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.

Sign up for GitHub

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

Jump to bottom

help the ranges concepts recognize standard contiguous iterators in c++14/17 #3202

Merged
merged 10 commits into from
Dec 21, 2024
Merged

help the ranges concepts recognize standard contiguous iterators in c++14/17 #3202

merged 10 commits into from
Dec 21, 2024

Conversation

ericniebler
Copy link
Collaborator

@ericniebler ericniebler commented Dec 19, 2024 •
edited
Loading

Description

In range-v3, @CaseyCarter and I found a sneaky way to "detect" the contiguous iterators of the 3 major stdlib implementations: libstdc++, libc++, and msvc's stdlib. This PR ports the implementation over so that the following standard containers are now recognized as contiguous:

  • std::vector
  • std::array
  • std::string
  • std::string_view
  • std::span

this PR also removes a hack in cudax that was only necessary because the std:: containers weren't considered contiguous in c++17.

Checklist

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

@ericniebler ericniebler requested review from a team as code owners December 19, 2024 20:17
@copy-pr-bot copy-pr-bot
Copy link

copy-pr-bot bot commented Dec 19, 2024

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

@ericniebler ericniebler force-pushed the cpp17-contiguous-ranges branch from d527d62 to 21907a4 Compare December 19, 2024 20:23
miscco
miscco reviewed Dec 19, 2024
View reviewed changes
Copy link
Collaborator

@miscco miscco left a comment

Choose a reason for hiding this comment

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

I really did not wanted to do that, but its most likely the easiest solution.

Can you please then fix our span to always use the proper range constructor and not the pre-ranges hack?

libcudacxx/include/cuda/std/__iterator/iterator_traits.h Show resolved Hide resolved
libcudacxx/include/cuda/std/__iterator/iterator_traits.h Outdated Show resolved Hide resolved
@github-actions GitHub Actions
Copy link
Contributor

🟩 CI finished in 2h 39m: Pass: 100%/170 | Total: 3d 18h | Avg: 31m 55s | Max: 1h 19m | Hits: 33%/22510

  • 🟩 libcudacxx: Pass: 100%/48 | Total: 14h 59m | Avg: 18m 44s | Max: 51m 36s | Hits: 46%/9814 

    🟩 cpu
  🟩 amd64              Pass: 100%/46  | Total: 14h 21m | Avg: 18m 43s | Max: 51m 36s | Hits:  46%/9814  
  🟩 arm64              Pass: 100%/2   | Total: 38m 05s | Avg: 19m 02s | Max: 21m 38s
🟩 ctk
  🟩 11.1               Pass: 100%/7   | Total:  2h 09m | Avg: 18m 29s | Max: 26m 23s | Hits:  49%/2239  
  🟩 12.5               Pass: 100%/2   | Total: 52m 26s | Avg: 26m 13s | Max: 28m 21s
  🟩 12.6               Pass: 100%/39  | Total: 11h 57m | Avg: 18m 24s | Max: 51m 36s | Hits:  45%/7575  
🟩 cudacxx
  🟩 ClangCUDA18        Pass: 100%/4   | Total:  1h 08m | Avg: 17m 12s | Max: 22m 39s
  🟩 nvcc11.1           Pass: 100%/7   | Total:  2h 09m | Avg: 18m 29s | Max: 26m 23s | Hits:  49%/2239  
  🟩 nvcc12.5           Pass: 100%/2   | Total: 52m 26s | Avg: 26m 13s | Max: 28m 21s
  🟩 nvcc12.6           Pass: 100%/35  | Total: 10h 48m | Avg: 18m 32s | Max: 51m 36s | Hits:  45%/7575  
🟩 cudacxx_family
  🟩 ClangCUDA          Pass: 100%/4   | Total:  1h 08m | Avg: 17m 12s | Max: 22m 39s
  🟩 nvcc               Pass: 100%/44  | Total: 13h 50m | Avg: 18m 52s | Max: 51m 36s | Hits:  46%/9814  
🟩 cxx
  🟩 Clang9             Pass: 100%/4   | Total:  1h 01m | Avg: 15m 18s | Max: 19m 59s
  🟩 Clang10            Pass: 100%/1   | Total: 17m 17s | Avg: 17m 17s | Max: 17m 17s
  🟩 Clang11            Pass: 100%/1   | Total: 15m 10s | Avg: 15m 10s | Max: 15m 10s
  🟩 Clang12            Pass: 100%/1   | Total: 15m 13s | Avg: 15m 13s | Max: 15m 13s
  🟩 Clang13            Pass: 100%/1   | Total: 15m 29s | Avg: 15m 29s | Max: 15m 29s
  🟩 Clang14            Pass: 100%/1   | Total: 15m 29s | Avg: 15m 29s | Max: 15m 29s
  🟩 Clang15            Pass: 100%/1   | Total: 17m 19s | Avg: 17m 19s | Max: 17m 19s
  🟩 Clang16            Pass: 100%/1   | Total: 18m 02s | Avg: 18m 02s | Max: 18m 02s
  🟩 Clang17            Pass: 100%/1   | Total: 16m 50s | Avg: 16m 50s | Max: 16m 50s
  🟩 Clang18            Pass: 100%/8   | Total:  2h 18m | Avg: 17m 21s | Max: 22m 39s
  🟩 GCC6               Pass: 100%/2   | Total: 35m 18s | Avg: 17m 39s | Max: 22m 14s
  🟩 GCC7               Pass: 100%/2   | Total: 25m 18s | Avg: 12m 39s | Max: 13m 23s
  🟩 GCC8               Pass: 100%/1   | Total: 14m 54s | Avg: 14m 54s | Max: 14m 54s
  🟩 GCC9               Pass: 100%/3   | Total: 52m 26s | Avg: 17m 28s | Max: 20m 24s
  🟩 GCC10              Pass: 100%/1   | Total: 16m 29s | Avg: 16m 29s | Max: 16m 29s
  🟩 GCC11              Pass: 100%/1   | Total: 16m 47s | Avg: 16m 47s | Max: 16m 47s
  🟩 GCC12              Pass: 100%/1   | Total: 17m 10s | Avg: 17m 10s | Max: 17m 10s
  🟩 GCC13              Pass: 100%/10  | Total:  3h 27m | Avg: 20m 42s | Max: 51m 36s
  🟩 Intel2023.2.0      Pass: 100%/1   | Total: 18m 12s | Avg: 18m 12s | Max: 18m 12s
  🟩 MSVC14.16          Pass: 100%/1   | Total: 26m 23s | Avg: 26m 23s | Max: 26m 23s | Hits:  49%/2239  
  🟩 MSVC14.29          Pass: 100%/1   | Total: 27m 55s | Avg: 27m 55s | Max: 27m 55s | Hits:  46%/2476  
  🟩 MSVC14.39          Pass: 100%/2   | Total: 58m 25s | Avg: 29m 12s | Max: 30m 16s | Hits:  45%/5099  
  🟩 NVHPC24.7          Pass: 100%/2   | Total: 52m 26s | Avg: 26m 13s | Max: 28m 21s
🟩 cxx_family
  🟩 Clang              Pass: 100%/20  | Total:  5h 30m | Avg: 16m 32s | Max: 22m 39s
  🟩 GCC                Pass: 100%/21  | Total:  6h 25m | Avg: 18m 21s | Max: 51m 36s
  🟩 Intel              Pass: 100%/1   | Total: 18m 12s | Avg: 18m 12s | Max: 18m 12s
  🟩 MSVC               Pass: 100%/4   | Total:  1h 52m | Avg: 28m 10s | Max: 30m 16s | Hits:  46%/9814  
  🟩 NVHPC              Pass: 100%/2   | Total: 52m 26s | Avg: 26m 13s | Max: 28m 21s
🟩 gpu
  🟩 v100               Pass: 100%/48  | Total: 14h 59m | Avg: 18m 44s | Max: 51m 36s | Hits:  46%/9814  
🟩 jobs
  🟩 Build              Pass: 100%/41  | Total: 12h 18m | Avg: 18m 00s | Max: 30m 16s | Hits:  46%/9814  
  🟩 NVRTC              Pass: 100%/4   | Total:  1h 28m | Avg: 22m 05s | Max: 24m 54s
  🟩 Test               Pass: 100%/2   | Total:  1h 11m | Avg: 35m 32s | Max: 51m 36s
  🟩 VerifyCodegen      Pass: 100%/1   | Total:  2m 00s | Avg:  2m 00s | Max:  2m 00s
🟩 sm
  🟩 90                 Pass: 100%/1   | Total: 13m 55s | Avg: 13m 55s | Max: 13m 55s
  🟩 90a                Pass: 100%/2   | Total: 24m 20s | Avg: 12m 10s | Max: 13m 17s
🟩 std
  🟩 11                 Pass: 100%/6   | Total:  1h 48m | Avg: 18m 08s | Max: 22m 14s
  🟩 14                 Pass: 100%/5   | Total:  1h 31m | Avg: 18m 15s | Max: 26m 23s | Hits:  49%/2239  
  🟩 17                 Pass: 100%/13  | Total:  4h 09m | Avg: 19m 09s | Max: 28m 09s | Hits:  46%/4952  
  🟩 20                 Pass: 100%/23  | Total:  7h 28m | Avg: 19m 29s | Max: 51m 36s | Hits:  44%/2623

  • 🟩 cub: Pass: 100%/47 | Total: 1d 15h | Avg: 50m 41s | Max: 1h 07m | Hits: 3%/3124 

    🟩 cpu
  🟩 amd64              Pass: 100%/45  | Total:  1d 13h | Avg: 50m 20s | Max:  1h 07m | Hits:   3%/3124  
  🟩 arm64              Pass: 100%/2   | Total:  1h 57m | Avg: 58m 34s | Max:  1h 00m
🟩 ctk
  🟩 11.1               Pass: 100%/7   | Total:  6h 03m | Avg: 51m 55s | Max: 57m 58s | Hits:   2%/781   
  🟩 12.5               Pass: 100%/2   | Total:  2h 15m | Avg:  1h 07m | Max:  1h 07m
  🟩 12.6               Pass: 100%/38  | Total:  1d 07h | Avg: 49m 34s | Max:  1h 05m | Hits:   3%/2343  
🟩 cudacxx
  🟩 ClangCUDA18        Pass: 100%/2   | Total:  2h 03m | Avg:  1h 01m | Max:  1h 02m
  🟩 nvcc11.1           Pass: 100%/7   | Total:  6h 03m | Avg: 51m 55s | Max: 57m 58s | Hits:   2%/781   
  🟩 nvcc12.5           Pass: 100%/2   | Total:  2h 15m | Avg:  1h 07m | Max:  1h 07m
  🟩 nvcc12.6           Pass: 100%/36  | Total:  1d 05h | Avg: 48m 53s | Max:  1h 05m | Hits:   3%/2343  
🟩 cudacxx_family
  🟩 ClangCUDA          Pass: 100%/2   | Total:  2h 03m | Avg:  1h 01m | Max:  1h 02m
  🟩 nvcc               Pass: 100%/45  | Total:  1d 13h | Avg: 50m 12s | Max:  1h 07m | Hits:   3%/3124  
🟩 cxx
  🟩 Clang9             Pass: 100%/4   | Total:  3h 36m | Avg: 54m 04s | Max: 58m 08s
  🟩 Clang10            Pass: 100%/1   | Total: 57m 05s | Avg: 57m 05s | Max: 57m 05s
  🟩 Clang11            Pass: 100%/1   | Total: 56m 39s | Avg: 56m 39s | Max: 56m 39s
  🟩 Clang12            Pass: 100%/1   | Total: 55m 15s | Avg: 55m 15s | Max: 55m 15s
  🟩 Clang13            Pass: 100%/1   | Total:  1h 01m | Avg:  1h 01m | Max:  1h 01m
  🟩 Clang14            Pass: 100%/1   | Total: 59m 20s | Avg: 59m 20s | Max: 59m 20s
  🟩 Clang15            Pass: 100%/1   | Total: 59m 36s | Avg: 59m 36s | Max: 59m 36s
  🟩 Clang16            Pass: 100%/1   | Total:  1h 04m | Avg:  1h 04m | Max:  1h 04m
  🟩 Clang17            Pass: 100%/1   | Total:  1h 01m | Avg:  1h 01m | Max:  1h 01m
  🟩 Clang18            Pass: 100%/7   | Total:  5h 35m | Avg: 47m 58s | Max:  1h 02m
  🟩 GCC6               Pass: 100%/2   | Total:  1h 40m | Avg: 50m 00s | Max: 51m 54s
  🟩 GCC7               Pass: 100%/2   | Total:  1h 47m | Avg: 53m 37s | Max: 53m 58s
  🟩 GCC8               Pass: 100%/1   | Total: 53m 22s | Avg: 53m 22s | Max: 53m 22s
  🟩 GCC9               Pass: 100%/3   | Total:  2h 43m | Avg: 54m 29s | Max:  1h 00m
  🟩 GCC10              Pass: 100%/1   | Total: 56m 59s | Avg: 56m 59s | Max: 56m 59s
  🟩 GCC11              Pass: 100%/1   | Total: 54m 42s | Avg: 54m 42s | Max: 54m 42s
  🟩 GCC12              Pass: 100%/3   | Total:  1h 36m | Avg: 32m 01s | Max: 55m 08s
  🟩 GCC13              Pass: 100%/8   | Total:  4h 36m | Avg: 34m 32s | Max:  1h 00m
  🟩 Intel2023.2.0      Pass: 100%/1   | Total:  1h 02m | Avg:  1h 02m | Max:  1h 02m
  🟩 MSVC14.16          Pass: 100%/1   | Total: 57m 58s | Avg: 57m 58s | Max: 57m 58s | Hits:   2%/781   
  🟩 MSVC14.29          Pass: 100%/1   | Total:  1h 02m | Avg:  1h 02m | Max:  1h 02m | Hits:   2%/781   
  🟩 MSVC14.39          Pass: 100%/2   | Total:  2h 08m | Avg:  1h 04m | Max:  1h 05m | Hits:   3%/1562  
  🟩 NVHPC24.7          Pass: 100%/2   | Total:  2h 15m | Avg:  1h 07m | Max:  1h 07m
🟩 cxx_family
  🟩 Clang              Pass: 100%/19  | Total: 17h 07m | Avg: 54m 04s | Max:  1h 04m
  🟩 GCC                Pass: 100%/21  | Total: 15h 08m | Avg: 43m 14s | Max:  1h 00m
  🟩 Intel              Pass: 100%/1   | Total:  1h 02m | Avg:  1h 02m | Max:  1h 02m
  🟩 MSVC               Pass: 100%/4   | Total:  4h 09m | Avg:  1h 02m | Max:  1h 05m | Hits:   3%/3124  
  🟩 NVHPC              Pass: 100%/2   | Total:  2h 15m | Avg:  1h 07m | Max:  1h 07m
🟩 gpu
  🟩 h100               Pass: 100%/2   | Total: 40m 57s | Avg: 20m 28s | Max: 25m 05s
  🟩 v100               Pass: 100%/45  | Total:  1d 15h | Avg: 52m 02s | Max:  1h 07m | Hits:   3%/3124  
🟩 jobs
  🟩 Build              Pass: 100%/40  | Total:  1d 13h | Avg: 56m 13s | Max:  1h 07m | Hits:   3%/3124  
  🟩 DeviceLaunch       Pass: 100%/1   | Total: 21m 45s | Avg: 21m 45s | Max: 21m 45s
  🟩 GraphCapture       Pass: 100%/1   | Total: 14m 35s | Avg: 14m 35s | Max: 14m 35s
  🟩 HostLaunch         Pass: 100%/3   | Total: 49m 16s | Avg: 16m 25s | Max: 17m 06s
  🟩 TestGPU            Pass: 100%/2   | Total: 47m 57s | Avg: 23m 58s | Max: 25m 42s
🟩 sm
  🟩 90                 Pass: 100%/2   | Total: 40m 57s | Avg: 20m 28s | Max: 25m 05s
  🟩 90a                Pass: 100%/1   | Total: 26m 35s | Avg: 26m 35s | Max: 26m 35s
🟩 std
  🟩 11                 Pass: 100%/5   | Total:  4h 23m | Avg: 52m 37s | Max: 58m 08s
  🟩 14                 Pass: 100%/4   | Total:  3h 35m | Avg: 53m 48s | Max: 57m 58s | Hits:   2%/781   
  🟩 17                 Pass: 100%/12  | Total: 11h 45m | Avg: 58m 48s | Max:  1h 07m | Hits:   2%/1562  
  🟩 20                 Pass: 100%/26  | Total: 19h 58m | Avg: 46m 05s | Max:  1h 07m | Hits:   3%/781

