Skip to content

Extracted Segmented Reduce kernels into NVRTC compilable header - #3727

Merged
shwina merged 4 commits into
NVIDIA:mainfrom
oleksandr-pavlyk:extract-segmented-reduce-scan-kernels-into-nvrtc-palatable-headers
Feb 7, 2025
Merged

shwina merged 4 commits into
NVIDIA:mainfrom
oleksandr-pavlyk:extract-segmented-reduce-scan-kernels-into-nvrtc-palatable-headers

Conversation

@oleksandr-pavlyk

@oleksandr-pavlyk oleksandr-pavlyk commented Feb 6, 2025 •

Copy link
Copy Markdown
Contributor

Moved definition of DeviceSegmentedReduceKernel out to NVRTC compilable header file

Moved the definition from "cub/device/dispatch/dispatch_reduce.cuh" to new file "cub/device/dispatch/kernels/segmented_reduce.cuh"

Added compilation of the new file to "cub/test/catch2_test_nvrtc.cu"

Needed to remove friended operator<<() overload for std::ostream from "arg_index_input_iterator.cuh", since this header is needed by DeviceSegmentedReduceKernel implementation, and use of std::ostream breaks compilation by nvrtc/

Description

closes gh-3712

Checklist

  • New or existing tests cover these changes.
  • The documentation is up to date with these changes.

…le header file

Moved the definition from cub/device/dispatch/dispatch_reduce.cuh to
new file cub/device/dispatch/kernels/segmented_reduce.cuh

Added compilation of the new file to cub/test/catch2_test_nvrtc.cu

Needed to remove friended operator<< overload for std::ostream from
arg_index_input_iterator.cuh, since this header is needed by
DeviceSegmentedReduceKernel implementation, and use of std::ostream
breaks compilation by nvrtc
@github-actions

github-actions Bot commented Feb 7, 2025

Copy link
Copy Markdown
Contributor
🟩 CI finished in 1h 39m: Pass: 100%/90 | Total: 2d 15h | Avg: 42m 23s | Max: 1h 13m | Hits: 74%/132401
  • 🟩 cub: Pass: 100%/44 | Total: 1d 15h | Avg: 53m 53s | Max: 1h 13m | Hits: 68%/52496

