 2097152 |     1     |   9.934 ms |       0.15% |   9.934 ms |       0.15% |  -0.271 us |  -0.00% |   SAME   |
|      F32      |      I32       |      I32      |      2^24      | 2097152 |     1     |  12.185 ms |       0.22% |  12.184 ms |       0.21% |  -0.696 us |  -0.01% |   SAME   |
|      F32      |      I32       |      I32      |      2^28      | 2097152 |     1     |  47.115 ms |       0.05% |  47.120 ms |       0.06% |   5.469 us |   0.01% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      |   32    |   0.201   |  12.979 us |       7.57% |  13.423 us |       3.03% |   0.444 us |   3.42% |   SLOW   |
|      F64      |      I32       |      I32      |      2^20      |   32    |   0.201   |  20.597 us |       5.31% |  20.905 us |       5.54% |   0.308 us |   1.50% |   SAME   |
|      F64      |      I32       |      I32      |      2^24      |   32    |   0.201   | 122.294 us |       1.47% | 122.766 us |       1.03% |   0.472 us |   0.39% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      |   32    |   0.201   |   1.783 ms |       0.06% |   1.784 ms |       0.06% |   0.411 us |   0.02% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      |   128   |   0.201   |  12.691 us |       7.95% |  12.839 us |       7.26% |   0.148 us |   1.17% |   SAME   |
|      F64      |      I32       |      I32      |      2^20      |   128   |   0.201   |  20.663 us |       5.01% |  20.994 us |       4.46% |   0.332 us |   1.61% |   SAME   |
|      F64      |      I32       |      I32      |      2^24      |   128   |   0.201   | 120.541 us |       1.38% | 120.913 us |       0.99% |   0.371 us |   0.31% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      |   128   |   0.201   |   1.824 ms |       0.03% |   1.824 ms |       0.01% |   0.114 us |   0.01% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      |  2048   |   0.201   |  17.327 us |       3.83% |  17.450 us |       2.87% |   0.123 us |   0.71% |   SAME   |
|      F64      |      I32       |      I32      |      2^20      |  2048   |   0.201   |  37.537 us |       3.86% |  37.826 us |       3.37% |   0.289 us |   0.77% |   SAME   |
|      F64      |      I32       |      I32      |      2^24      |  2048   |   0.201   | 344.432 us |       1.23% | 344.659 us |       1.19% |   0.227 us |   0.07% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      |  2048   |   0.201   |   5.335 ms |       0.39% |   5.336 ms |       0.40% |   0.340 us |   0.01% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      | 2097152 |   0.201   |   7.777 ms |       0.06% |   7.778 ms |       0.06% |   1.626 us |   0.02% |   SAME   |
|      F64      |      I32       |      I32      |      2^20      | 2097152 |   0.201   |  10.776 ms |       0.15% |  10.776 ms |       0.14% |   0.122 us |   0.00% |   SAME   |
|      F64      |      I32       |      I32      |      2^24      | 2097152 |   0.201   |  11.247 ms |       0.24% |  11.246 ms |       0.23% |  -1.653 us |  -0.01% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      | 2097152 |   0.201   |  16.746 ms |       0.34% |  16.749 ms |       0.32% |   2.307 us |   0.01% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      |   32    |     1     |  13.306 us |       6.08% |  13.407 us |       5.90% |   0.102 us |   0.76% |   SAME   |
|      F64      |      I32       |      I32      |      2^20      |   32    |     1     |  18.388 us |       4.57% |  19.319 us |       2.62% |   0.931 us |   5.07% |   SLOW   |
|      F64      |      I32       |      I32      |      2^24      |   32    |     1     | 116.407 us |       1.92% | 116.868 us |       1.79% |   0.462 us |   0.40% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      |   32    |     1     |   1.312 ms |       3.16% |   1.311 ms |       2.77% |  -1.107 us |  -0.08% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      |   128   |     1     |  13.480 us |       5.43% |  13.486 us |       4.69% |   0.006 us |   0.05% |   SAME   |
|      F64      |      I32       |      I32      |      2^20      |   128   |     1     |  21.541 us |       6.37% |  21.946 us |       4.06% |   0.406 us |   1.88% |   SAME   |
|      F64      |      I32       |      I32      |      2^24      |   128   |     1     | 115.810 us |       2.17% | 116.324 us |       2.05% |   0.514 us |   0.44% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      |   128   |     1     |   1.347 ms |       2.65% |   1.344 ms |       2.83% |  -2.439 us |  -0.18% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      |  2048   |     1     |  19.717 us |       2.66% |  19.707 us |       2.15% |  -0.010 us |  -0.05% |   SAME   |
|      F64      |      I32       |      I32      |      2^20      |  2048   |     1     |  41.367 us |       3.23% |  41.450 us |       3.11% |   0.082 us |   0.20% |   SAME   |
|      F64      |      I32       |      I32      |      2^24      |  2048   |     1     | 322.109 us |       0.93% | 323.050 us |       0.85% |   0.941 us |   0.29% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      |  2048   |     1     |   4.788 ms |       0.27% |   4.789 ms |       0.29% |   1.303 us |   0.03% |   SAME   |
|      F64      |      I32       |      I32      |      2^16      | 2097152 |     1     |   7.756 ms |       0.05% |   7.774 ms |       0.18% |  17.880 us |   0.23% |   SLOW   |
|      F64      |      I32       |      I32      |      2^20      | 2097152 |     1     |  10.931 ms |       0.10% |  10.959 ms |       0.17% |  28.219 us |   0.26% |   SLOW   |
|      F64      |      I32       |      I32      |      2^24      | 2097152 |     1     |  13.178 ms |       0.11% |  13.182 ms |       0.09% |   4.132 us |   0.03% |   SAME   |
|      F64      |      I32       |      I32      |      2^28      | 2097152 |     1     |  48.765 ms |       0.08% |  48.772 ms |       0.04% |   7.105 us |   0.01% |   SAME   |

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
Status: Done
Development

Successfully merging this pull request may close these issues.

4 participants