  • 🟩 thrust: Pass: 100%/46 | Total: 1d 05h | Avg: 37m 57s | Max: 1h 19m | Hits: 30%/9260 

    🟩 cmake_options
  🟩 -DTHRUST_DISPATCH_TYPE=Force32bit Pass: 100%/2   | Total: 47m 18s | Avg: 23m 39s | Max: 35m 09s
🟩 cpu
  🟩 amd64              Pass: 100%/44  | Total:  1d 03h | Avg: 37m 57s | Max:  1h 19m | Hits:  30%/9260  
  🟩 arm64              Pass: 100%/2   | Total:  1h 15m | Avg: 37m 48s | Max: 39m 51s
🟩 ctk
  🟩 11.1               Pass: 100%/7   | Total:  4h 13m | Avg: 36m 15s | Max:  1h 06m | Hits:  12%/1852  
  🟩 12.5               Pass: 100%/2   | Total:  2h 23m | Avg:  1h 11m | Max:  1h 15m
  🟩 12.6               Pass: 100%/37  | Total: 22h 28m | Avg: 36m 27s | Max:  1h 19m | Hits:  34%/7408  
🟩 cudacxx
  🟩 ClangCUDA18        Pass: 100%/2   | Total:  1h 06m | Avg: 33m 05s | Max: 33m 09s
  🟩 nvcc11.1           Pass: 100%/7   | Total:  4h 13m | Avg: 36m 15s | Max:  1h 06m | Hits:  12%/1852  
  🟩 nvcc12.5           Pass: 100%/2   | Total:  2h 23m | Avg:  1h 11m | Max:  1h 15m
  🟩 nvcc12.6           Pass: 100%/35  | Total: 21h 22m | Avg: 36m 38s | Max:  1h 19m | Hits:  34%/7408  
🟩 cudacxx_family
  🟩 ClangCUDA          Pass: 100%/2   | Total:  1h 06m | Avg: 33m 05s | Max: 33m 09s
  🟩 nvcc               Pass: 100%/44  | Total:  1d 03h | Avg: 38m 10s | Max:  1h 19m | Hits:  30%/9260  
🟩 cxx
  🟩 Clang9             Pass: 100%/4   | Total:  2h 12m | Avg: 33m 13s | Max: 35m 15s
  🟩 Clang10            Pass: 100%/1   | Total: 43m 10s | Avg: 43m 10s | Max: 43m 10s
  🟩 Clang11            Pass: 100%/1   | Total: 40m 26s | Avg: 40m 26s | Max: 40m 26s
  🟩 Clang12            Pass: 100%/1   | Total: 39m 02s | Avg: 39m 02s | Max: 39m 02s
  🟩 Clang13            Pass: 100%/1   | Total: 39m 29s | Avg: 39m 29s | Max: 39m 29s
  🟩 Clang14            Pass: 100%/1   | Total: 37m 00s | Avg: 37m 00s | Max: 37m 00s
  🟩 Clang15            Pass: 100%/1   | Total: 37m 25s | Avg: 37m 25s | Max: 37m 25s
  🟩 Clang16            Pass: 100%/1   | Total: 37m 41s | Avg: 37m 41s | Max: 37m 41s
  🟩 Clang17            Pass: 100%/1   | Total: 40m 20s | Avg: 40m 20s | Max: 40m 20s
  🟩 Clang18            Pass: 100%/7   | Total:  3h 16m | Avg: 28m 02s | Max: 36m 36s
  🟩 GCC6               Pass: 100%/2   | Total: 58m 26s | Avg: 29m 13s | Max: 32m 45s
  🟩 GCC7               Pass: 100%/2   | Total:  1h 03m | Avg: 31m 37s | Max: 34m 55s
  🟩 GCC8               Pass: 100%/1   | Total: 39m 40s | Avg: 39m 40s | Max: 39m 40s
  🟩 GCC9               Pass: 100%/3   | Total:  1h 45m | Avg: 35m 17s | Max: 40m 19s
  🟩 GCC10              Pass: 100%/1   | Total: 38m 58s | Avg: 38m 58s | Max: 38m 58s
  🟩 GCC11              Pass: 100%/1   | Total: 38m 51s | Avg: 38m 51s | Max: 38m 51s
  🟩 GCC12              Pass: 100%/1   | Total: 39m 54s | Avg: 39m 54s | Max: 39m 54s
  🟩 GCC13              Pass: 100%/8   | Total:  3h 34m | Avg: 26m 49s | Max: 40m 25s
  🟩 Intel2023.2.0      Pass: 100%/1   | Total: 50m 05s | Avg: 50m 05s | Max: 50m 05s
  🟩 MSVC14.16          Pass: 100%/1   | Total:  1h 06m | Avg:  1h 06m | Max:  1h 06m | Hits:  12%/1852  
  🟩 MSVC14.29          Pass: 100%/1   | Total:  1h 08m | Avg:  1h 08m | Max:  1h 08m | Hits:  12%/1852  
  🟩 MSVC14.39          Pass: 100%/3   | Total:  2h 55m | Avg: 58m 21s | Max:  1h 19m | Hits:  41%/5556  
  🟩 NVHPC24.7          Pass: 100%/2   | Total:  2h 23m | Avg:  1h 11m | Max:  1h 15m
🟩 cxx_family
  🟩 Clang              Pass: 100%/19  | Total: 10h 43m | Avg: 33m 52s | Max: 43m 10s
  🟩 GCC                Pass: 100%/19  | Total:  9h 59m | Avg: 31m 33s | Max: 40m 25s
  🟩 Intel              Pass: 100%/1   | Total: 50m 05s | Avg: 50m 05s | Max: 50m 05s
  🟩 MSVC               Pass: 100%/5   | Total:  5h 09m | Avg:  1h 01m | Max:  1h 19m | Hits:  30%/9260  
  🟩 NVHPC              Pass: 100%/2   | Total:  2h 23m | Avg:  1h 11m | Max:  1h 15m
🟩 gpu
  🟩 v100               Pass: 100%/46  | Total:  1d 05h | Avg: 37m 57s | Max:  1h 19m | Hits:  30%/9260  
🟩 jobs
  🟩 Build              Pass: 100%/40  | Total:  1d 03h | Avg: 41m 42s | Max:  1h 19m | Hits:  12%/7408  
  🟩 TestCPU            Pass: 100%/3   | Total: 38m 37s | Avg: 12m 52s | Max: 24m 00s | Hits:  99%/1852  
  🟩 TestGPU            Pass: 100%/3   | Total: 39m 03s | Avg: 13m 01s | Max: 15m 30s
🟩 sm
  🟩 90a                Pass: 100%/1   | Total: 29m 03s | Avg: 29m 03s | Max: 29m 03s
🟩 std
  🟩 11                 Pass: 100%/5   | Total:  2h 26m | Avg: 29m 16s | Max: 34m 18s
  🟩 14                 Pass: 100%/4   | Total:  2h 48m | Avg: 42m 09s | Max:  1h 06m | Hits:  12%/1852  
  🟩 17                 Pass: 100%/12  | Total:  9h 27m | Avg: 47m 18s | Max:  1h 15m | Hits:  12%/3704  
  🟩 20                 Pass: 100%/23  | Total: 13h 35m | Avg: 35m 27s | Max:  1h 19m | Hits:  56%/3704

  • 🟩 cudax: Pass: 100%/26 | Total: 5h 55m | Avg: 13m 40s | Max: 17m 44s | Hits: 34%/312 

    🟩 cpu
  🟩 amd64              Pass: 100%/22  | Total:  5h 02m | Avg: 13m 45s | Max: 17m 44s | Hits:  34%/312   
  🟩 arm64              Pass: 100%/4   | Total: 52m 58s | Avg: 13m 14s | Max: 14m 32s
🟩 ctk
  🟩 12.0               Pass: 100%/3   | Total: 38m 41s | Avg: 12m 53s | Max: 14m 19s | Hits:  34%/156   
  🟩 12.5               Pass: 100%/2   | Total: 18m 43s | Avg:  9m 21s | Max:  9m 27s
  🟩 12.6               Pass: 100%/21  | Total:  4h 58m | Avg: 14m 12s | Max: 17m 44s | Hits:  34%/156   
🟩 cudacxx
  🟩 nvcc12.0           Pass: 100%/3   | Total: 38m 41s | Avg: 12m 53s | Max: 14m 19s | Hits:  34%/156   
  🟩 nvcc12.5           Pass: 100%/2   | Total: 18m 43s | Avg:  9m 21s | Max:  9m 27s
  🟩 nvcc12.6           Pass: 100%/21  | Total:  4h 58m | Avg: 14m 12s | Max: 17m 44s | Hits:  34%/156   
🟩 cudacxx_family
  🟩 nvcc               Pass: 100%/26  | Total:  5h 55m | Avg: 13m 40s | Max: 17m 44s | Hits:  34%/312   
🟩 cxx
  🟩 Clang9             Pass: 100%/1   | Total: 14m 19s | Avg: 14m 19s | Max: 14m 19s
  🟩 Clang10            Pass: 100%/1   | Total: 15m 50s | Avg: 15m 50s | Max: 15m 50s
  🟩 Clang11            Pass: 100%/1   | Total: 14m 49s | Avg: 14m 49s | Max: 14m 49s
  🟩 Clang12            Pass: 100%/1   | Total: 14m 46s | Avg: 14m 46s | Max: 14m 46s
  🟩 Clang13            Pass: 100%/1   | Total: 13m 19s | Avg: 13m 19s | Max: 13m 19s
  🟩 Clang14            Pass: 100%/1   | Total: 13m 01s | Avg: 13m 01s | Max: 13m 01s
  🟩 Clang15            Pass: 100%/1   | Total: 14m 36s | Avg: 14m 36s | Max: 14m 36s
  🟩 Clang16            Pass: 100%/1   | Total: 14m 58s | Avg: 14m 58s | Max: 14m 58s
  🟩 Clang17            Pass: 100%/1   | Total: 14m 32s | Avg: 14m 32s | Max: 14m 32s
  🟩 Clang18            Pass: 100%/4   | Total: 57m 14s | Avg: 14m 18s | Max: 17m 28s
  🟩 GCC9               Pass: 100%/1   | Total: 14m 00s | Avg: 14m 00s | Max: 14m 00s
  🟩 GCC10              Pass: 100%/1   | Total: 16m 16s | Avg: 16m 16s | Max: 16m 16s
  🟩 GCC11              Pass: 100%/1   | Total: 14m 46s | Avg: 14m 46s | Max: 14m 46s
  🟩 GCC12              Pass: 100%/2   | Total: 33m 12s | Avg: 16m 36s | Max: 17m 44s
  🟩 GCC13              Pass: 100%/4   | Total: 50m 31s | Avg: 12m 37s | Max: 14m 32s
  🟩 MSVC14.36          Pass: 100%/1   | Total: 10m 22s | Avg: 10m 22s | Max: 10m 22s | Hits:  34%/156   
  🟩 MSVC14.39          Pass: 100%/1   | Total: 10m 24s | Avg: 10m 24s | Max: 10m 24s | Hits:  34%/156   
  🟩 NVHPC24.7          Pass: 100%/2   | Total: 18m 43s | Avg:  9m 21s | Max:  9m 27s
🟩 cxx_family
  🟩 Clang              Pass: 100%/13  | Total:  3h 07m | Avg: 14m 24s | Max: 17m 28s
  🟩 GCC                Pass: 100%/9   | Total:  2h 08m | Avg: 14m 18s | Max: 17m 44s
  🟩 MSVC               Pass: 100%/2   | Total: 20m 46s | Avg: 10m 23s | Max: 10m 24s | Hits:  34%/312   
  🟩 NVHPC              Pass: 100%/2   | Total: 18m 43s | Avg:  9m 21s | Max:  9m 27s
🟩 gpu
  🟩 v100               Pass: 100%/26  | Total:  5h 55m | Avg: 13m 40s | Max: 17m 44s | Hits:  34%/312   
🟩 jobs
  🟩 Build              Pass: 100%/24  | Total:  5h 20m | Avg: 13m 21s | Max: 16m 16s | Hits:  34%/312   
  🟩 Test               Pass: 100%/2   | Total: 35m 12s | Avg: 17m 36s | Max: 17m 44s
🟩 sm
  🟩 90                 Pass: 100%/1   | Total: 11m 18s | Avg: 11m 18s | Max: 11m 18s
  🟩 90a                Pass: 100%/1   | Total: 11m 45s | Avg: 11m 45s | Max: 11m 45s
🟩 std
  🟩 17                 Pass: 100%/6   | Total:  1h 13m | Avg: 12m 19s | Max: 14m 19s
  🟩 20                 Pass: 100%/20  | Total:  4h 41m | Avg: 14m 05s | Max: 17m 44s | Hits:  34%/312

  • 🟩 cccl_c_parallel: Pass: 100%/2 | Total: 13m 02s | Avg: 6m 31s | Max: 10m 56s

     
    🟩 cpu
  🟩 amd64              Pass: 100%/2   | Total: 13m 02s | Avg:  6m 31s | Max: 10m 56s
🟩 ctk
  🟩 12.6               Pass: 100%/2   | Total: 13m 02s | Avg:  6m 31s | Max: 10m 56s
🟩 cudacxx
  🟩 nvcc12.6           Pass: 100%/2   | Total: 13m 02s | Avg:  6m 31s | Max: 10m 56s
🟩 cudacxx_family
  🟩 nvcc               Pass: 100%/2   | Total: 13m 02s | Avg:  6m 31s | Max: 10m 56s
🟩 cxx
  🟩 GCC13              Pass: 100%/2   | Total: 13m 02s | Avg:  6m 31s | Max: 10m 56s
🟩 cxx_family
  🟩 GCC                Pass: 100%/2   | Total: 13m 02s | Avg:  6m 31s | Max: 10m 56s
🟩 gpu
  🟩 v100               Pass: 100%/2   | Total: 13m 02s | Avg:  6m 31s | Max: 10m 56s
🟩 jobs
  🟩 Build              Pass: 100%/1   | Total:  2m 06s | Avg:  2m 06s | Max:  2m 06s
  🟩 Test               Pass: 100%/1   | Total: 10m 56s | Avg: 10m 56s | Max: 10m 56s

  • 🟩 python: Pass: 100%/1 | Total: 29m 24s | Avg: 29m 24s | Max: 29m 24s

     
    🟩 cpu
  🟩 amd64              Pass: 100%/1   | Total: 29m 24s | Avg: 29m 24s | Max: 29m 24s
🟩 ctk
  🟩 12.6               Pass: 100%/1   | Total: 29m 24s | Avg: 29m 24s | Max: 29m 24s
🟩 cudacxx
  🟩 nvcc12.6           Pass: 100%/1   | Total: 29m 24s | Avg: 29m 24s | Max: 29m 24s
🟩 cudacxx_family
  🟩 nvcc               Pass: 100%/1   | Total: 29m 24s | Avg: 29m 24s | Max: 29m 24s
🟩 cxx
  🟩 GCC13              Pass: 100%/1   | Total: 29m 24s | Avg: 29m 24s | Max: 29m 24s
🟩 cxx_family
  🟩 GCC                Pass: 100%/1   | Total: 29m 24s | Avg: 29m 24s | Max: 29m 24s
🟩 gpu
  🟩 v100               Pass: 100%/1   | Total: 29m 24s | Avg: 29m 24s | Max: 29m 24s
🟩 jobs
  🟩 Test               Pass: 100%/1   | Total: 29m 24s | Avg: 29m 24s | Max: 29m 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: 170)
# Runner
125 linux-amd64-cpu16
19 linux-amd64-gpu-v100-latest-1
15 windows-amd64-cpu16
10 linux-arm64-cpu16
1 linux-amd64-gpu-h100-latest-1-testing

@ericniebler
Copy link
Collaborator Author

Can you please then fix our span to always use the proper range constructor and not the pre-ranges hack?

i started down this rabbit hole, but the range stuff is only defined in C++17, and (for now) span still supports C++14.