    🟩 cpu
      🟩 amd64              Pass: 100%/42  | Total:  1d 13h | Avg: 53m 29s | Max:  1h 13m | Hits:  68%/50056 
      🟩 arm64              Pass: 100%/2   | Total:  2h 04m | Avg:  1h 02m | Max:  1h 03m | Hits:  67%/2440  
    🟩 ctk
      🟩 12.0               Pass: 100%/5   | Total:  4h 59m | Avg: 59m 49s | Max:  1h 02m | Hits:  58%/5934  
      🟩 12.5               Pass: 100%/2   | Total:  2h 16m | Avg:  1h 08m | Max:  1h 09m | Hits:  67%/2258  
      🟩 12.8               Pass: 100%/37  | Total:  1d 08h | Avg: 52m 19s | Max:  1h 13m | Hits:  70%/44304 
    🟩 cudacxx
      🟩 ClangCUDA18        Pass: 100%/2   | Total:  2h 00m | Avg:  1h 00m | Max:  1h 00m | Hits:  73%/2112  
      🟩 nvcc12.0           Pass: 100%/5   | Total:  4h 59m | Avg: 59m 49s | Max:  1h 02m | Hits:  58%/5934  
      🟩 nvcc12.5           Pass: 100%/2   | Total:  2h 16m | Avg:  1h 08m | Max:  1h 09m | Hits:  67%/2258  
      🟩 nvcc12.8           Pass: 100%/35  | Total:  1d 06h | Avg: 51m 52s | Max:  1h 13m | Hits:  70%/42192 
    🟩 cudacxx_family
      🟩 ClangCUDA          Pass: 100%/2   | Total:  2h 00m | Avg:  1h 00m | Max:  1h 00m | Hits:  73%/2112  
      🟩 nvcc               Pass: 100%/42  | Total:  1d 13h | Avg: 53m 35s | Max:  1h 13m | Hits:  68%/50384 
    🟩 cxx
      🟩 Clang14            Pass: 100%/4   | Total:  3h 55m | Avg: 58m 48s | Max:  1h 00m | Hits:  67%/4888  
      🟩 Clang15            Pass: 100%/2   | Total:  1h 58m | Avg: 59m 09s | Max: 59m 57s | Hits:  67%/2440  
      🟩 Clang16            Pass: 100%/2   | Total:  1h 49m | Avg: 54m 47s | Max: 55m 58s | Hits:  67%/2440  
      🟩 Clang17            Pass: 100%/2   | Total:  1h 52m | Avg: 56m 20s | Max: 59m 29s | Hits:  67%/2440  
      🟩 Clang18            Pass: 100%/7   | Total:  5h 44m | Avg: 49m 10s | Max:  1h 01m | Hits:  78%/8212  
      🟩 GCC7               Pass: 100%/2   | Total:  1h 57m | Avg: 58m 48s | Max:  1h 01m | Hits:  67%/2444  
      🟩 GCC8               Pass: 100%/1   | Total: 57m 48s | Avg: 57m 48s | Max: 57m 48s | Hits:  67%/1222  
      🟩 GCC9               Pass: 100%/2   | Total:  1h 53m | Avg: 56m 35s | Max: 57m 07s | Hits:  67%/2444  
      🟩 GCC10              Pass: 100%/2   | Total:  1h 57m | Avg: 58m 53s | Max:  1h 02m | Hits:  67%/2444  
      🟩 GCC11              Pass: 100%/2   | Total:  1h 55m | Avg: 57m 58s | Max:  1h 01m | Hits:  67%/2440  
      🟩 GCC12              Pass: 100%/2   | Total:  1h 54m | Avg: 57m 25s | Max: 59m 19s | Hits:  67%/2440  
      🟩 GCC13              Pass: 100%/10  | Total:  6h 37m | Avg: 39m 46s | Max:  1h 13m | Hits:  83%/12200 
      🟩 MSVC14.29          Pass: 100%/2   | Total:  2h 14m | Avg:  1h 07m | Max:  1h 11m | Hits:  14%/2092  
      🟩 MSVC14.42          Pass: 100%/2   | Total:  2h 26m | Avg:  1h 13m | Max:  1h 13m | Hits:  14%/2092  
      🟩 NVHPC24.7          Pass: 100%/2   | Total:  2h 16m | Avg:  1h 08m | Max:  1h 09m | Hits:  67%/2258  
    🟩 cxx_family
      🟩 Clang              Pass: 100%/17  | Total: 15h 20m | Avg: 54m 07s | Max:  1h 01m | Hits:  72%/20420 
      🟩 GCC                Pass: 100%/21  | Total: 17h 14m | Avg: 49m 16s | Max:  1h 13m | Hits:  75%/25634 
      🟩 MSVC               Pass: 100%/4   | Total:  4h 40m | Avg:  1h 10m | Max:  1h 13m | Hits:  14%/4184  
      🟩 NVHPC              Pass: 100%/2   | Total:  2h 16m | Avg:  1h 08m | Max:  1h 09m | Hits:  67%/2258  
    🟩 gpu
      🟩 h100               Pass: 100%/2   | Total: 51m 12s | Avg: 25m 36s | Max: 26m 14s | Hits:  83%/2440  
      🟩 rtx2080            Pass: 100%/34  | Total:  1d 10h | Avg:  1h 00m | Max:  1h 13m | Hits:  62%/40296 
      🟩 rtxa6000           Pass: 100%/8   | Total:  4h 15m | Avg: 31m 57s | Max:  1h 01m | Hits:  91%/9760  
    🟩 jobs
      🟩 Build              Pass: 100%/37  | Total:  1d 12h | Avg: 59m 47s | Max:  1h 13m | Hits:  62%/43956 
      🟩 DeviceLaunch       Pass: 100%/1   | Total: 23m 21s | Avg: 23m 21s | Max: 23m 21s | Hits:  99%/1220  
      🟩 GraphCapture       Pass: 100%/1   | Total: 16m 45s | Avg: 16m 45s | Max: 16m 45s | Hits:  99%/1220  
      🟩 HostLaunch         Pass: 100%/3   | Total:  1h 15m | Avg: 25m 08s | Max: 26m 14s | Hits:  99%/3660  
      🟩 TestGPU            Pass: 100%/2   | Total: 43m 53s | Avg: 21m 56s | Max: 23m 10s | Hits:  99%/2440  
    🟩 sm
      🟩 90                 Pass: 100%/2   | Total: 51m 12s | Avg: 25m 36s | Max: 26m 14s | Hits:  83%/2440  
      🟩 90;90a;100         Pass: 100%/1   | Total:  1h 13m | Avg:  1h 13m | Max:  1h 13m | Hits:  67%/1220  
    🟩 std
      🟩 17                 Pass: 100%/20  | Total: 20h 00m | Avg:  1h 00m | Max:  1h 13m | Hits:  60%/23639 
      🟩 20                 Pass: 100%/24  | Total: 19h 31m | Avg: 48m 48s | Max:  1h 13m | Hits:  75%/28857 
    
  • 🟩 thrust: Pass: 100%/43 | Total: 23h 24m | Avg: 32m 40s | Max: 1h 02m | Hits: 78%/79625