@github-actions GitHub Actions
Copy link
Contributor

🟩 CI finished in 1h 59m: Pass: 100%/170 | Total: 1d 14h | Avg: 13m 34s | Max: 1h 18m | Hits: 34%/22530

  • 🟩 libcudacxx: Pass: 100%/48 | Total: 11h 46m | Avg: 14m 42s | Max: 31m 28s | Hits: 48%/9814 

    🟩 cpu
  🟩 amd64              Pass: 100%/46  | Total: 11h 26m | Avg: 14m 55s | Max: 31m 28s | Hits:  48%/9814  
  🟩 arm64              Pass: 100%/2   | Total: 19m 42s | Avg:  9m 51s | Max: 16m 12s
🟩 ctk
  🟩 11.1               Pass: 100%/7   | Total:  1h 19m | Avg: 11m 23s | Max: 26m 02s | Hits:  49%/2239  
  🟩 12.5               Pass: 100%/2   | Total: 47m 51s | Avg: 23m 55s | Max: 24m 40s
  🟩 12.6               Pass: 100%/39  | Total:  9h 38m | Avg: 14m 50s | Max: 31m 28s | Hits:  47%/7575  
🟩 cudacxx
  🟩 ClangCUDA18        Pass: 100%/4   | Total:  1h 02m | Avg: 15m 37s | Max: 19m 03s
  🟩 nvcc11.1           Pass: 100%/7   | Total:  1h 19m | Avg: 11m 23s | Max: 26m 02s | Hits:  49%/2239  
  🟩 nvcc12.5           Pass: 100%/2   | Total: 47m 51s | Avg: 23m 55s | Max: 24m 40s
  🟩 nvcc12.6           Pass: 100%/35  | Total:  8h 36m | Avg: 14m 44s | Max: 31m 28s | Hits:  47%/7575  
🟩 cudacxx_family
  🟩 ClangCUDA          Pass: 100%/4   | Total:  1h 02m | Avg: 15m 37s | Max: 19m 03s
  🟩 nvcc               Pass: 100%/44  | Total: 10h 43m | Avg: 14m 37s | Max: 31m 28s | Hits:  48%/9814  
🟩 cxx
  🟩 Clang9             Pass: 100%/4   | Total: 41m 03s | Avg: 10m 15s | Max: 12m 44s
  🟩 Clang10            Pass: 100%/1   | Total: 17m 51s | Avg: 17m 51s | Max: 17m 51s
  🟩 Clang11            Pass: 100%/1   | Total:  4m 14s | Avg:  4m 14s | Max:  4m 14s
  🟩 Clang12            Pass: 100%/1   | Total: 14m 26s | Avg: 14m 26s | Max: 14m 26s
  🟩 Clang13            Pass: 100%/1   | Total: 14m 35s | Avg: 14m 35s | Max: 14m 35s
  🟩 Clang14            Pass: 100%/1   | Total:  4m 06s | Avg:  4m 06s | Max:  4m 06s
  🟩 Clang15            Pass: 100%/1   | Total: 16m 41s | Avg: 16m 41s | Max: 16m 41s
  🟩 Clang16            Pass: 100%/1   | Total:  4m 13s | Avg:  4m 13s | Max:  4m 13s
  🟩 Clang17            Pass: 100%/1   | Total: 16m 55s | Avg: 16m 55s | Max: 16m 55s
  🟩 Clang18            Pass: 100%/8   | Total:  2h 08m | Avg: 16m 01s | Max: 19m 03s
  🟩 GCC6               Pass: 100%/2   | Total: 24m 03s | Avg: 12m 01s | Max: 21m 19s
  🟩 GCC7               Pass: 100%/2   | Total: 22m 10s | Avg: 11m 05s | Max: 11m 28s
  🟩 GCC8               Pass: 100%/1   | Total: 15m 12s | Avg: 15m 12s | Max: 15m 12s
  🟩 GCC9               Pass: 100%/3   | Total: 17m 30s | Avg:  5m 50s | Max: 11m 16s
  🟩 GCC10              Pass: 100%/1   | Total: 15m 23s | Avg: 15m 23s | Max: 15m 23s
  🟩 GCC11              Pass: 100%/1   | Total:  3m 34s | Avg:  3m 34s | Max:  3m 34s
  🟩 GCC12              Pass: 100%/1   | Total: 18m 04s | Avg: 18m 04s | Max: 18m 04s
  🟩 GCC13              Pass: 100%/10  | Total:  2h 40m | Avg: 16m 03s | Max: 31m 28s
  🟩 Intel2023.2.0      Pass: 100%/1   | Total:  5m 52s | Avg:  5m 52s | Max:  5m 52s
  🟩 MSVC14.16          Pass: 100%/1   | Total: 26m 02s | Avg: 26m 02s | Max: 26m 02s | Hits:  49%/2239  
  🟩 MSVC14.29          Pass: 100%/1   | Total: 28m 28s | Avg: 28m 28s | Max: 28m 28s | Hits:  46%/2476  
  🟩 MSVC14.39          Pass: 100%/2   | Total: 59m 11s | Avg: 29m 35s | Max: 30m 05s | Hits:  47%/5099  
  🟩 NVHPC24.7          Pass: 100%/2   | Total: 47m 51s | Avg: 23m 55s | Max: 24m 40s
🟩 cxx_family
  🟩 Clang              Pass: 100%/20  | Total:  4h 22m | Avg: 13m 06s | Max: 19m 03s
  🟩 GCC                Pass: 100%/21  | Total:  4h 36m | Avg: 13m 09s | Max: 31m 28s
  🟩 Intel              Pass: 100%/1   | Total:  5m 52s | Avg:  5m 52s | Max:  5m 52s
  🟩 MSVC               Pass: 100%/4   | Total:  1h 53m | Avg: 28m 25s | Max: 30m 05s | Hits:  48%/9814  
  🟩 NVHPC              Pass: 100%/2   | Total: 47m 51s | Avg: 23m 55s | Max: 24m 40s
🟩 gpu
  🟩 v100               Pass: 100%/48  | Total: 11h 46m | Avg: 14m 42s | Max: 31m 28s | Hits:  48%/9814  
🟩 jobs
  🟩 Build              Pass: 100%/41  | Total:  9h 14m | Avg: 13m 31s | Max: 30m 05s | Hits:  48%/9814  
  🟩 NVRTC              Pass: 100%/4   | Total:  1h 54m | Avg: 28m 34s | Max: 31m 28s
  🟩 Test               Pass: 100%/2   | Total: 35m 17s | Avg: 17m 38s | Max: 18m 07s
  🟩 VerifyCodegen      Pass: 100%/1   | Total:  1m 59s | Avg:  1m 59s | Max:  1m 59s
🟩 sm
  🟩 90                 Pass: 100%/1   | Total: 12m 06s | Avg: 12m 06s | Max: 12m 06s
  🟩 90a                Pass: 100%/2   | Total: 17m 55s | Avg:  8m 57s | Max: 13m 53s
🟩 std
  🟩 11                 Pass: 100%/6   | Total:  1h 19m | Avg: 13m 14s | Max: 28m 19s
  🟩 14                 Pass: 100%/5   | Total:  1h 23m | Avg: 16m 41s | Max: 31m 28s | Hits:  49%/2239  
  🟩 17                 Pass: 100%/13  | Total:  3h 44m | Avg: 17m 14s | Max: 29m 18s | Hits:  49%/4952  
  🟩 20                 Pass: 100%/23  | Total:  5h 17m | Avg: 13m 47s | Max: 30m 05s | Hits:  44%/2623

  • 🟩 cub: Pass: 100%/47 | Total: 12h 25m | Avg: 15m 52s | Max: 1h 13m | Hits: 2%/3144 

    🟩 cpu
  🟩 amd64              Pass: 100%/45  | Total: 12h 16m | Avg: 16m 21s | Max:  1h 13m | Hits:   2%/3144  
  🟩 arm64              Pass: 100%/2   | Total:  9m 33s | Avg:  4m 46s | Max:  4m 50s
🟩 ctk
  🟩 11.1               Pass: 100%/7   | Total:  1h 28m | Avg: 12m 35s | Max:  1h 01m | Hits:   2%/786   
  🟩 12.5               Pass: 100%/2   | Total:  2h 23m | Avg:  1h 11m | Max:  1h 13m
  🟩 12.6               Pass: 100%/38  | Total:  8h 34m | Avg: 13m 32s | Max:  1h 12m | Hits:   2%/2358  
🟩 cudacxx
  🟩 ClangCUDA18        Pass: 100%/2   | Total:  8m 53s | Avg:  4m 26s | Max:  4m 33s
  🟩 nvcc11.1           Pass: 100%/7   | Total:  1h 28m | Avg: 12m 35s | Max:  1h 01m | Hits:   2%/786   
  🟩 nvcc12.5           Pass: 100%/2   | Total:  2h 23m | Avg:  1h 11m | Max:  1h 13m
  🟩 nvcc12.6           Pass: 100%/36  | Total:  8h 25m | Avg: 14m 02s | Max:  1h 12m | Hits:   2%/2358  
🟩 cudacxx_family
  🟩 ClangCUDA          Pass: 100%/2   | Total:  8m 53s | Avg:  4m 26s | Max:  4m 33s
  🟩 nvcc               Pass: 100%/45  | Total: 12h 17m | Avg: 16m 22s | Max:  1h 13m | Hits:   2%/3144  
🟩 cxx
  🟩 Clang9             Pass: 100%/4   | Total: 21m 01s | Avg:  5m 15s | Max:  5m 55s
  🟩 Clang10            Pass: 100%/1   | Total:  6m 17s | Avg:  6m 17s | Max:  6m 17s
  🟩 Clang11            Pass: 100%/1   | Total:  5m 04s | Avg:  5m 04s | Max:  5m 04s
  🟩 Clang12            Pass: 100%/1   | Total:  5m 15s | Avg:  5m 15s | Max:  5m 15s
  🟩 Clang13            Pass: 100%/1   | Total:  5m 43s | Avg:  5m 43s | Max:  5m 43s
  🟩 Clang14            Pass: 100%/1   | Total:  5m 30s | Avg:  5m 30s | Max:  5m 30s
  🟩 Clang15            Pass: 100%/1   | Total:  5m 36s | Avg:  5m 36s | Max:  5m 36s
  🟩 Clang16            Pass: 100%/1   | Total:  5m 37s | Avg:  5m 37s | Max:  5m 37s
  🟩 Clang17            Pass: 100%/1   | Total:  5m 32s | Avg:  5m 32s | Max:  5m 32s
  🟩 Clang18            Pass: 100%/7   | Total:  1h 06m | Avg:  9m 27s | Max: 24m 28s
  🟩 GCC6               Pass: 100%/2   | Total:  8m 47s | Avg:  4m 23s | Max:  4m 36s
  🟩 GCC7               Pass: 100%/2   | Total: 10m 21s | Avg:  5m 10s | Max:  5m 31s
  🟩 GCC8               Pass: 100%/1   | Total:  5m 21s | Avg:  5m 21s | Max:  5m 21s
  🟩 GCC9               Pass: 100%/3   | Total: 14m 24s | Avg:  4m 48s | Max:  5m 44s
  🟩 GCC10              Pass: 100%/1   | Total:  5m 46s | Avg:  5m 46s | Max:  5m 46s
  🟩 GCC11              Pass: 100%/1   | Total:  5m 46s | Avg:  5m 46s | Max:  5m 46s
  🟩 GCC12              Pass: 100%/3   | Total: 26m 30s | Avg:  8m 50s | Max: 15m 51s
  🟩 GCC13              Pass: 100%/8   | Total:  1h 59m | Avg: 14m 58s | Max: 36m 51s
  🟩 Intel2023.2.0      Pass: 100%/1   | Total:  6m 28s | Avg:  6m 28s | Max:  6m 28s
  🟩 MSVC14.16          Pass: 100%/1   | Total:  1h 01m | Avg:  1h 01m | Max:  1h 01m | Hits:   2%/786   
  🟩 MSVC14.29          Pass: 100%/1   | Total:  1h 06m | Avg:  1h 06m | Max:  1h 06m | Hits:   2%/786   
  🟩 MSVC14.39          Pass: 100%/2   | Total:  2h 20m | Avg:  1h 10m | Max:  1h 12m | Hits:   2%/1572  
  🟩 NVHPC24.7          Pass: 100%/2   | Total:  2h 23m | Avg:  1h 11m | Max:  1h 13m
🟩 cxx_family
  🟩 Clang              Pass: 100%/19  | Total:  2h 11m | Avg:  6m 56s | Max: 24m 28s
  🟩 GCC                Pass: 100%/21  | Total:  3h 16m | Avg:  9m 22s | Max: 36m 51s
  🟩 Intel              Pass: 100%/1   | Total:  6m 28s | Avg:  6m 28s | Max:  6m 28s
  🟩 MSVC               Pass: 100%/4   | Total:  4h 27m | Avg:  1h 06m | Max:  1h 12m | Hits:   2%/3144  
  🟩 NVHPC              Pass: 100%/2   | Total:  2h 23m | Avg:  1h 11m | Max:  1h 13m
🟩 gpu
  🟩 h100               Pass: 100%/2   | Total: 20m 22s | Avg: 10m 11s | Max: 15m 51s
  🟩 v100               Pass: 100%/45  | Total: 12h 05m | Avg: 16m 07s | Max:  1h 13m | Hits:   2%/3144  
🟩 jobs
  🟩 Build              Pass: 100%/40  | Total:  9h 49m | Avg: 14m 44s | Max:  1h 13m | Hits:   2%/3144  
  🟩 DeviceLaunch       Pass: 100%/1   | Total: 36m 51s | Avg: 36m 51s | Max: 36m 51s
  🟩 GraphCapture       Pass: 100%/1   | Total: 17m 43s | Avg: 17m 43s | Max: 17m 43s
  🟩 HostLaunch         Pass: 100%/3   | Total: 55m 46s | Avg: 18m 35s | Max: 22m 44s
  🟩 TestGPU            Pass: 100%/2   | Total: 46m 17s | Avg: 23m 08s | Max: 24m 28s
🟩 sm
  🟩 90                 Pass: 100%/2   | Total: 20m 22s | Avg: 10m 11s | Max: 15m 51s
  🟩 90a                Pass: 100%/1   | Total:  4m 20s | Avg:  4m 20s | Max:  4m 20s
🟩 std
  🟩 11                 Pass: 100%/5   | Total: 23m 18s | Avg:  4m 39s | Max:  5m 51s
  🟩 14                 Pass: 100%/4   | Total:  1h 17m | Avg: 19m 21s | Max:  1h 01m | Hits:   2%/786   
  🟩 17                 Pass: 100%/12  | Total:  4h 21m | Avg: 21m 45s | Max:  1h 13m | Hits:   2%/1572  
  🟩 20                 Pass: 100%/26  | Total:  6h 24m | Avg: 14m 46s | Max:  1h 10m | Hits:   2%/786

  • 🟩 thrust: Pass: 100%/46 | Total: 11h 11m | Avg: 14m 35s | Max: 1h 18m | Hits: 30%/9260 

    🟩 cmake_options
  🟩 -DTHRUST_DISPATCH_TYPE=Force32bit Pass: 100%/2   | Total: 17m 55s | Avg:  8m 57s | Max: 12m 16s
🟩 cpu
  🟩 amd64              Pass: 100%/44  | Total: 11h 01m | Avg: 15m 02s | Max:  1h 18m | Hits:  30%/9260  
  🟩 arm64              Pass: 100%/2   | Total:  9m 26s | Avg:  4m 43s | Max:  5m 00s
🟩 ctk
  🟩 11.1               Pass: 100%/7   | Total:  1h 29m | Avg: 12m 49s | Max:  1h 04m | Hits:  12%/1852  
  🟩 12.5               Pass: 100%/2   | Total:  2h 27m | Avg:  1h 13m | Max:  1h 18m
  🟩 12.6               Pass: 100%/37  | Total:  7h 13m | Avg: 11m 42s | Max:  1h 09m | Hits:  34%/7408  
🟩 cudacxx
  🟩 ClangCUDA18        Pass: 100%/2   | Total:  9m 50s | Avg:  4m 55s | Max:  5m 01s
  🟩 nvcc11.1           Pass: 100%/7   | Total:  1h 29m | Avg: 12m 49s | Max:  1h 04m | Hits:  12%/1852  
  🟩 nvcc12.5           Pass: 100%/2   | Total:  2h 27m | Avg:  1h 13m | Max:  1h 18m
  🟩 nvcc12.6           Pass: 100%/35  | Total:  7h 03m | Avg: 12m 06s | Max:  1h 09m | Hits:  34%/7408  
🟩 cudacxx_family
  🟩 ClangCUDA          Pass: 100%/2   | Total:  9m 50s | Avg:  4m 55s | Max:  5m 01s
  🟩 nvcc               Pass: 100%/44  | Total: 11h 01m | Avg: 15m 01s | Max:  1h 18m | Hits:  30%/9260  
🟩 cxx
  🟩 Clang9             Pass: 100%/4   | Total: 20m 42s | Avg:  5m 10s | Max:  6m 44s
  🟩 Clang10            Pass: 100%/1   | Total:  7m 16s | Avg:  7m 16s | Max:  7m 16s
  🟩 Clang11            Pass: 100%/1   | Total:  4m 57s | Avg:  4m 57s | Max:  4m 57s
  🟩 Clang12            Pass: 100%/1   | Total:  5m 22s | Avg:  5m 22s | Max:  5m 22s
  🟩 Clang13            Pass: 100%/1   | Total:  5m 15s | Avg:  5m 15s | Max:  5m 15s
  🟩 Clang14            Pass: 100%/1   | Total:  5m 12s | Avg:  5m 12s | Max:  5m 12s
  🟩 Clang15            Pass: 100%/1   | Total:  5m 45s | Avg:  5m 45s | Max:  5m 45s
  🟩 Clang16            Pass: 100%/1   | Total:  5m 28s | Avg:  5m 28s | Max:  5m 28s
  🟩 Clang17            Pass: 100%/1   | Total:  5m 54s | Avg:  5m 54s | Max:  5m 54s
  🟩 Clang18            Pass: 100%/7   | Total: 47m 56s | Avg:  6m 50s | Max: 14m 58s
  🟩 GCC6               Pass: 100%/2   | Total:  8m 01s | Avg:  4m 00s | Max:  4m 20s
  🟩 GCC7               Pass: 100%/2   | Total: 10m 23s | Avg:  5m 11s | Max:  5m 23s
  🟩 GCC8               Pass: 100%/1   | Total:  5m 36s | Avg:  5m 36s | Max:  5m 36s
  🟩 GCC9               Pass: 100%/3   | Total: 14m 16s | Avg:  4m 45s | Max:  5m 54s
  🟩 GCC10              Pass: 100%/1   | Total:  5m 21s | Avg:  5m 21s | Max:  5m 21s
  🟩 GCC11              Pass: 100%/1   | Total:  5m 47s | Avg:  5m 47s | Max:  5m 47s
  🟩 GCC12              Pass: 100%/1   | Total:  6m 07s | Avg:  6m 07s | Max:  6m 07s
  🟩 GCC13              Pass: 100%/8   | Total:  1h 00m | Avg:  7m 30s | Max: 12m 16s
  🟩 Intel2023.2.0      Pass: 100%/1   | Total:  7m 10s | Avg:  7m 10s | Max:  7m 10s
  🟩 MSVC14.16          Pass: 100%/1   | Total:  1h 04m | Avg:  1h 04m | Max:  1h 04m | Hits:  12%/1852  
  🟩 MSVC14.29          Pass: 100%/1   | Total:  1h 04m | Avg:  1h 04m | Max:  1h 04m | Hits:  12%/1852  
  🟩 MSVC14.39          Pass: 100%/3   | Total:  2h 37m | Avg: 52m 23s | Max:  1h 09m | Hits:  41%/5556  
  🟩 NVHPC24.7          Pass: 100%/2   | Total:  2h 27m | Avg:  1h 13m | Max:  1h 18m
🟩 cxx_family
  🟩 Clang              Pass: 100%/19  | Total:  1h 53m | Avg:  5m 59s | Max: 14m 58s
  🟩 GCC                Pass: 100%/19  | Total:  1h 55m | Avg:  6m 05s | Max: 12m 16s
  🟩 Intel              Pass: 100%/1   | Total:  7m 10s | Avg:  7m 10s | Max:  7m 10s
  🟩 MSVC               Pass: 100%/5   | Total:  4h 46m | Avg: 57m 21s | Max:  1h 09m | Hits:  30%/9260  
  🟩 NVHPC              Pass: 100%/2   | Total:  2h 27m | Avg:  1h 13m | Max:  1h 18m
🟩 gpu
  🟩 v100               Pass: 100%/46  | Total: 11h 11m | Avg: 14m 35s | Max:  1h 18m | Hits:  30%/9260  
🟩 jobs
  🟩 Build              Pass: 100%/40  | Total:  9h 54m | Avg: 14m 52s | Max:  1h 18m | Hits:  12%/7408  
  🟩 TestCPU            Pass: 100%/3   | Total: 37m 55s | Avg: 12m 38s | Max: 21m 13s | Hits:  99%/1852  
  🟩 TestGPU            Pass: 100%/3   | Total: 38m 25s | Avg: 12m 48s | Max: 14m 58s
🟩 sm
  🟩 90a                Pass: 100%/1   | Total:  4m 57s | Avg:  4m 57s | Max:  4m 57s
🟩 std
  🟩 11                 Pass: 100%/5   | Total: 21m 49s | Avg:  4m 21s | Max:  5m 20s
  🟩 14                 Pass: 100%/4   | Total:  1h 21m | Avg: 20m 18s | Max:  1h 04m | Hits:  12%/1852  
  🟩 17                 Pass: 100%/12  | Total:  4h 12m | Avg: 21m 02s | Max:  1h 09m | Hits:  12%/3704  
  🟩 20                 Pass: 100%/23  | Total:  4h 57m | Avg: 12m 56s | Max:  1h 18m | Hits:  56%/3704

  • 🟩 cudax: Pass: 100%/26 | Total: 2h 18m | Avg: 5m 20s | Max: 20m 04s | Hits: 34%/312 

    🟩 cpu
  🟩 amd64              Pass: 100%/22  | Total:  2h 08m | Avg:  5m 50s | Max: 20m 04s | Hits:  34%/312   
  🟩 arm64              Pass: 100%/4   | Total: 10m 25s | Avg:  2m 36s | Max:  2m 41s
🟩 ctk
  🟩 12.0               Pass: 100%/3   | Total: 16m 14s | Avg:  5m 24s | Max:  9m 54s | Hits:  34%/156   
  🟩 12.5               Pass: 100%/2   | Total: 17m 28s | Avg:  8m 44s | Max:  8m 53s
  🟩 12.6               Pass: 100%/21  | Total:  1h 45m | Avg:  5m 00s | Max: 20m 04s | Hits:  34%/156   
🟩 cudacxx
  🟩 nvcc12.0           Pass: 100%/3   | Total: 16m 14s | Avg:  5m 24s | Max:  9m 54s | Hits:  34%/156   
  🟩 nvcc12.5           Pass: 100%/2   | Total: 17m 28s | Avg:  8m 44s | Max:  8m 53s
  🟩 nvcc12.6           Pass: 100%/21  | Total:  1h 45m | Avg:  5m 00s | Max: 20m 04s | Hits:  34%/156   
🟩 cudacxx_family
  🟩 nvcc               Pass: 100%/26  | Total:  2h 18m | Avg:  5m 20s | Max: 20m 04s | Hits:  34%/312   
🟩 cxx
  🟩 Clang9             Pass: 100%/1   | Total:  3m 10s | Avg:  3m 10s | Max:  3m 10s
  🟩 Clang10            Pass: 100%/1   | Total:  3m 27s | Avg:  3m 27s | Max:  3m 27s
  🟩 Clang11            Pass: 100%/1   | Total:  2m 56s | Avg:  2m 56s | Max:  2m 56s
  🟩 Clang12            Pass: 100%/1   | Total:  3m 38s | Avg:  3m 38s | Max:  3m 38s
  🟩 Clang13            Pass: 100%/1   | Total:  3m 10s | Avg:  3m 10s | Max:  3m 10s
  🟩 Clang14            Pass: 100%/1   | Total:  3m 07s | Avg:  3m 07s | Max:  3m 07s
  🟩 Clang15            Pass: 100%/1   | Total:  3m 27s | Avg:  3m 27s | Max:  3m 27s
  🟩 Clang16            Pass: 100%/1   | Total:  3m 07s | Avg:  3m 07s | Max:  3m 07s
  🟩 Clang17            Pass: 100%/1   | Total:  3m 20s | Avg:  3m 20s | Max:  3m 20s
  🟩 Clang18            Pass: 100%/4   | Total: 27m 41s | Avg:  6m 55s | Max: 19m 27s
  🟩 GCC9               Pass: 100%/1   | Total:  3m 10s | Avg:  3m 10s | Max:  3m 10s
  🟩 GCC10              Pass: 100%/1   | Total:  3m 09s | Avg:  3m 09s | Max:  3m 09s
  🟩 GCC11              Pass: 100%/1   | Total:  3m 19s | Avg:  3m 19s | Max:  3m 19s
  🟩 GCC12              Pass: 100%/2   | Total: 23m 19s | Avg: 11m 39s | Max: 20m 04s
  🟩 GCC13              Pass: 100%/4   | Total: 10m 56s | Avg:  2m 44s | Max:  2m 56s
  🟩 MSVC14.36          Pass: 100%/1   | Total:  9m 54s | Avg:  9m 54s | Max:  9m 54s | Hits:  34%/156   
  🟩 MSVC14.39          Pass: 100%/1   | Total: 10m 28s | Avg: 10m 28s | Max: 10m 28s | Hits:  34%/156   
  🟩 NVHPC24.7          Pass: 100%/2   | Total: 17m 28s | Avg:  8m 44s | Max:  8m 53s
🟩 cxx_family
  🟩 Clang              Pass: 100%/13  | Total: 57m 03s | Avg:  4m 23s | Max: 19m 27s
  🟩 GCC                Pass: 100%/9   | Total: 43m 53s | Avg:  4m 52s | Max: 20m 04s
  🟩 MSVC               Pass: 100%/2   | Total: 20m 22s | Avg: 10m 11s | Max: 10m 28s | Hits:  34%/312   
  🟩 NVHPC              Pass: 100%/2   | Total: 17m 28s | Avg:  8m 44s | Max:  8m 53s
🟩 gpu
  🟩 v100               Pass: 100%/26  | Total:  2h 18m | Avg:  5m 20s | Max: 20m 04s | Hits:  34%/312   
🟩 jobs
  🟩 Build              Pass: 100%/24  | Total:  1h 39m | Avg:  4m 08s | Max: 10m 28s | Hits:  34%/312   
  🟩 Test               Pass: 100%/2   | Total: 39m 31s | Avg: 19m 45s | Max: 20m 04s
🟩 sm
  🟩 90                 Pass: 100%/1   | Total:  2m 50s | Avg:  2m 50s | Max:  2m 50s
  🟩 90a                Pass: 100%/1   | Total:  2m 56s | Avg:  2m 56s | Max:  2m 56s
🟩 std
  🟩 17                 Pass: 100%/6   | Total: 23m 10s | Avg:  3m 51s | Max:  8m 53s
  🟩 20                 Pass: 100%/20  | Total:  1h 55m | Avg:  5m 46s | Max: 20m 04s | Hits:  34%/312

  • 🟩 cccl_c_parallel: Pass: 100%/2 | Total: 10m 28s | Avg: 5m 14s | Max: 8m 31s

     
    🟩 cpu
  🟩 amd64              Pass: 100%/2   | Total: 10m 28s | Avg:  5m 14s | Max:  8m 31s
🟩 ctk
  🟩 12.6               Pass: 100%/2   | Total: 10m 28s | Avg:  5m 14s | Max:  8m 31s
🟩 cudacxx
  🟩 nvcc12.6           Pass: 100%/2   | Total: 10m 28s | Avg:  5m 14s | Max:  8m 31s
🟩 cudacxx_family
  🟩 nvcc               Pass: 100%/2   | Total: 10m 28s | Avg:  5m 14s | Max:  8m 31s
🟩 cxx
  🟩 GCC13              Pass: 100%/2   | Total: 10m 28s | Avg:  5m 14s | Max:  8m 31s
🟩 cxx_family
  🟩 GCC                Pass: 100%/2   | Total: 10m 28s | Avg:  5m 14s | Max:  8m 31s
🟩 gpu
  🟩 v100               Pass: 100%/2   | Total: 10m 28s | Avg:  5m 14s | Max:  8m 31s
🟩 jobs
  🟩 Build              Pass: 100%/1   | Total:  1m 57s | Avg:  1m 57s | Max:  1m 57s
  🟩 Test               Pass: 100%/1   | Total:  8m 31s | Avg:  8m 31s | Max:  8m 31s

  • 🟩 python: Pass: 100%/1 | Total: 35m 00s | Avg: 35m 00s | Max: 35m 00s

     
    🟩 cpu
  🟩 amd64              Pass: 100%/1   | Total: 35m 00s | Avg: 35m 00s | Max: 35m 00s
🟩 ctk
  🟩 12.6               Pass: 100%/1   | Total: 35m 00s | Avg: 35m 00s | Max: 35m 00s
🟩 cudacxx
  🟩 nvcc12.6           Pass: 100%/1   | Total: 35m 00s | Avg: 35m 00s | Max: 35m 00s
🟩 cudacxx_family
  🟩 nvcc               Pass: 100%/1   | Total: 35m 00s | Avg: 35m 00s | Max: 35m 00s
🟩 cxx
  🟩 GCC13              Pass: 100%/1   | Total: 35m 00s | Avg: 35m 00s | Max: 35m 00s
🟩 cxx_family
  🟩 GCC                Pass: 100%/1   | Total: 35m 00s | Avg: 35m 00s | Max: 35m 00s
🟩 gpu
  🟩 v100               Pass: 100%/1   | Total: 35m 00s | Avg: 35m 00s | Max: 35m 00s
🟩 jobs
  🟩 Test               Pass: 100%/1   | Total: 35m 00s | Avg: 35m 00s | Max: 35m 00s

👃 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: 170)
# Runner
125 linux-amd64-cpu16
19 linux-amd64-gpu-v100-latest-1
15 windows-amd64-cpu16
10 linux-arm64-cpu16
1 linux-amd64-gpu-h100-latest-1-testing

@ericniebler ericniebler enabled auto-merge (squash) December 21, 2024 01:47
@ericniebler ericniebler merged commit faca86c into NVIDIA:main Dec 21, 2024
185 checks passed
@ericniebler ericniebler deleted the cpp17-contiguous-ranges branch December 23, 2024 18:44
davebayer pushed a commit to davebayer/cccl that referenced this pull request Jan 18, 2025
@ericniebler @davebayer
help the ranges concepts recognize standard contiguous iterators in c…
549fe69 
…++14/17 (NVIDIA#3202)
davebayer added a commit to davebayer/cccl that referenced this pull request Jan 20, 2025
@davebayer
add builtins
bf99eb3 
implement `add_sat`

split `signed`/`unsigned` implementation, improve implementation for MSVC

improve device `add_sat` implementation

add `add_sat` test

improve generic `add_sat` implementation for signed types

implement `sub_sat`

allow more msvc intrinsics on x86

add op tests

partially implement `mul_sat`

implement `div_sat` and `saturate_cast`

add `saturate_cast` test

simplify `div_sat` test

Deprectate C++11 and C++14 for libcu++ (#3173)

* Deprectate C++11 and C++14 for libcu++

Co-authored-by: Bernhard Manfred Gruber <[email protected]>

Implement `abs` and `div` from `cstdlib` (#3153)

* implement integer abs functions
* improve tests, fix constexpr support
* just use the our implementation
* implement `cuda::std::div`
* prefer host's `div_t` like types
* provide `cuda::std::abs` overloads for floats
* allow fp abs for NVRTC
* silence msvc's warning about conversion from floating point to integral

Fix missing radix sort policies (#3174)

Fixes NVBug 5009941

Introduces new `DeviceReduce::Arg{Min,Max}` interface with two output iterators (#3148)

* introduces new arg{min,max} interface with two output iterators

* adds fp inf tests

* fixes docs

* improves code example

* fixes exec space specifier

* trying to fix deprecation warning for more compilers

* inlines unzip operator

* trying to fix deprecation warning for nvhpc

* integrates supression fixes in diagnostics

* pre-ctk 11.5 deprecation suppression

* fixes icc

* fix for pre-ctk11.5

* cleans up deprecation suppression

* cleanup

Extend tuning documentation (#3179)

Add codespell pre-commit hook, fix typos in CCCL (#3168)

* Add codespell pre-commit hook
* Automatic changes from codespell.
* Manual changes.