    🟩 cmake_options
      🟩 -DTHRUST_DISPATCH_TYPE=Force32bit Pass: 100%/2   | Total: 39m 23s | Avg: 19m 41s | Max: 28m 00s | Hits:  89%/3706  
    🟩 cpu
      🟩 amd64              Pass: 100%/41  | Total: 22h 26m | Avg: 32m 49s | Max:  1h 02m | Hits:  78%/75920 
      🟩 arm64              Pass: 100%/2   | Total: 58m 54s | Avg: 29m 27s | Max: 31m 28s | Hits:  78%/3705  
    🟩 ctk
      🟩 12.0               Pass: 100%/5   | Total:  3h 04m | Avg: 36m 51s | Max: 56m 18s | Hits:  73%/9256  
      🟩 12.5               Pass: 100%/2   | Total:  1h 44m | Avg: 52m 15s | Max: 52m 38s | Hits:  73%/3704  
      🟩 12.8               Pass: 100%/36  | Total: 18h 36m | Avg: 31m 00s | Max:  1h 02m | Hits:  79%/66665 
    🟩 cudacxx
      🟩 ClangCUDA18        Pass: 100%/2   | Total: 55m 26s | Avg: 27m 43s | Max: 28m 44s | Hits:  78%/3704  
      🟩 nvcc12.0           Pass: 100%/5   | Total:  3h 04m | Avg: 36m 51s | Max: 56m 18s | Hits:  73%/9256  
      🟩 nvcc12.5           Pass: 100%/2   | Total:  1h 44m | Avg: 52m 15s | Max: 52m 38s | Hits:  73%/3704  
      🟩 nvcc12.8           Pass: 100%/34  | Total: 17h 40m | Avg: 31m 11s | Max:  1h 02m | Hits:  79%/62961 
    🟩 cudacxx_family
      🟩 ClangCUDA          Pass: 100%/2   | Total: 55m 26s | Avg: 27m 43s | Max: 28m 44s | Hits:  78%/3704  
      🟩 nvcc               Pass: 100%/41  | Total: 22h 29m | Avg: 32m 54s | Max:  1h 02m | Hits:  78%/75921 
    🟩 cxx
      🟩 Clang14            Pass: 100%/4   | Total:  2h 02m | Avg: 30m 44s | Max: 32m 49s | Hits:  78%/7408  
      🟩 Clang15            Pass: 100%/2   | Total:  1h 02m | Avg: 31m 01s | Max: 32m 22s | Hits:  78%/3704  
      🟩 Clang16            Pass: 100%/2   | Total:  1h 02m | Avg: 31m 19s | Max: 32m 59s | Hits:  78%/3704  
      🟩 Clang17            Pass: 100%/2   | Total:  1h 00m | Avg: 30m 22s | Max: 31m 22s | Hits:  78%/3704  
      🟩 Clang18            Pass: 100%/7   | Total:  2h 44m | Avg: 23m 27s | Max: 32m 21s | Hits:  84%/12964 
      🟩 GCC7               Pass: 100%/2   | Total:  1h 03m | Avg: 31m 59s | Max: 33m 25s | Hits:  78%/3706  
      🟩 GCC8               Pass: 100%/1   | Total: 30m 17s | Avg: 30m 17s | Max: 30m 17s | Hits:  78%/1853  
      🟩 GCC9               Pass: 100%/2   | Total:  1h 07m | Avg: 33m 52s | Max: 35m 42s | Hits:  78%/3706  
      🟩 GCC10              Pass: 100%/2   | Total:  1h 11m | Avg: 35m 30s | Max: 35m 40s | Hits:  78%/3706  
      🟩 GCC11              Pass: 100%/2   | Total:  1h 05m | Avg: 32m 39s | Max: 33m 30s | Hits:  78%/3706  
      🟩 GCC12              Pass: 100%/2   | Total:  1h 08m | Avg: 34m 09s | Max: 35m 39s | Hits:  78%/3706  
      🟩 GCC13              Pass: 100%/8   | Total:  3h 13m | Avg: 24m 14s | Max: 36m 03s | Hits:  86%/14824 
      🟩 MSVC14.29          Pass: 100%/2   | Total:  1h 56m | Avg: 58m 12s | Max:  1h 00m | Hits:  53%/3692  
      🟩 MSVC14.42          Pass: 100%/3   | Total:  2h 30m | Avg: 50m 19s | Max:  1h 02m | Hits:  58%/5538  
      🟩 NVHPC24.7          Pass: 100%/2   | Total:  1h 44m | Avg: 52m 15s | Max: 52m 38s | Hits:  73%/3704  
    🟩 cxx_family
      🟩 Clang              Pass: 100%/17  | Total:  7h 52m | Avg: 27m 47s | Max: 32m 59s | Hits:  81%/31484 
      🟩 GCC                Pass: 100%/19  | Total:  9h 20m | Avg: 29m 30s | Max: 36m 03s | Hits:  82%/35207 
      🟩 MSVC               Pass: 100%/5   | Total:  4h 27m | Avg: 53m 28s | Max:  1h 02m | Hits:  56%/9230  
      🟩 NVHPC              Pass: 100%/2   | Total:  1h 44m | Avg: 52m 15s | Max: 52m 38s | Hits:  73%/3704  
    🟩 gpu
      🟩 rtx2080            Pass: 100%/33  | Total: 19h 27m | Avg: 35m 22s | Max:  1h 00m | Hits:  76%/61112 
      🟩 rtx4090            Pass: 100%/10  | Total:  3h 57m | Avg: 23m 46s | Max:  1h 02m | Hits:  85%/18513 
    🟩 jobs
      🟩 Build              Pass: 100%/37  | Total: 22h 05m | Avg: 35m 48s | Max:  1h 02m | Hits:  75%/68516 
      🟩 TestCPU            Pass: 100%/3   | Total: 47m 00s | Avg: 15m 40s | Max: 31m 23s | Hits:  89%/5551  
      🟩 TestGPU            Pass: 100%/3   | Total: 32m 50s | Avg: 10m 56s | Max: 11m 23s | Hits:  99%/5558  
    🟩 sm
      🟩 90;90a;100         Pass: 100%/1   | Total: 36m 03s | Avg: 36m 03s | Max: 36m 03s | Hits:  78%/1853  
    🟩 std
      🟩 17                 Pass: 100%/20  | Total: 12h 16m | Avg: 36m 50s | Max:  1h 00m | Hits:  74%/37031 
      🟩 20                 Pass: 100%/21  | Total: 10h 28m | Avg: 29m 56s | Max:  1h 02m | Hits:  80%/38888 
    
  • 🟩 cccl_c_parallel: Pass: 100%/2 | Total: 8m 11s | Avg: 4m 05s | Max: 5m 40s | Hits: 98%/280

    🟩 cpu
      🟩 amd64              Pass: 100%/2   | Total:  8m 11s | Avg:  4m 05s | Max:  5m 40s | Hits:  98%/280   
    🟩 ctk
      🟩 12.8               Pass: 100%/2   | Total:  8m 11s | Avg:  4m 05s | Max:  5m 40s | Hits:  98%/280   
    🟩 cudacxx
      🟩 nvcc12.8           Pass: 100%/2   | Total:  8m 11s | Avg:  4m 05s | Max:  5m 40s | Hits:  98%/280   
    🟩 cudacxx_family
      🟩 nvcc               Pass: 100%/2   | Total:  8m 11s | Avg:  4m 05s | Max:  5m 40s | Hits:  98%/280   
    🟩 cxx
      🟩 GCC13              Pass: 100%/2   | Total:  8m 11s | Avg:  4m 05s | Max:  5m 40s | Hits:  98%/280   
    🟩 cxx_family
      🟩 GCC                Pass: 100%/2   | Total:  8m 11s | Avg:  4m 05s | Max:  5m 40s | Hits:  98%/280   
    🟩 gpu
      🟩 rtx2080            Pass: 100%/2   | Total:  8m 11s | Avg:  4m 05s | Max:  5m 40s | Hits:  98%/280   
    🟩 jobs
      🟩 Build              Pass: 100%/1   | Total:  2m 31s | Avg:  2m 31s | Max:  2m 31s | Hits:  97%/140   
      🟩 Test               Pass: 100%/1   | Total:  5m 40s | Avg:  5m 40s | Max:  5m 40s | Hits:  98%/140   
    
  • 🟩 python: Pass: 100%/1 | Total: 30m 44s | Avg: 30m 44s | Max: 30m 44s

    🟩 cpu
      🟩 amd64              Pass: 100%/1   | Total: 30m 44s | Avg: 30m 44s | Max: 30m 44s
    🟩 ctk
      🟩 12.8               Pass: 100%/1   | Total: 30m 44s | Avg: 30m 44s | Max: 30m 44s
    🟩 cudacxx
      🟩 nvcc12.8           Pass: 100%/1   | Total: 30m 44s | Avg: 30m 44s | Max: 30m 44s
    🟩 cudacxx_family
      🟩 nvcc               Pass: 100%/1   | Total: 30m 44s | Avg: 30m 44s | Max: 30m 44s
    🟩 cxx
      🟩 GCC13              Pass: 100%/1   | Total: 30m 44s | Avg: 30m 44s | Max: 30m 44s
    🟩 cxx_family
      🟩 GCC                Pass: 100%/1   | Total: 30m 44s | Avg: 30m 44s | Max: 30m 44s
    🟩 gpu
      🟩 rtx2080            Pass: 100%/1   | Total: 30m 44s | Avg: 30m 44s | Max: 30m 44s
    🟩 jobs
      🟩 Test               Pass: 100%/1   | Total: 30m 44s | Avg: 30m 44s | Max: 30m 44s
    

πŸ‘ƒ 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

πŸƒβ€ 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

@shwina shwina left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM pending @bernhardmgruber's comments!