Fix parameter space for TUNE_LOAD in scan benchmark (#3176)

fix various old compiler checks (#3178)

implement C++26 `std::projected` (#3175)

Fix pre-commit config for codespell and remaining typos (#3182)

Massive cleanup of our config (#3155)

Fix UB in atomics with automatic storage (#2586)

* Adds specialized local cuda atomics and injects them into most atomics paths.

Co-authored-by: Georgy Evtushenko <[email protected]>
Co-authored-by: gonzalobg <[email protected]>

* Allow CUDA 12.2 to keep perf, this addresses earlier comments in #478

* Remove extraneous double brackets in unformatted code.

* Merge unsafe atomic logic into `__cuda_is_local`.

* Use `const_cast` for type conversions in cuda_local.h

* Fix build issues from interface changes

* Fix missing __nanosleep on sm70-

* Guard __isLocal from NVHPC

* Use PTX instead of running nothing from NVHPC

* fixup /s/nvrtc/nvhpc

* Fixup missing CUDA ifdef surrounding device code

* Fix codegen

* Bypass some sort of compiler bug on GCC7

* Apply suggestions from code review

* Use unsafe automatic storage atomics in codegen tests

---------

Co-authored-by: Georgy Evtushenko <[email protected]>
Co-authored-by: gonzalobg <[email protected]>
Co-authored-by: Michael Schellenberger Costa <[email protected]>

Refactor the source code layout for `cuda.parallel` (#3177)

* Refactor the source layout for cuda.parallel

* Add copyright

* Address review feedback

* Don't import anything into `experimental` namespace

* fix import

---------

Co-authored-by: Ashwin Srinath <[email protected]>

new type-erased memory resources (#2824)

s/_LIBCUDACXX_DECLSPEC_EMPTY_BASES/_CCCL_DECLSPEC_EMPTY_BASES/g (#3186)

Document address stability of `thrust::transform` (#3181)

* Do not document _LIBCUDACXX_MARK_CAN_COPY_ARGUMENTS
* Reformat and fix UnaryFunction/BinaryFunction in transform docs
* Mention transform can use proclaim_copyable_arguments
* Document cuda::proclaims_copyable_arguments better
* Deprecate depending on transform functor argument addresses

Fixes: #3053

turn off cuda version check for clangd (#3194)

[STF] jacobi example based on parallel_for (#3187)

* Simple jacobi example with parallel for and reductions

* clang-format

* remove useless capture list

fixes pre-nv_diag suppression issues (#3189)

Prefer c2h::type_name over c2h::demangle (#3195)

Fix memcpy_async* tests (#3197)

* memcpy_async_tx: Fix bug in test

Two bugs, one of which occurs in practice:

1. There is a missing fence.proxy.space::global between the writes to
   global memory and the memcpy_async_tx. (Occurs in practice)

2. The end of the kernel should be fenced with `__syncthreads()`,
   because the barrier is invalidated in the destructor. If other
   threads are still waiting on it, there will be UB. (Has not yet
   manifested itself)

* cp_async_bulk_tensor: Pre-emptively fence more in test

Add type annotations and mypy checks for `cuda.parallel`  (#3180)

* Refactor the source layout for cuda.parallel

* Add initial type annotations

* Update pre-commit config

* More typing

* Fix bad merge

* Fix TYPE_CHECKING and numpy annotations

* typing bindings.py correctly

* Address review feedback

---------

Co-authored-by: Ashwin Srinath <[email protected]>

Fix rendering of cuda.parallel docs (#3192)

* Fix pre-commit config for codespell and remaining typos

* Fix rendering of docs for cuda.parallel

---------

Co-authored-by: Ashwin Srinath <[email protected]>

Enable PDL for DeviceMergeSortBlockSortKernel (#3199)

The kernel already contains a call to _CCCL_PDL_GRID_DEPENDENCY_SYNC.
This commit enables PDL when launching the kernel.

Adds support for large `num_items` to `DeviceReduce::{ArgMin,ArgMax}` (#2647)

* adds benchmarks for reduce::arg{min,max}

* preliminary streaming arg-extremum reduction

* fixes implicit conversion

* uses streaming dispatch class

* changes arg benches to use new streaming reduce

* streaming arg-extrema reduction

* fixes style

* fixes compilation failures

* cleanups

* adds rst style comments

* declare vars const and use clamp

* consolidates argmin argmax benchmarks

* fixes thrust usage

* drops offset type in arg-extrema benchmarks

* fixes clang cuda

* exec space macros

* switch to signed global offset type for slightly better perf

* clarifies documentation

* applies minor benchmark style changes from review comments

* fixes interface documentation and comments

* list-init accumulating output op

* improves style, comments, and tests

* cleans up aggregate init

* renames dispatch class usage in benchmarks

* fixes merge conflicts

* addresses review comments

* addresses review comments

* fixes assertion

* removes superseded implementation

* changes large problem tests to use new interface

* removes obsolete tests for deprecated interface

Fixes for Python 3.7 docs environment (#3206)

Co-authored-by: Ashwin Srinath <[email protected]>

Adds support for large number of items to `DeviceTransform` (#3172)

* moves large problem test helper to common file

* adds support for large num items to device transform

* adds tests for large number of items to device interface

* fixes format

* addresses review comments

cp_async_bulk: Fix test (#3198)

* memcpy_async_tx: Fix bug in test

Two bugs, one of which occurs in practice:

1. There is a missing fence.proxy.space::global between the writes to
   global memory and the memcpy_async_tx. (Occurs in practice)

2. The end of the kernel should be fenced with `__syncthreads()`,
   because the barrier is invalidated in the destructor. If other
   threads are still waiting on it, there will be UB. (Has not yet
   manifested itself)

* cp_async_bulk_tensor: Pre-emptively fence more in test

* cp_async_bulk: Fix test

The global memory pointer could be misaligned.

cudax fixes for msvc 14.41 (#3200)

avoid instantiating class templates in `is_same` implementation when possible (#3203)

Fix: make launchers a CUB detail; make kernel source functions hidden. (#3209)

* Fix: make launchers a CUB detail; make kernel source functions hidden.

* [pre-commit.ci] auto code formatting

* Address review comments, fix which macro gets fixed.

help the ranges concepts recognize standard contiguous iterators in c++14/17 (#3202)

unify macros and cmake options that control the suppression of deprecation warnings (#3220)

* unify macros and cmake options that control the suppression of deprecation warnings

* suppress nvcc warning #186 in thrust header tests

* suppress c++ dialect deprecation warnings in libcudacxx header tests

Fx thread-reduce performance regression (#3225)

cuda.parallel: In-memory caching of build objects (#3216)

* Define __eq__ and __hash__ for Iterators

* Define cache_with_key utility and use it to cache Reduce objects

* Add tests for caching Reduce objects

* Tighten up types

* Updates to support 3.7

* Address review feedback

* Introduce IteratorKind to hold iterator type information

* Use the .kind to generate an abi_name

* Remove __eq__ and __hash__ methods from IteratorBase

* Move helper function

* Formatting

* Don't unpack tuple in cache key

---------

Co-authored-by: Ashwin Srinath <[email protected]>

Just enough ranges for c++14 `span` (#3211)

use generalized concepts portability macros to simplify the `range` concept (#3217)

fixes some issues in the concepts portability macros and then re-implements the `range` concept with `_CCCL_REQUIRES_EXPR`

Use Ruff to sort imports (#3230)

* Update pyproject.tomls for import sorting

* Update files after running pre-commit

* Move ruff config to pyproject.toml

---------

Co-authored-by: Ashwin Srinath <[email protected]>

fix tuning_scan sm90 config issue (#3236)

Co-authored-by: Shijie Chen <[email protected]>

[STF] Logical token (#3196)

* Split the implementation of the void interface into the definition of the interface, and its implementations on streams and graphs.

* Add missing files

* Check if a task implementation can match a prototype where the void_interface arguments are ignored

* Implement ctx.abstract_logical_data() which relies on a void data interface

* Illustrate how to use abstract handles in local contexts

* Introduce an is_void_interface() virtual method in the data interface to potentially optimize some stages

* Small improvements in the examples

* Do not try to allocate or move void data

* Do not use I as a variable

* fix linkage error

* rename abtract_logical_data into logical_token

* Document logical token

* fix spelling error

* fix sphinx error

* reflect name changes

* use meaningful variable names

* simplify logical_token implementation because writeback is already disabled

* add a unit test for token elision

* implement token elision in host_launch

* Remove unused type

* Implement helpers to check if a function can be invoked from a tuple, or from a tuple where we removed tokens

* Much simpler is_tuple_invocable_with_filtered implementation

* Fix buggy test

* Factorize code

* Document that we can ignore tokens for task and host_launch

* Documentation for logical data freeze

Fix ReduceByKey tuning (#3240)

Fix RLE tuning (#3239)

cuda.parallel: Forbid non-contiguous arrays as inputs (or outputs) (#3233)

* Forbid non-contiguous arrays as inputs (or outputs)

* Implement a more robust way to check for contiguity

* Don't bother if cublas unavailable

* Fix how we check for zero-element arrays

* sort imports

---------

Co-authored-by: Ashwin Srinath <[email protected]>

expands support for more offset types in segmented benchmark (#3231)

Add escape hatches to the cmake configuration of the header tests so that we can tests deprecated compilers / dialects (#3253)

* Add escape hatches to the cmake configuration of the header tests so that we can tests deprecated compilers / dialects

* Do not add option twice

ptx: Add add_instruction.py (#3190)

This file helps create the necessary structure for new PTX instructions.

Co-authored-by: Allard Hendriksen <[email protected]>

Bump main to 2.9.0. (#3247)

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

Drop cub::Mutex (#3251)

Fixes: #3250

Remove legacy macros from CUB util_arch.cuh (#3257)

Fixes: #3256

Remove thrust::[unary|binary]_traits (#3260)

Fixes: #3259

Architecture and OS identification macros (#3237)

Bump main to 3.0.0. (#3265)

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

Drop thrust not1 and not2 (#3264)

Fixes: #3263

CCCL Internal macro documentation (#3238)

Deprecate GridBarrier and GridBarrierLifetime (#3258)

Fixes: #1389

Require at least gcc7 (#3268)

Fixes: #3267

Drop thrust::[unary|binary]_function (#3274)

Fixes: #3273

Drop ICC from CI (#3277)

[STF] Corruption of the capture list of an extended lambda with a parallel_for construct on a host execution place (#3270)

* Add a test to reproduce a bug observed with parallel_for on a host place

* clang-format

* use _CCCL_ASSERT

* Attempt to debug

* do not create a tuple with a universal reference that is out of scope when we use it, use an lvalue instead

* fix lambda expression

* clang-format

Enable thrust::identity test for non-MSVC (#3281)

This seems to be an oversight when the test was added

Co-authored-by: Michael Schellenberger Costa <[email protected]>

Enable PDL in triple chevron launch (#3282)

It seems PDL was disabled by accident when _THRUST_HAS_PDL was renamed
to _CCCL_HAS_PDL during the review introducing the feature.

Disambiguate line continuations and macro continuations in <nv/target> (#3244)

Drop VS 2017 from CI (#3287)

Fixes: #3286

Drop ICC support in code (#3279)

* Drop ICC from code

Fixes: #3278

Co-authored-by: Michael Schellenberger Costa <[email protected]>

Make CUB NVRTC commandline arguments come from a cmake template (#3292)

Propose the same components (thrust, cub, libc++, cudax, cuda.parallel,...) in the bug report template than in the feature request template (#3295)

Use process isolation instead of default hyper-v for Windows. (#3294)

Try improving build times by using process isolation instead of hyper-v

Co-authored-by: Michael Schellenberger Costa <[email protected]>

[pre-commit.ci] pre-commit autoupdate (#3248)

* [pre-commit.ci] pre-commit autoupdate

updates:
- [github.com/pre-commit/mirrors-clang-format: v18.1.8 → v19.1.6](https://github.com/pre-commit/mirrors-clang-format/compare/v18.1.8...v19.1.6)
- [github.com/astral-sh/ruff-pre-commit: v0.8.3 → v0.8.6](https://github.com/astral-sh/ruff-pre-commit/compare/v0.8.3...v0.8.6)
- [github.com/pre-commit/mirrors-mypy: v1.13.0 → v1.14.1](https://github.com/pre-commit/mirrors-mypy/compare/v1.13.0...v1.14.1)

Co-authored-by: Michael Schellenberger Costa <[email protected]>

Drop Thrust legacy arch macros (#3298)

Which were disabled and could be re-enabled using THRUST_PROVIDE_LEGACY_ARCH_MACROS

Drop Thrust's compiler_fence.h (#3300)

Drop CTK 11.x from CI (#3275)

* Add cuda12.0-gcc7 devcontainer
* Move MSVC2017 jobs to CTK 12.6
Those is the only combination where rapidsai has devcontainers
* Add /Zc:__cplusplus for the libcudacxx tests
* Only add excape hatch for affected CTKs
* Workaround missing cudaLaunchKernelEx on MSVC
cudaLaunchKernelEx requires C++11, but unfortunately <cuda_runtime.h> checks this using the __cplusplus macro, which is reported wrongly for MSVC. CTK 12.3 fixed this by additionally detecting _MSV_VER. As a workaround, we provide our own copy of cudaLaunchKernelEx when it is not available from the CTK.
* Workaround nvcc+MSVC issue
* Regenerate devcontainers

Fixes: #3249

Co-authored-by: Michael Schellenberger Costa <[email protected]>

Drop CUB's util_compiler.cuh (#3302)

All contained macros were deprecated

Update packman and repo_docs versions (#3293)

Co-authored-by: Ashwin Srinath <[email protected]>

Drop Thrust's deprecated compiler macros (#3301)

Drop CUB_RUNTIME_ENABLED and __THRUST_HAS_CUDART__ (#3305)

Adds support for large number of items to `DevicePartition::If` with the `ThreeWayPartition` overload (#2506)

* adds support for large number of items to three-way partition

* adapts interface to use choose_signed_offset_t

* integrates applicable feedback from device-select pr

* changes behavior for empty problems

* unifies grid constant macro

* fixes kernel template specialization mismatch

* integrates _CCCL_GRID_CONSTANT changes

* resolve merge conflicts

* fixes checks in test

* fixes test verification

* improves tests

* makes few improvements to streaming dispatch

* improves code comment on test

* fixes unrelated compiler error

* minor style improvements

Refactor scan tunings (#3262)

Require C++17 for compiling Thrust and CUB (#3255)