@github-actions

github-actions Bot commented Feb 7, 2025

Copy link
Copy Markdown
Contributor
🟩 CI finished in 1h 39m: Pass: 100%/90 | Total: 2d 16h | Avg: 42m 47s | Max: 1h 20m | Hits: 74%/132401
  • 🟩 cub: Pass: 100%/44 | Total: 1d 16h | Avg: 54m 57s | Max: 1h 20m | Hits: 68%/52496

    🟩 cpu
      🟩 amd64              Pass: 100%/42  | Total:  1d 14h | Avg: 54m 42s | Max:  1h 20m | Hits:  68%/50056 
      🟩 arm64              Pass: 100%/2   | Total:  2h 00m | Avg:  1h 00m | Max:  1h 00m | Hits:  67%/2440  
    🟩 ctk
      🟩 12.0               Pass: 100%/5   | Total:  5h 15m | Avg:  1h 03m | Max:  1h 09m | Hits:  58%/5934  
      🟩 12.5               Pass: 100%/2   | Total:  2h 17m | Avg:  1h 08m | Max:  1h 12m | Hits:  67%/2258  
      🟩 12.8               Pass: 100%/37  | Total:  1d 08h | Avg: 53m 07s | Max:  1h 20m | Hits:  70%/44304 
    🟩 cudacxx
      🟩 ClangCUDA18        Pass: 100%/2   | Total:  2h 04m | Avg:  1h 02m | Max:  1h 04m | Hits:  73%/2112  
      🟩 nvcc12.0           Pass: 100%/5   | Total:  5h 15m | Avg:  1h 03m | Max:  1h 09m | Hits:  58%/5934  
      🟩 nvcc12.5           Pass: 100%/2   | Total:  2h 17m | Avg:  1h 08m | Max:  1h 12m | Hits:  67%/2258  
      🟩 nvcc12.8           Pass: 100%/35  | Total:  1d 06h | Avg: 52m 35s | Max:  1h 20m | Hits:  70%/42192 
    🟩 cudacxx_family
      🟩 ClangCUDA          Pass: 100%/2   | Total:  2h 04m | Avg:  1h 02m | Max:  1h 04m | Hits:  73%/2112  
      🟩 nvcc               Pass: 100%/42  | Total:  1d 14h | Avg: 54m 35s | Max:  1h 20m | Hits:  68%/50384 
    🟩 cxx
      🟩 Clang14            Pass: 100%/4   | Total:  3h 56m | Avg: 59m 10s | Max:  1h 01m | Hits:  67%/4888  
      🟩 Clang15            Pass: 100%/2   | Total:  1h 57m | Avg: 58m 36s | Max:  1h 02m | Hits:  67%/2440  
      🟩 Clang16            Pass: 100%/2   | Total:  1h 54m | Avg: 57m 11s | Max: 57m 20s | Hits:  67%/2440  
      🟩 Clang17            Pass: 100%/2   | Total:  2h 05m | Avg:  1h 02m | Max:  1h 02m | Hits:  67%/2440  
      🟩 Clang18            Pass: 100%/7   | Total:  5h 45m | Avg: 49m 24s | Max:  1h 04m | Hits:  78%/8212  
      🟩 GCC7               Pass: 100%/2   | Total:  2h 03m | Avg:  1h 01m | Max:  1h 02m | Hits:  67%/2444  
      🟩 GCC8               Pass: 100%/1   | Total: 57m 10s | Avg: 57m 10s | Max: 57m 10s | Hits:  67%/1222  
      🟩 GCC9               Pass: 100%/2   | Total:  2h 07m | Avg:  1h 03m | Max:  1h 04m | Hits:  67%/2444  
      🟩 GCC10              Pass: 100%/2   | Total:  1h 54m | Avg: 57m 08s | Max: 58m 55s | Hits:  67%/2444  