* Issue an unsuppressable warning when compiling with < C++17
* Remove C++11/14 presets
* Remove CCCL_IGNORE_DEPRECATED_CPP_DIALECT from headers
* Remove [CUB|THRUST|TCT]_IGNORE_DEPRECATED_CPP_[11|14]
* Remove CUB_ENABLE_DIALECT_CPP[11|14]
* Update CI runs
* Remove C++11/14 CI runs for CUB and Thrust
* Raise compiler minimum versions for C++17
* Update ReadMe
* Drop Thrust's cpp14_required.h
* Add escape hatch for C++17 removal

Fixes: #3252

Implement `views::empty` (#3254)

* Disable pair conversion of subrange with clang in C++17

* Fix namespace views

* Implement `views::empty`

This implements `std::ranges::views::empty`, see https://en.cppreference.com/w/cpp/ranges/empty_view

Refactor `limits` and `climits` (#3221)

* implement builtins for huge val, nan and nans

* change `INFINITY` and `NAN` implementation for NVRTC

cuda.parallel: Add documentation for the current iterators along with examples and tests (#3311)

* Add tests demonstrating usage of different iterators

* Update documentation of reduce_into by merging import code snippet with the rest of the example

* Add documentation for current iterators

* Run pre-commit checks and update accordingly

* Fix comments to refer to the proper lines in the code snippets in the docs

Drop clang<14 from CI, update devcontainers. (#3309)

Co-authored-by: Bernhard Manfred Gruber <[email protected]>

[STF] Cleanup task dependencies object constructors (#3291)

* Define tag types for access modes

* - Rework how we build task_dep objects based on access mode tags
- pack_state is now responsible for using a const_cast for read only data

* Greatly simplify the previous attempt : do not define new types, but use integral constants based on the enums

* It seems the const_cast was not necessarily so we can simplify it and not even do some dispatch based on access modes

Disable test with a gcc-14 regression (#3297)

Deprecate Thrust's cpp_compatibility.h macros (#3299)

Remove dropped function objects from docs (#3319)

Document `NV_TARGET` macros (#3313)

[STF] Define ctx.pick_stream() which was missing for the unified context (#3326)

* Define ctx.pick_stream() which was missing for the unified context

* clang-format

Deprecate cub::IterateThreadStore (#3337)

Drop CUB's BinaryFlip operator (#3332)

Deprecate cub::Swap (#3333)

Clarify transform output can overlap input (#3323)

Drop CUB APIs with a debug_synchronous parameter (#3330)

Fixes: #3329

Drop CUB's util_compiler.cuh for real (#3340)

PR #3302 planned to drop the file, but only dropped its content. This
was an oversight. So let's drop the entire file.

Drop cub::ValueCache (#3346)

limits offset types for merge sort (#3328)

Drop CDPv1 (#3344)

Fixes: #3341

Drop thrust::void_t (#3362)

Use cuda::std::addressof in Thrust (#3363)

Fix all_of documentation for empty ranges (#3358)

all_of always returns true on an empty range.

[STF] Do not keep track of dangling events in a CUDA graph backend (#3327)

* Unlike the CUDA stream backend, nodes in a CUDA graph are necessarily done when
the CUDA graph completes. Therefore keeping track of "dangling events" is a
waste of time and resources.

* replace can_ignore_dangling_events by track_dangling_events which leads to more readable code

* When not storing the dangling events, we must still perform the deinit operations that were producing these events !

Extract scan kernels into NVRTC-compilable header (#3334)

* Extract scan kernels into NVRTC-compilable header

* Update cub/cub/device/dispatch/dispatch_scan.cuh

Co-authored-by: Georgii Evtushenko <[email protected]>

---------

Co-authored-by: Ashwin Srinath <[email protected]>
Co-authored-by: Georgii Evtushenko <[email protected]>

Drop deprecated aliases in Thrust functional (#3272)

Fixes: #3271

Drop cub::DivideAndRoundUp (#3347)

Use cuda::std::min/max in Thrust (#3364)

Implement `cuda::std::numeric_limits` for `__half` and `__nv_bfloat16` (#3361)

* implement `cuda::std::numeric_limits` for `__half` and `__nv_bfloat16`

Cleanup util_arch (#2773)

Deprecate thrust::null_type (#3367)

Deprecate cub::DeviceSpmv (#3320)

Fixes: #896

Improves `DeviceSegmentedSort` test run time for large number of items and segments (#3246)

* fixes segment offset generation

* switches to analytical verification

* switches to analytical verification for pairs

* fixes spelling

* adds tests for large number of segments

* fixes narrowing conversion in tests

* addresses review comments

* fixes includes

Compile basic infra test with C++17 (#3377)

Adds support for large number of items and large number of segments to `DeviceSegmentedSort` (#3308)

* fixes segment offset generation

* switches to analytical verification

* switches to analytical verification for pairs

* addresses review comments

* introduces segment offset type

* adds tests for large number of segments

* adds support for large number of segments

* drops segment offset type

* fixes thrust namespace

* removes about-to-be-deprecated cub iterators

* no exec specifier on defaulted ctor

* fixes gcc7 linker error

* uses local_segment_index_t throughout

* determine offset type based on type returned by segment iterator begin/end iterators

* minor style improvements

Exit with error when RAPIDS CI fails. (#3385)

cuda.parallel: Support structured types as algorithm inputs (#3218)

* Introduce gpu_struct decorator and typing

* Enable `reduce` to accept arrays of structs as inputs

* Add test for reducing arrays-of-struct

* Update documentation

* Use a numpy array rather than ctypes object

* Change zeros -> empty for output array and temp storage

* Add a TODO for typing GpuStruct

* Documentation udpates

* Remove test_reduce_struct_type from test_reduce.py

* Revert to `to_cccl_value()` accepting ndarray + GpuStruct

* Bump copyrights

---------

Co-authored-by: Ashwin Srinath <[email protected]>

Deprecate thrust::async (#3324)

Fixes: #100

Review/Deprecate CUB `util.ptx` for CCCL 2.x (#3342)

Fix broken `_CCCL_BUILTIN_ASSUME` macro (#3314)

* add compiler-specific path
* fix device code path
* add _CCC_ASSUME

Deprecate thrust::numeric_limits (#3366)

Replace `typedef` with `using` in libcu++ (#3368)

Deprecate thrust::optional (#3307)

Fixes: #3306

Upgrade to Catch2 3.8  (#3310)

Fixes: #1724

refactor `<cuda/std/cstdint>` (#3325)

Co-authored-by: Bernhard Manfred Gruber <[email protected]>

Update CODEOWNERS (#3331)

* Update CODEOWNERS

* Update CODEOWNERS

* Update CODEOWNERS

* [pre-commit.ci] auto code formatting

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

Fix sign-compare warning (#3408)

Implement more cmath functions to be usable on host and device (#3382)

* Implement more cmath functions to be usable on host and device

* Implement math roots functions

* Implement exponential functions

Redefine and deprecate thrust::remove_cvref (#3394)

* Redefine and deprecate thrust::remove_cvref

Co-authored-by: Michael Schellenberger Costa <[email protected]>

Fix assert definition for NVHPC due to constexpr issues (#3418)

NVHPC cannot decide at compile time where the code would run so _CCCL_ASSERT within a constexpr function breaks it.

Fix this by always using the host definition which should also work on device.

Fixes #3411

Extend CUB reduce benchmarks (#3401)

* Rename max.cu to custom.cu, since it uses a custom operator
* Extend types covered my min.cu to all fundamental types
* Add some notes on how to collect tuning parameters

Fixes: #3283

Update upload-pages-artifact to v3 (#3423)

* Update upload-pages-artifact to v3

* Empty commit

---------

Co-authored-by: Ashwin Srinath <[email protected]>

Replace and deprecate thrust::cuda_cub::terminate (#3421)

`std::linalg` accessors and `transposed_layout` (#2962)

Add round up/down to multiple (#3234)

[FEA]: Introduce Python module with CCCL headers (#3201)

* Add cccl/python/cuda_cccl directory and use from cuda_parallel, cuda_cooperative

* Run `copy_cccl_headers_to_aude_include()` before `setup()`

* Create python/cuda_cccl/cuda/_include/__init__.py, then simply import cuda._include to find the include path.

* Add cuda.cccl._version exactly as for cuda.cooperative and cuda.parallel

* Bug fix: cuda/_include only exists after shutil.copytree() ran.

* Use `f"cuda-cccl @ file://{cccl_path}/python/cuda_cccl"` in setup.py

* Remove CustomBuildCommand, CustomWheelBuild in cuda_parallel/setup.py (they are equivalent to the default functions)

* Replace := operator (needs Python 3.8+)

* Fix oversights: remove `pip3 install ./cuda_cccl` lines from README.md

* Restore original README.md: `pip3 install -e` now works on first pass.

* cuda_cccl/README.md: FOR INTERNAL USE ONLY

* Remove `$pymajor.$pyminor.` prefix in cuda_cccl _version.py (as suggested under https://github.com/NVIDIA/cccl/pull/3201#discussion_r1894035917)

Command used: ci/update_version.sh 2 8 0

* Modernize pyproject.toml, setup.py

Trigger for this change:

* https://github.com/NVIDIA/cccl/pull/3201#discussion_r1894043178

* https://github.com/NVIDIA/cccl/pull/3201#discussion_r1894044996

* Install CCCL headers under cuda.cccl.include

Trigger for this change:

* https://github.com/NVIDIA/cccl/pull/3201#discussion_r1894048562

Unexpected accidental discovery: cuda.cooperative unit tests pass without CCCL headers entirely.

* Factor out cuda_cccl/cuda/cccl/include_paths.py

* Reuse cuda_cccl/cuda/cccl/include_paths.py from cuda_cooperative

* Add missing Copyright notice.

* Add missing __init__.py (cuda.cccl)

* Add `"cuda.cccl"` to `autodoc.mock_imports`

* Move cuda.cccl.include_paths into function where it is used. (Attempt to resolve Build and Verify Docs failure.)

* Add # TODO: move this to a module-level import

* Modernize cuda_cooperative/pyproject.toml, setup.py

* Convert cuda_cooperative to use hatchling as build backend.

* Revert "Convert cuda_cooperative to use hatchling as build backend."

This reverts commit 61637d608da06fcf6851ef6197f88b5e7dbc3bbe.

* Move numpy from [build-system] requires -> [project] dependencies

* Move pyproject.toml [project] dependencies -> setup.py install_requires, to be able to use CCCL_PATH

* Remove copy_license() and use license_files=["../../LICENSE"] instead.

* Further modernize cuda_cccl/setup.py to use pathlib

* Trivial simplifications in cuda_cccl/pyproject.toml

* Further simplify cuda_cccl/pyproject.toml, setup.py: remove inconsequential code

* Make cuda_cooperative/pyproject.toml more similar to cuda_cccl/pyproject.toml

* Add taplo-pre-commit to .pre-commit-config.yaml

* taplo-pre-commit auto-fixes

* Use pathlib in cuda_cooperative/setup.py

* CCCL_PYTHON_PATH in cuda_cooperative/setup.py

* Modernize cuda_parallel/pyproject.toml, setup.py

* Use pathlib in cuda_parallel/setup.py

* Add `# TOML lint & format` comment.

* Replace MANIFEST.in with `[tool.setuptools.package-data]` section in pyproject.toml

* Use pathlib in cuda/cccl/include_paths.py

* pre-commit autoupdate (EXCEPT clang-format, which was manually restored)

* Fixes after git merge main

* Resolve warning: AttributeError: '_Reduce' object has no attribute 'build_result'

```
=========================================================================== warnings summary ===========================================================================
tests/test_reduce.py::test_reduce_non_contiguous
  /home/coder/cccl/python/devenv/lib/python3.12/site-packages/_pytest/unraisableexception.py:85: PytestUnraisableExceptionWarning: Exception ignored in: <function _Reduce.__del__ at 0x7bf123139080>

  Traceback (most recent call last):
    File "/home/coder/cccl/python/cuda_parallel/cuda/parallel/experimental/algorithms/reduce.py", line 132, in __del__
      bindings.cccl_device_reduce_cleanup(ctypes.byref(self.build_result))
                                                       ^^^^^^^^^^^^^^^^^
  AttributeError: '_Reduce' object has no attribute 'build_result'

    warnings.warn(pytest.PytestUnraisableExceptionWarning(msg))

-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
============================================================= 1 passed, 93 deselected, 1 warning in 0.44s ==============================================================
```

* Move `copy_cccl_headers_to_cuda_cccl_include()` functionality to `class CustomBuildPy`

* Introduce cuda_cooperative/constraints.txt

* Also add cuda_parallel/constraints.txt

* Add `--constraint constraints.txt` in ci/test_python.sh

* Update Copyright dates

* Switch to https://github.com/ComPWA/taplo-pre-commit (the other repo has been archived by the owner on Jul 1, 2024)

For completeness: The other repo took a long time to install into the pre-commit cache; so long it lead to timeouts in the CCCL CI.

* Remove unused cuda_parallel jinja2 dependency (noticed by chance).

* Remove constraints.txt files, advertise running `pip install cuda-cccl` first instead.

* Make cuda_cooperative, cuda_parallel testing completely independent.

* Run only test_python.sh [skip-rapids][skip-matx][skip-docs][skip-vdc]

* Try using another runner (because V100 runners seem to be stuck) [skip-rapids][skip-matx][skip-docs][skip-vdc]

* Fix sign-compare warning (#3408) [skip-rapids][skip-matx][skip-docs][skip-vdc]

* Revert "Try using another runner (because V100 runners seem to be stuck) [skip-rapids][skip-matx][skip-docs][skip-vdc]"

This reverts commit ea33a218ed77a075156cd1b332047202adb25aa2.

Error message: https://github.com/NVIDIA/cccl/pull/3201#issuecomment-2594012971

* Try using A100 runner (because V100 runners still seem to be stuck) [skip-rapids][skip-matx][skip-docs][skip-vdc]

* Also show cuda-cooperative site-packages, cuda-parallel site-packages (after pip install) [skip-rapids][skip-matx][skip-docs][skip-vdc]

* Try using l4 runner (because V100 runners still seem to be stuck) [skip-rapids][skip-matx][skip-docs][skip-vdc]

* Restore original ci/matrix.yaml [skip-rapids]

* Use for loop in test_python.sh to avoid code duplication.

* Run only test_python.sh [skip-rapids][skip-matx][skip-docs][skip-vdc][skip pre-commit.ci]

* Comment out taplo-lint in pre-commit config [skip-rapids][skip-matx][skip-docs][skip-vdc]

* Revert "Run only test_python.sh [skip-rapids][skip-matx][skip-docs][skip-vdc][skip pre-commit.ci]"

This reverts commit ec206fd8b50a6a293e00a5825b579e125010b13d.

* Implement suggestion by @shwina (https://github.com/NVIDIA/cccl/pull/3201#pullrequestreview-2556918460)

* Address feedback by @leofang

---------

Co-authored-by: Bernhard Manfred Gruber <[email protected]>

cuda.parallel: Add optional stream argument to reduce_into() (#3348)

* Add optional stream argument to reduce_into()

* Add tests to check for reduce_into() stream behavior

* Move protocol related utils to separate file and rework __cuda_stream__ error messages

* Fix synchronization issue in stream test and add one more invalid stream test case

* Rename cuda stream validation function after removing leading underscore

* Unpack values from __cuda_stream__ instead of indexing

* Fix linting errors

* Handle TypeError when unpacking invalid __cuda_stream__ return

* Use stream to allocate cupy memory in new stream test

Upgrade to actions/deploy-pages@v4 (from v2), as suggested by @leofang (#3434)

Deprecate `cub::{min, max}` and replace internal uses with those from libcu++ (#3419)

* Deprecate `cub::{min, max}` and replace internal uses with those from libcu++

Fixes #3404

move to c++17, finalize device optimization

fix msvc compilation, update tests

Deprectate C++11 and C++14 for libcu++ (#3173)

* Deprectate C++11 and C++14 for libcu++

Co-authored-by: Bernhard Manfred Gruber <[email protected]>

Implement `abs` and `div` from `cstdlib` (#3153)

* implement integer abs functions
* improve tests, fix constexpr support
* just use the our implementation
* implement `cuda::std::div`
* prefer host's `div_t` like types
* provide `cuda::std::abs` overloads for floats
* allow fp abs for NVRTC
* silence msvc's warning about conversion from floating point to integral

Fix missing radix sort policies (#3174)

Fixes NVBug 5009941

Introduces new `DeviceReduce::Arg{Min,Max}` interface with two output iterators (#3148)

* introduces new arg{min,max} interface with two output iterators

* adds fp inf tests

* fixes docs

* improves code example

* fixes exec space specifier

* trying to fix deprecation warning for more compilers

* inlines unzip operator

* trying to fix deprecation warning for nvhpc

* integrates supression fixes in diagnostics

* pre-ctk 11.5 deprecation suppression

* fixes icc

* fix for pre-ctk11.5

* cleans up deprecation suppression

* cleanup

Extend tuning documentation (#3179)

Add codespell pre-commit hook, fix typos in CCCL (#3168)

* Add codespell pre-commit hook
* Automatic changes from codespell.
* Manual changes.