      🟩 GCC11              Pass: 100%/2   | Total:  2h 04m | Avg:  1h 02m | Max:  1h 02m | Hits:  67%/2440  
      🟩 GCC12              Pass: 100%/2   | Total:  1h 53m | Avg: 56m 56s | Max: 57m 12s | Hits:  67%/2440  
      🟩 GCC13              Pass: 100%/10  | Total:  6h 24m | Avg: 38m 24s | Max:  1h 12m | Hits:  83%/12200 
      🟩 MSVC14.29          Pass: 100%/2   | Total:  2h 21m | Avg:  1h 10m | Max:  1h 11m | Hits:  14%/2092  
      🟩 MSVC14.42          Pass: 100%/2   | Total:  2h 35m | Avg:  1h 17m | Max:  1h 20m | Hits:  14%/2092  
      🟩 NVHPC24.7          Pass: 100%/2   | Total:  2h 17m | Avg:  1h 08m | Max:  1h 12m | Hits:  67%/2258  
    🟩 cxx_family
      🟩 Clang              Pass: 100%/17  | Total: 15h 39m | Avg: 55m 14s | Max:  1h 04m | Hits:  72%/20420 
      🟩 GCC                Pass: 100%/21  | Total: 17h 25m | Avg: 49m 46s | Max:  1h 12m | Hits:  75%/25634 
      🟩 MSVC               Pass: 100%/4   | Total:  4h 56m | Avg:  1h 14m | Max:  1h 20m | Hits:  14%/4184  
      🟩 NVHPC              Pass: 100%/2   | Total:  2h 17m | Avg:  1h 08m | Max:  1h 12m | Hits:  67%/2258  
    🟩 gpu
      🟩 h100               Pass: 100%/2   | Total: 49m 14s | Avg: 24m 37s | Max: 25m 14s | Hits:  83%/2440  
      🟩 rtx2080            Pass: 100%/34  | Total:  1d 11h | Avg:  1h 02m | Max:  1h 20m | Hits:  62%/40296 
      🟩 rtxa6000           Pass: 100%/8   | Total:  4h 09m | Avg: 31m 13s | Max: 59m 01s | Hits:  91%/9760  
    🟩 jobs
      🟩 Build              Pass: 100%/37  | Total:  1d 13h | Avg:  1h 01m | Max:  1h 20m | Hits:  62%/43956 
      🟩 DeviceLaunch       Pass: 100%/1   | Total: 21m 49s | Avg: 21m 49s | Max: 21m 49s | Hits:  99%/1220  
      🟩 GraphCapture       Pass: 100%/1   | Total: 17m 38s | Avg: 17m 38s | Max: 17m 38s | Hits:  99%/1220  
      🟩 HostLaunch         Pass: 100%/3   | Total:  1h 13m | Avg: 24m 38s | Max: 25m 26s | Hits:  99%/3660  
      🟩 TestGPU            Pass: 100%/2   | Total: 46m 03s | Avg: 23m 01s | Max: 23m 43s | Hits:  99%/2440  
    🟩 sm
      🟩 90                 Pass: 100%/2   | Total: 49m 14s | Avg: 24m 37s | Max: 25m 14s | Hits:  83%/2440  
      🟩 90;90a;100         Pass: 100%/1   | Total:  1h 12m | Avg:  1h 12m | Max:  1h 12m | Hits:  67%/1220  
    🟩 std
      🟩 17                 Pass: 100%/20  | Total: 20h 35m | Avg:  1h 01m | Max:  1h 15m | Hits:  60%/23639 
      🟩 20                 Pass: 100%/24  | Total: 19h 42m | Avg: 49m 15s | Max:  1h 20m | Hits:  75%/28857 
    
  • 🟩 thrust: Pass: 100%/43 | Total: 23h 13m | Avg: 32m 23s | Max: 56m 48s | Hits: 78%/79625

    🟩 cmake_options
      🟩 -DTHRUST_DISPATCH_TYPE=Force32bit Pass: 100%/2   | Total: 38m 41s | Avg: 19m 20s | Max: 27m 11s | Hits:  89%/3706  
    🟩 cpu
      🟩 amd64              Pass: 100%/41  | Total: 22h 14m | Avg: 32m 32s | Max: 56m 48s | Hits:  78%/75920 
      🟩 arm64              Pass: 100%/2   | Total: 59m 01s | Avg: 29m 30s | Max: 31m 43s | Hits:  78%/3705  
    🟩 ctk
      🟩 12.0               Pass: 100%/5   | Total:  3h 01m | Avg: 36m 22s | Max: 56m 48s | Hits:  73%/9256  
      🟩 12.5               Pass: 100%/2   | Total:  1h 44m | Avg: 52m 05s | Max: 53m 19s | Hits:  73%/3704  
      🟩 12.8               Pass: 100%/36  | Total: 18h 27m | Avg: 30m 45s | Max: 56m 41s | Hits:  79%/66665 
    🟩 cudacxx
      🟩 ClangCUDA18        Pass: 100%/2   | Total: 57m 48s | Avg: 28m 54s | Max: 30m 11s | Hits:  78%/3704  
      🟩 nvcc12.0           Pass: 100%/5   | Total:  3h 01m | Avg: 36m 22s | Max: 56m 48s | Hits:  73%/9256  
      🟩 nvcc12.5           Pass: 100%/2   | Total:  1h 44m | Avg: 52m 05s | Max: 53m 19s | Hits:  73%/3704  
      🟩 nvcc12.8           Pass: 100%/34  | Total: 17h 29m | Avg: 30m 51s | Max: 56m 41s | Hits:  79%/62961 
    🟩 cudacxx_family
      🟩 ClangCUDA          Pass: 100%/2   | Total: 57m 48s | Avg: 28m 54s | Max: 30m 11s | Hits:  78%/3704  
      🟩 nvcc               Pass: 100%/41  | Total: 22h 15m | Avg: 32m 34s | Max: 56m 48s | Hits:  78%/75921 
    🟩 cxx
      🟩 Clang14            Pass: 100%/4   | Total:  2h 02m | Avg: 30m 41s | Max: 31m 21s | Hits:  78%/7408  
      🟩 Clang15            Pass: 100%/2   | Total:  1h 05m | Avg: 32m 42s | Max: 33m 15s | Hits:  78%/3704  
      🟩 Clang16            Pass: 100%/2   | Total:  1h 03m | Avg: 31m 53s | Max: 33m 38s | Hits:  78%/3704  
      🟩 Clang17            Pass: 100%/2   | Total:  1h 05m | Avg: 32m 36s | Max: 33m 39s | Hits:  78%/3704  
      🟩 Clang18            Pass: 100%/7   | Total:  2h 45m | Avg: 23m 35s | Max: 32m 19s | Hits:  84%/12964 
      🟩 GCC7               Pass: 100%/2   | Total:  1h 05m | Avg: 32m 54s | Max: 33m 07s | Hits:  78%/3706  
      🟩 GCC8               Pass: 100%/1   | Total: 29m 41s | Avg: 29m 41s | Max: 29m 41s | Hits:  78%/1853  
      🟩 GCC9               Pass: 100%/2   | Total:  1h 06m | Avg: 33m 12s | Max: 34m 46s | Hits:  78%/3706  
      🟩 GCC10              Pass: 100%/2   | Total:  1h 05m | Avg: 32m 49s | Max: 34m 33s | Hits:  78%/3706  
      🟩 GCC11              Pass: 100%/2   | Total:  1h 06m | Avg: 33m 00s | Max: 33m 14s | Hits:  78%/3706  
      🟩 GCC12              Pass: 100%/2   | Total:  1h 07m | Avg: 33m 33s | Max: 33m 51s | Hits:  78%/3706  
      🟩 GCC13              Pass: 100%/8   | Total:  3h 10m | Avg: 23m 47s | Max: 33m 42s | Hits:  86%/14824 
      🟩 MSVC14.29          Pass: 100%/2   | Total:  1h 51m | Avg: 55m 42s | Max: 56m 48s | Hits:  53%/3692  
      🟩 MSVC14.42          Pass: 100%/3   | Total:  2h 24m | Avg: 48m 06s | Max: 56m 41s | Hits:  58%/5538  
      🟩 NVHPC24.7          Pass: 100%/2   | Total:  1h 44m | Avg: 52m 05s | Max: 53m 19s | Hits:  73%/3704  
    🟩 cxx_family
      🟩 Clang              Pass: 100%/17  | Total:  8h 02m | Avg: 28m 22s | Max: 33m 39s | Hits:  81%/31484 
      🟩 GCC                Pass: 100%/19  | Total:  9h 10m | Avg: 28m 59s | Max: 34m 46s | Hits:  82%/35207 
      🟩 MSVC               Pass: 100%/5   | Total:  4h 15m | Avg: 51m 08s | Max: 56m 48s | Hits:  56%/9230  
      🟩 NVHPC              Pass: 100%/2   | Total:  1h 44m | Avg: 52m 05s | Max: 53m 19s | Hits:  73%/3704  
    🟩 gpu
      🟩 rtx2080            Pass: 100%/33  | Total: 19h 22m | Avg: 35m 14s | Max: 56m 48s | Hits:  76%/61112 
      🟩 rtx4090            Pass: 100%/10  | Total:  3h 50m | Avg: 23m 01s | Max: 55m 21s | Hits:  85%/18513 
    🟩 jobs
      🟩 Build              Pass: 100%/37  | Total: 21h 50m | Avg: 35m 25s | Max: 56m 48s | Hits:  75%/68516 
      🟩 TestCPU            Pass: 100%/3   | Total: 48m 55s | Avg: 16m 18s | Max: 32m 16s | Hits:  89%/5551  
      🟩 TestGPU            Pass: 100%/3   | Total: 33m 31s | Avg: 11m 10s | Max: 11m 39s | Hits:  99%/5558  
    🟩 sm
      🟩 90;90a;100         Pass: 100%/1   | Total: 33m 09s | Avg: 33m 09s | Max: 33m 09s | Hits:  78%/1853  
    🟩 std
      🟩 17                 Pass: 100%/20  | Total: 12h 09m | Avg: 36m 27s | Max: 56m 48s | Hits:  74%/37031 
      🟩 20                 Pass: 100%/21  | Total: 10h 25m | Avg: 29m 46s | Max: 55m 21s | Hits:  80%/38888 
    