Fix parameter space for TUNE_LOAD in scan benchmark (#3176)

fix various old compiler checks (#3178)

implement C++26 `std::projected` (#3175)

Fix pre-commit config for codespell and remaining typos (#3182)

Massive cleanup of our config (#3155)

Fix UB in atomics with automatic storage (#2586)

* Adds specialized local cuda atomics and injects them into most atomics paths.

Co-authored-by: Georgy Evtushenko <[email protected]>
Co-authored-by: gonzalobg <[email protected]>

* Allow CUDA 12.2 to keep perf, this addresses earlier comments in #478

* Remove extraneous double brackets in unformatted code.

* Merge unsafe atomic logic into `__cuda_is_local`.

* Use `const_cast` for type conversions in cuda_local.h

* Fix build issues from interface changes

* Fix missing __nanosleep on sm70-

* Guard __isLocal from NVHPC

* Use PTX instead of running nothing from NVHPC

* fixup /s/nvrtc/nvhpc

* Fixup missing CUDA ifdef surrounding device code

* Fix codegen

* Bypass some sort of compiler bug on GCC7

* Apply suggestions from code review

* Use unsafe automatic storage atomics in codegen tests

---------

Co-authored-by: Georgy Evtushenko <[email protected]>
Co-authored-by: gonzalobg <[email protected]>
Co-authored-by: Michael Schellenberger Costa <[email protected]>

Refactor the source code layout for `cuda.parallel` (#3177)

* Refactor the source layout for cuda.parallel

* Add copyright

* Address review feedback

* Don't import anything into `experimental` namespace

* fix import

---------

Co-authored-by: Ashwin Srinath <[email protected]>

new type-erased memory resources (#2824)

s/_LIBCUDACXX_DECLSPEC_EMPTY_BASES/_CCCL_DECLSPEC_EMPTY_BASES/g (#3186)

Document address stability of `thrust::transform` (#3181)

* Do not document _LIBCUDACXX_MARK_CAN_COPY_ARGUMENTS
* Reformat and fix UnaryFunction/BinaryFunction in transform docs
* Mention transform can use proclaim_copyable_arguments
* Document cuda::proclaims_copyable_arguments better
* Deprecate depending on transform functor argument addresses

Fixes: #3053

turn off cuda version check for clangd (#3194)

[STF] jacobi example based on parallel_for (#3187)

* Simple jacobi example with parallel for and reductions

* clang-format

* remove useless capture list

fixes pre-nv_diag suppression issues (#3189)

Prefer c2h::type_name over c2h::demangle (#3195)

Fix memcpy_async* tests (#3197)

* memcpy_async_tx: Fix bug in test

Two bugs, one of which occurs in practice:

1. There is a missing fence.proxy.space::global between the writes to
   global memory and the memcpy_async_tx. (Occurs in practice)

2. The end of the kernel should be fenced with `__syncthreads()`,
   because the barrier is invalidated in the destructor. If other
   threads are still waiting on it, there will be UB. (Has not yet
   manifested itself)

* cp_async_bulk_tensor: Pre-emptively fence more in test

Add type annotations and mypy checks for `cuda.parallel`  (#3180)

* Refactor the source layout for cuda.parallel

* Add initial type annotations

* Update pre-commit config

* More typing

* Fix bad merge

* Fix TYPE_CHECKING and numpy annotations

* typing bindings.py correctly

* Address review feedback

---------

Co-authored-by: Ashwin Srinath <[email protected]>

Fix rendering of cuda.parallel docs (#3192)

* Fix pre-commit config for codespell and remaining typos

* Fix rendering of docs for cuda.parallel

---------

Co-authored-by: Ashwin Srinath <[email protected]>

Enable PDL for DeviceMergeSortBlockSortKernel (#3199)

The kernel already contains a call to _CCCL_PDL_GRID_DEPENDENCY_SYNC.
This commit enables PDL when launching the kernel.

Adds support for large `num_items` to `DeviceReduce::{ArgMin,ArgMax}` (#2647)

* adds benchmarks for reduce::arg{min,max}

* preliminary streaming arg-extremum reduction

* fixes implicit conversion

* uses streaming dispatch class

* changes arg benches to use new streaming reduce

* streaming arg-extrema reduction

* fixes style

* fixes compilation failures

* cleanups

* adds rst style comments

* declare vars const and use clamp

* consolidates argmin argmax benchmarks

* fixes thrust usage

* drops offset type in arg-extrema benchmarks

* fixes clang cuda

* exec space macros

* switch to signed global offset type for slightly better perf

* clarifies documentation

* applies minor benchmark style changes from review comments

* fixes interface documentation and comments

* list-init accumulating output op

* improves style, comments, and tests

* cleans up aggregate init

* renames dispatch class usage in benchmarks

* fixes merge conflicts

* addresses review comments

* addresses review comments

* fixes assertion

* removes superseded implementation

* changes large problem tests to use new interface

* removes obsolete tests for deprecated interface

Fixes for Python 3.7 docs environment (#3206)

Co-authored-by: Ashwin Srinath <[email protected]>

Adds support for large number of items to `DeviceTransform` (#3172)

* moves large problem test helper to common file

* adds support for large num items to device transform

* adds tests for large number of items to device interface

* fixes format

* addresses review comments

cp_async_bulk: Fix test (#3198)

* memcpy_async_tx: Fix bug in test

Two bugs, one of which occurs in practice:

1. There is a missing fence.proxy.space::global between the writes to
   global memory and the memcpy_async_tx. (Occurs in practice)

2. The end of the kernel should be fenced with `__syncthreads()`,
   because the barrier is invalidated in the destructor. If other
   threads are still waiting on it, there will be UB. (Has not yet
   manifested itself)

* cp_async_bulk_tensor: Pre-emptively fence more in test

* cp_async_bulk: Fix test

The global memory pointer could be misaligned.

cudax fixes for msvc 14.41 (#3200)

avoid instantiating class templates in `is_same` implementation when possible (#3203)

Fix: make launchers a CUB detail; make kernel source functions hidden. (#3209)

* Fix: make launchers a CUB detail; make kernel source functions hidden.

* [pre-commit.ci] auto code formatting

* Address review comments, fix which macro gets fixed.

help the ranges concepts recognize standard contiguous iterators in c++14/17 (#3202)

unify macros and cmake options that control the suppression of deprecation warnings (#3220)

* unify macros and cmake options that control the suppression of deprecation warnings

* suppress nvcc warning #186 in thrust header tests

* suppress c++ dialect deprecation warnings in libcudacxx header tests

Fx thread-reduce performance regression (#3225)

cuda.parallel: In-memory caching of build objects (#3216)

* Define __eq__ and __hash__ for Iterators

* Define cache_with_key utility and use it to cache Reduce objects

* Add tests for caching Reduce objects

* Tighten up types

* Updates to support 3.7

* Address review feedback

* Introduce IteratorKind to hold iterator type information

* Use the .kind to generate an abi_name

* Remove __eq__ and __hash__ methods from IteratorBase

* Move helper function

* Formatting

* Don't unpack tuple in cache key

---------

Co-authored-by: Ashwin Srinath <[email protected]>

Just enough ranges for c++14 `span` (#3211)

use generalized concepts portability macros to simplify the `range` concept (#3217)

fixes some issues in the concepts portability macros and then re-implements the `range` concept with `_CCCL_REQUIRES_EXPR`

Use Ruff to sort imports (#3230)

* Update pyproject.tomls for import sorting

* Update files after running pre-commit

* Move ruff config to pyproject.toml

---------

Co-authored-by: Ashwin Srinath <[email protected]>

fix tuning_scan sm90 config issue (#3236)

Co-authored-by: Shijie Chen <[email protected]>

[STF] Logical token (#3196)

* Split the implementation of the void interface into the definition of the interface, and its implementations on streams and graphs.

* Add missing files

* Check if a task implementation can match a prototype where the void_interface arguments are ignored

* Implement ctx.abstract_logical_data() which relies on a void data interface

* Illustrate how to use abstract handles in local contexts

* Introduce an is_void_interface() virtual method in the data interface to potentially optimize some stages

* Small improvements in the examples

* Do not try to allocate or move void data

* Do not use I as a variable

* fix linkage error

* rename abtract_logical_data into logical_token

* Document logical token

* fix spelling error

* fix sphinx error

* reflect name changes

* use meaningful variable names

* simplify logical_token implementation because writeback is already disabled

* add a unit test for token elision

* implement token elision in host_launch

* Remove unused type

* Implement helpers to check if a function can be invoked from a tuple, or from a tuple where we removed tokens

* Much simpler is_tuple_invocable_with_filtered implementation

* Fix buggy test

* Factorize code

* Document that we can ignore tokens for task and host_launch

* Documentation for logical data freeze

Fix ReduceByKey tuning (#3240)

Fix RLE tuning (#3239)

cuda.parallel: Forbid non-contiguous arrays as inputs (or outputs) (#3233)

* Forbid non-contiguous arrays as inputs (or outputs)

* Implement a more robust way to check for contiguity

* Don't bother if cublas unavailable

* Fix how we check for zero-element arrays

* sort imports

---------

Co-authored-by: Ashwin Srinath <[email protected]>

expands support for more offset types in segmented benchmark (#3231)

Add escape hatches to the cmake configuration of the header tests so that we can tests deprecated compilers / dialects (#3253)

* Add escape hatches to the cmake configuration of the header tests so that we can tests deprecated compilers / dialects

* Do not add option twice

ptx: Add add_instruction.py (#3190)

This file helps create the necessary structure for new PTX instructions.

Co-authored-by: Allard Hendriksen <[email protected]>

Bump main to 2.9.0. (#3247)

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

Drop cub::Mutex (#3251)

Fixes: #3250

Remove legacy macros from CUB util_arch.cuh (#3257)

Fixes: #3256

Remove thrust::[unary|binary]_traits (#3260)

Fixes: #3259

Architecture and OS identification macros (#3237)

Bump main to 3.0.0. (#3265)

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

Drop thrust not1 and not2 (#3264)

Fixes: #3263

CCCL Internal macro documentation (#3238)

Deprecate GridBarrier and GridBarrierLifetime (#3258)

Fixes: #1389

Require at least gcc7 (#3268)

Fixes: #3267

Drop thrust::[unary|binary]_function (#3274)

Fixes: #3273

Drop ICC from CI (#3277)

[STF] Corruption of the capture list of an extended lambda with a parallel_for construct on a host execution place (#3270)

* Add a test to reproduce a bug observed with parallel_for on a host place

* clang-format

* use _CCCL_ASSERT

* Attempt to debug

* do not create a tuple with a universal reference that is out of scope when we use it, use an lvalue instead

* fix lambda expression

* clang-format

Enable thrust::identity test for non-MSVC (#3281)

This seems to be an oversight when the test was added

Co-authored-by: Michael Schellenberger Costa <[email protected]>

Enable PDL in triple chevron launch (#3282)

It seems PDL was disabled by accident when _THRUST_HAS_PDL was renamed
to _CCCL_HAS_PDL during the review introducing the feature.

Disambiguate line continuations and macro continuations in <nv/target> (#3244)

Drop VS 2017 from CI (#3287)

Fixes: #3286

Drop ICC support in code (#3279)

* Drop ICC from code

Fixes: #3278

Co-authored-by: Michael Schellenberger Costa <[email protected]>

Make CUB NVRTC commandline arguments come from a cmake template (#3292)

Propose the same components (thrust, cub, libc++, cudax, cuda.parallel,...) in the bug report template than in the feature request template (#3295)

Use process isolation instead of default hyper-v for Windows. (#3294)

Try improving build times by using process isolation instead of hyper-v

Co-authored-by: Michael Schellenberger Costa <[email protected]>

[pre-commit.ci] pre-commit autoupdate (#3248)

* [pre-commit.ci] pre-commit autoupdate

updates:
- [github.com/pre-commit/mirrors-clang-format: v18.1.8 → v19.1.6](https://github.com/pre-commit/mirrors-clang-format/compare/v18.1.8...v19.1.6)
- [github.com/astral-sh/ruff-pre-commit: v0.8.3 → v0.8.6](https://github.com/astral-sh/ruff-pre-commit/compare/v0.8.3...v0.8.6)
- [github.com/pre-commit/mirrors-mypy: v1.13.0 → v1.14.1](https://github.com/pre-commit/mirrors-mypy/compare/v1.13.0...v1.14.1)

Co-authored-by: Michael Schellenberger Costa <[email protected]>

Drop Thrust legacy arch macros (#3298)

Which were disabled and could be re-enabled using THRUST_PROVIDE_LEGACY_ARCH_MACROS

Drop Thrust's compiler_fence.h (#3300)

Drop CTK 11.x from CI (#3275)

* Add cuda12.0-gcc7 devcontainer
* Move MSVC2017 jobs to CTK 12.6
Those is the only combination where rapidsai has devcontainers
* Add /Zc:__cplusplus for the libcudacxx tests
* Only add excape hatch for affected CTKs
* Workaround missing cudaLaunchKernelEx on MSVC
cudaLaunchKernelEx requires C++11, but unfortunately <cuda_runtime.h> checks this using the __cplusplus macro, which is reported wrongly for MSVC. CTK 12.3 fixed this by additionally detecting _MSV_VER. As a workaround, we provide our own copy of cudaLaunchKernelEx when it is not available from the CTK.
* Workaround nvcc+MSVC issue
* Regenerate devcontainers

Fixes: #3249

Co-authored-by: Michael Schellenberger Costa <[email protected]>

Update packman and repo_docs versions (#3293)

Co-authored-by: Ashwin Srinath <[email protected]>

Drop Thrust's deprecated compiler macros (#3301)

Drop CUB_RUNTIME_ENABLED and __THRUST_HAS_CUDART__ (#3305)

Adds support for large number of items to `DevicePartition::If` with the `ThreeWayPartition` overload (#2506)

* adds support for large number of items to three-way partition

* adapts interface to use choose_signed_offset_t

* integrates applicable feedback from device-select pr

* changes behavior for empty problems

* unifies grid constant macro

* fixes kernel template specialization mismatch

* integrates _CCCL_GRID_CONSTANT changes

* resolve merge conflicts

* fixes checks in test

* fixes test verification

* improves tests

* makes few improvements to streaming dispatch

* improves code comment on test

* fixes unrelated compiler error

* minor style improvements

Refactor scan tunings (#3262)

Require C++17 for compiling Thrust and CUB (#3255)

* Issue an unsuppressable warning when compiling with < C++17
* Remove C++11/14 presets
* Remove CCCL_IGNORE_DEPRECATED_CPP_DIALECT from headers
* Remove [CUB|THRUST|TCT]_IGNORE_DEPRECATED_CPP_[11|14]
* Remove CUB_ENABLE_DIALECT_CPP[11|14]
* Update CI runs
* Remove C++11/14 CI runs for CUB and Thrust
* Raise compiler minimum versions for C++17
* Update ReadMe
* Drop Thrust's cpp14_required.h
* Add escape hatch for C++17 removal

Fixes: #3252

Implement `views::empty` (#3254)

* Disable pair conversion of subrange with clang in C++17

* Fix namespace views

* Implement `views::empty`

This implements `std::ranges::views::empty`, see https://en.cppreference.com/w/cpp/ranges/empty_view

Refactor `limits` and `climits` (#3221)

* implement builtins for huge val, nan and nans

* change `INFINITY` and `NAN` implementation for NVRTC

cuda.parallel: Add documentation for the current iterators along with examples and tests (#3311)

* Add tests demonstrating usage of different iterators

* Update documentation of reduce_into by merging import code snippet with the rest of the example

* Add documentation for current iterators

* Run pre-commit checks and update accordingly

* Fix comments to refer to the proper lines in the code snippets in the docs

Drop clang<14 from CI, update devcontainers. (#3309)

Co-authored-by: Bernhard Manfred Gruber <[email protected]>

[STF] Cleanup task dependencies object constructors (#3291)

* Define tag types for access modes

* - Rework how we build task_dep objects based on access mode tags
- pack_state is now responsible for using a const_cast for read only data

* Greatly simplify the previous attempt : do not define new types, but use integral constants based on the enums

* It seems the const_cast was not necessarily so we can simplify it and not even do some dispatch based on access modes

Disable test with a gcc-14 regression (#3297)

Deprecate Thrust's cpp_compatibility.h macros (#3299)

Remove dropped function objects from docs (#3319)

Document `NV_TARGET` macros (#3313)

[STF] Define ctx.pick_stream() which was missing for the unified context (#3326)

* Define ctx.pick_stream() which was missing for the unified context

* clang-format

Deprecate cub::IterateThreadStore (#3337)

Drop CUB's BinaryFlip operator (#3332)

Deprecate cub::Swap (#3333)

Clarify transform output can overlap input (#3323)

Drop CUB APIs with a debug_synchronous parameter (#3330)

Fixes: #3329

Drop CUB's util_compiler.cuh for real (#3340)

PR #3302 planned to drop the file, but only dropped its content. This
was an oversight. So let's drop the entire file.