  • 🟩 cccl_c_parallel: Pass: 100%/2 | Total: 9m 23s | Avg: 4m 41s | Max: 6m 59s | Hits: 98%/280

    🟩 cpu
      🟩 amd64              Pass: 100%/2   | Total:  9m 23s | Avg:  4m 41s | Max:  6m 59s | Hits:  98%/280   
    🟩 ctk
      🟩 12.8               Pass: 100%/2   | Total:  9m 23s | Avg:  4m 41s | Max:  6m 59s | Hits:  98%/280   
    🟩 cudacxx
      🟩 nvcc12.8           Pass: 100%/2   | Total:  9m 23s | Avg:  4m 41s | Max:  6m 59s | Hits:  98%/280   
    🟩 cudacxx_family
      🟩 nvcc               Pass: 100%/2   | Total:  9m 23s | Avg:  4m 41s | Max:  6m 59s | Hits:  98%/280   
    🟩 cxx
      🟩 GCC13              Pass: 100%/2   | Total:  9m 23s | Avg:  4m 41s | Max:  6m 59s | Hits:  98%/280   
    🟩 cxx_family
      🟩 GCC                Pass: 100%/2   | Total:  9m 23s | Avg:  4m 41s | Max:  6m 59s | Hits:  98%/280   
    🟩 gpu
      🟩 rtx2080            Pass: 100%/2   | Total:  9m 23s | Avg:  4m 41s | Max:  6m 59s | Hits:  98%/280   
    🟩 jobs
      🟩 Build              Pass: 100%/1   | Total:  2m 24s | Avg:  2m 24s | Max:  2m 24s | Hits:  97%/140   
      🟩 Test               Pass: 100%/1   | Total:  6m 59s | Avg:  6m 59s | Max:  6m 59s | Hits:  98%/140   
    
  • 🟩 python: Pass: 100%/1 | Total: 30m 02s | Avg: 30m 02s | Max: 30m 02s

    🟩 cpu
      🟩 amd64              Pass: 100%/1   | Total: 30m 02s | Avg: 30m 02s | Max: 30m 02s
    🟩 ctk
      🟩 12.8               Pass: 100%/1   | Total: 30m 02s | Avg: 30m 02s | Max: 30m 02s
    🟩 cudacxx
      🟩 nvcc12.8           Pass: 100%/1   | Total: 30m 02s | Avg: 30m 02s | Max: 30m 02s
    🟩 cudacxx_family
      🟩 nvcc               Pass: 100%/1   | Total: 30m 02s | Avg: 30m 02s | Max: 30m 02s
    🟩 cxx
      🟩 GCC13              Pass: 100%/1   | Total: 30m 02s | Avg: 30m 02s | Max: 30m 02s
    🟩 cxx_family
      🟩 GCC                Pass: 100%/1   | Total: 30m 02s | Avg: 30m 02s | Max: 30m 02s
    🟩 gpu
      🟩 rtx2080            Pass: 100%/1   | Total: 30m 02s | Avg: 30m 02s | Max: 30m 02s
    🟩 jobs
      🟩 Test               Pass: 100%/1   | Total: 30m 02s | Avg: 30m 02s | Max: 30m 02s
    

πŸ‘ƒ 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

πŸƒβ€ 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

@shwina
shwina merged commit a067839 into NVIDIA:main Feb 7, 2025
shwina pushed a commit to shwina/cccl that referenced this pull request Feb 10, 2025
…IA#3727)

* Moved definition of DeviceSegmentedReduceKernel out to NVRTC compilable header file

Moved the definition from cub/device/dispatch/dispatch_reduce.cuh to
new file cub/device/dispatch/kernels/segmented_reduce.cuh

Added compilation of the new file to cub/test/catch2_test_nvrtc.cu

Needed to remove friended operator<< overload for std::ostream from
arg_index_input_iterator.cuh, since this header is needed by
DeviceSegmentedReduceKernel implementation, and use of std::ostream
breaks compilation by nvrtc

* Include cub/util_type.cuh for value_t and non_void_value_t

* Include cub/detail/type_traits.cuh

* Restore friend operator ostream conditionally for nvcc/host compiler
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

Archived in project

Development

Successfully merging this pull request may close these issues.

Extract Segmented Reduce kernels into NVRTC compilable header

4 participants