Drop cub::ValueCache (#3346)

limits offset types for merge sort (#3328)

Drop CDPv1 (#3344)

Fixes: #3341

Drop thrust::void_t (#3362)

Use cuda::std::addressof in Thrust (#3363)

Fix all_of documentation for empty ranges (#3358)

all_of always returns true on an empty range.

[STF] Do not keep track of dangling events in a CUDA graph backend (#3327)

* Unlike the CUDA stream backend, nodes in a CUDA graph are necessarily done when
the CUDA graph completes. Therefore keeping track of "dangling events" is a
waste of time and resources.

* replace can_ignore_dangling_events by track_dangling_events which leads to more readable code

* When not storing the dangling events, we must still perform the deinit operations that were producing these events !

Extract scan kernels into NVRTC-compilable header (#3334)

* Extract scan kernels into NVRTC-compilable header

* Update cub/cub/device/dispatch/dispatch_scan.cuh

Co-authored-by: Georgii Evtushenko <[email protected]>

---------

Co-authored-by: Ashwin Srinath <[email protected]>
Co-authored-by: Georgii Evtushenko <[email protected]>

Drop deprecated aliases in Thrust functional (#3272)

Fixes: #3271

Drop cub::DivideAndRoundUp (#3347)

Use cuda::std::min/max in Thrust (#3364)

Implement `cuda::std::numeric_limits` for `__half` and `__nv_bfloat16` (#3361)

* implement `cuda::std::numeric_limits` for `__half` and `__nv_bfloat16`

Cleanup util_arch (#2773)

Deprecate thrust::null_type (#3367)

Deprecate cub::DeviceSpmv (#3320)

Fixes: #896

Improves `DeviceSegmentedSort` test run time for large number of items and segments (#3246)

* fixes segment offset generation

* switches to analytical verification

* switches to analytical verification for pairs

* fixes spelling

* adds tests for large number of segments

* fixes narrowing conversion in tests

* addresses review comments

* fixes includes

Compile basic infra test with C++17 (#3377)

Adds support for large number of items and large number of segments to `DeviceSegmentedSort` (#3308)

* fixes segment offset generation

* switches to analytical verification

* switches to analytical verification for pairs

* addresses review comments

* introduces segment offset type

* adds tests for large number of segments

* adds support for large number of segments

* drops segment offset type

* fixes thrust namespace

* removes about-to-be-deprecated cub iterators

* no exec specifier on defaulted ctor

* fixes gcc7 linker error

* uses local_segment_index_t throughout

* determine offset type based on type returned by segment iterator begin/end iterators

* minor style improvements

Exit with error when RAPIDS CI fails. (#3385)

cuda.parallel: Support structured types as algorithm inputs (#3218)

* Introduce gpu_struct decorator and typing

* Enable `reduce` to accept arrays of structs as inputs

* Add test for reducing arrays-of-struct

* Update documentation

* Use a numpy array rather than ctypes object

* Change zeros -> empty for output array and temp storage

* Add a TODO for typing GpuStruct

* Documentation udpates

* Remove test_reduce_struct_type from test_reduce.py

* Revert to `to_cccl_value()` accepting ndarray + GpuStruct

* Bump copyrights

---------

Co-authored-by: Ashwin Srinath <[email protected]>

Deprecate thrust::async (#3324)

Fixes: #100

Review/Deprecate CUB `util.ptx` for CCCL 2.x (#3342)

Fix broken `_CCCL_BUILTIN_ASSUME` macro (#3314)

* add compiler-specific path
* fix device code path
* add _CCC_ASSUME

Deprecate thrust::numeric_limits (#3366)

Replace `typedef` with `using` in libcu++ (#3368)

Deprecate thrust::optional (#3307)

Fixes: #3306

Upgrade to Catch2 3.8  (#3310)

Fixes: #1724

refactor `<cuda/std/cstdint>` (#3325)

Co-authored-by: Bernhard Manfred Gruber <[email protected]>

Update CODEOWNERS (#3331)

* Update CODEOWNERS

* Update CODEOWNERS

* Update CODEOWNERS

* [pre-commit.ci] auto code formatting

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

Fix sign-compare warning (#3408)

Implement more cmath functions to be usable on host and device (#3382)

* Implement more cmath functions to be usable on host and device

* Implement math roots functions

* Implement exponential functions

Redefine and deprecate thrust::remove_cvref (#3394)

* Redefine and deprecate thrust::remove_cvref

Co-authored-by: Michael Schellenberger Costa <[email protected]>

Fix assert definition for NVHPC due to constexpr issues (#3418)

NVHPC cannot decide at compile time where the code would run so _CCCL_ASSERT within a constexpr function breaks it.

Fix this by always using the host definition which should also work on device.

Fixes #3411

Extend CUB reduce benchmarks (#3401)

* Rename max.cu to custom.cu, since it uses a custom operator
* Extend types covered my min.cu to all fundamental types
* Add some notes on how to collect tuning parameters

Fixes: #3283

Update upload-pages-artifact to v3 (#3423)

* Update upload-pages-artifact to v3

* Empty commit

---------

Co-authored-by: Ashwin Srinath <[email protected]>

Replace and deprecate thrust::cuda_cub::terminate (#3421)

`std::linalg` accessors and `transposed_layout` (#2962)

Add round up/down to multiple (#3234)

[FEA]: Introduce Python module with CCCL headers (#3201)

* Add cccl/python/cuda_cccl directory and use from cuda_parallel, cuda_cooperative

* Run `copy_cccl_headers_to_aude_include()` before `setup()`

* Create python/cuda_cccl/cuda/_include/__init__.py, then simply import cuda._include to find the include path.

* Add cuda.cccl._version exactly as for cuda.cooperative and cuda.parallel

* Bug fix: cuda/_include only exists after shutil.copytree() ran.

* Use `f"cuda-cccl @ file://{cccl_path}/python/cuda_cccl"` in setup.py

* Remove CustomBuildCommand, CustomWheelBuild in cuda_parallel/setup.py (they are equivalent to the default functions)

* Replace := operator (needs Python 3.8+)

* Fix oversights: remove `pip3 install ./cuda_cccl` lines from README.md

* Restore original README.md: `pip3 install -e` now works on first pass.

* cuda_cccl/README.md: FOR INTERNAL USE ONLY

* Remove `$pymajor.$pyminor.` prefix in cuda_cccl _version.py (as suggested under https://github.com/NVIDIA/cccl/pull/3201#discussion_r1894035917)

Command used: ci/update_version.sh 2 8 0

* Modernize pyproject.toml, setup.py

Trigger for this change:

* https://github.com/NVIDIA/cccl/pull/3201#discussion_r1894043178

* https://github.com/NVIDIA/cccl/pull/3201#discussion_r1894044996

* Install CCCL headers under cuda.cccl.include

Trigger for this change:

* https://github.com/NVIDIA/cccl/pull/3201#discussion_r1894048562

Unexpected accidental discovery: cuda.cooperative unit tests pass without CCCL headers entirely.

* Factor out cuda_cccl/cuda/cccl/include_paths.py

* Reuse cuda_cccl/cuda/cccl/include_paths.py from cuda_cooperative

* Add missing Copyright notice.

* Add missing __init__.py (cuda.cccl)

* Add `"cuda.cccl"` to `autodoc.mock_imports`

* Move cuda.cccl.include_paths into function where it is used. (Attempt to resolve Build and Verify Docs failure.)

* Add # TODO: move this to a module-level import

* Modernize cuda_cooperative/pyproject.toml, setup.py

* Convert cuda_cooperative to use hatchling as build backend.

* Revert "Convert cuda_cooperative to use hatchling as build backend."

This reverts commit 61637d608da06fcf6851ef6197f88b5e7dbc3bbe.

* Move numpy from [build-system] requires -> [project] dependencies

* Move pyproject.toml [project] dependencies -> setup.py install_requires, to be able to use CCCL_PATH

* Remove copy_license() and use license_files=["../../LICENSE"] instead.

* Further modernize cuda_cccl/setup.py to use pathlib

* Trivial simplifications in cuda_cccl/pyproject.toml

* Further simplify cuda_cccl/pyproject.toml, setup.py: remove inconsequential code

* Make cuda_cooperative/pyproject.toml more similar to cuda_cccl/pyproject.toml

* Add taplo-pre-commit to .pre-commit-config.yaml

* taplo-pre-commit auto-fixes

* Use pathlib in cuda_cooperative/setup.py

* CCCL_PYTHON_PATH in cuda_cooperative/setup.py

* Modernize cuda_parallel/pyproject.toml, setup.py

* Use pathlib in cuda_parallel/setup.py

* Add `# TOML lint & format` comment.

* Replace MANIFEST.in with `[tool.setuptools.package-data]` section in pyproject.toml

* Use pathlib in cuda/cccl/include_paths.py

* pre-commit autoupdate (EXCEPT clang-format, which was manually restored)

* Fixes after git merge main

* Resolve warning: AttributeError: '_Reduce' object has no attribute 'build_result'

```
=========================================================================== warnings summary ===========================================================================
tests/test_reduce.py::test_reduce_non_contiguous
  /home/coder/cccl/python/devenv/lib/python3.12/site-packages/_pytest/unraisableexception.py:85: PytestUnraisableExceptionWarning: Exception ignored in: <function _Reduce.__del__ at 0x7bf123139080>

  Traceback (most recent call last):
    File "/home/coder/cccl/python/cuda_parallel/cuda/parallel/experimental/algorithms/reduce.py", line 132, in __del__
      bindings.cccl_device_reduce_cleanup(ctypes.byref(self.build_result))
                                                       ^^^^^^^^^^^^^^^^^
  AttributeError: '_Reduce' object has no attribute 'build_result'

    warnings.warn(pytest.PytestUnraisableExceptionWarning(msg))

-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
============================================================= 1 passed, 93 deselected, 1 warning in 0.44s ==============================================================
```

* Move `copy_cccl_headers_to_cuda_cccl_include()` functionality to `class CustomBuildPy`

* Introduce cuda_cooperative/constraints.txt

* Also add cuda_parallel/constraints.txt

* Add `--constraint constraints.txt` in ci/test_python.sh

* Update Copyright dates

* Switch to https://github.com/ComPWA/taplo-pre-commit (the other repo has been archived by the owner on Jul 1, 2024)

For completeness: The other repo took a long time to install into the pre-commit cache; so long it lead to timeouts in the CCCL CI.

* Remove unused cuda_parallel jinja2 dependency (noticed by chance).

* Remove constraints.txt files, advertise running `pip install cuda-cccl` first instead.

* Make cuda_cooperative, cuda_parallel testing completely independent.

* Run only test_python.sh [skip-rapids][skip-matx][skip-docs][skip-vdc]

* Try using another runner (because V100 runners seem to be stuck) [skip-rapids][skip-matx][skip-docs][skip-vdc]

* Fix sign-compare warning (#3408) [skip-rapids][skip-matx][skip-docs][skip-vdc]

* Revert "Try using another runner (because V100 runners seem to be stuck) [skip-rapids][skip-matx][skip-docs][skip-vdc]"

This reverts commit ea33a218ed77a075156cd1b332047202adb25aa2.

Error message: https://github.com/NVIDIA/cccl/pull/3201#issuecomment-2594012971

* Try using A100 runner (because V100 runners still seem to be stuck) [skip-rapids][skip-matx][skip-docs][skip-vdc]

* Also show cuda-cooperative site-packages, cuda-parallel site-packages (after pip install) [skip-rapids][skip-matx][skip-docs][skip-vdc]

* Try using l4 runner (because V100 runners still seem to be stuck) [skip-rapids][skip-matx][skip-docs][skip-vdc]

* Restore original ci/matrix.yaml [skip-rapids]

* Use for loop in test_python.sh to avoid code duplication.

* Run only test_python.sh [skip-rapids][skip-matx][skip-docs][skip-vdc][skip pre-commit.ci]

* Comment out taplo-lint in pre-commit config [skip-rapids][skip-matx][skip-docs][skip-vdc]

* Revert "Run only test_python.sh [skip-rapids][skip-matx][skip-docs][skip-vdc][skip pre-commit.ci]"

This reverts commit ec206fd8b50a6a293e00a5825b579e125010b13d.

* Implement suggestion by @shwina (https://github.com/NVIDIA/cccl/pull/3201#pullrequestreview-2556918460)

* Address feedback by @leofang

---------

Co-authored-by: Bernhard Manfred Gruber <[email protected]>

cuda.parallel: Add optional stream argument to reduce_into() (#3348)

* Add optional stream argument to reduce_into()

* Add tests to check for reduce_into() stream behavior

* Move protocol related utils to separate file and rework __cuda_stream__ error messages

* Fix synchronization issue in stream test and add one more invalid stream test case

* Rename cuda stream validation function after removing leading underscore

* Unpack values from __cuda_stream__ instead of indexing

* Fix linting errors

* Handle TypeError when unpacking invalid __cuda_stream__ return

* Use stream to allocate cupy memory in new stream test

Upgrade to actions/deploy-pages@v4 (from v2), as suggested by @leofang (#3434)

Deprecate `cub::{min, max}` and replace internal uses with those from libcu++ (#3419)

* Deprecate `cub::{min, max}` and replace internal uses with those from libcu++

Fixes #3404

Fix CI issues (#3443)

update docs

fix review

restrict allowed types

replace constexpr implementations with generic

optimize `__is_arithmetic_integral`
@Jorgeperezr
Copy link

It seems like a good solution for detecting contiguous iterators in older versions of C++🙂. Have you considered any cases where compiler optimizations might affect this detection?

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

@CaseyCarter CaseyCarter CaseyCarter left review comments

@miscco miscco miscco approved these changes

@wmaxey wmaxey Awaiting requested review from wmaxey wmaxey is a code owner automatically assigned from NVIDIA/cccl-libcudacxx-codeowners

@fbusato fbusato Awaiting requested review from fbusato fbusato was automatically assigned from NVIDIA/cccl-codeowners

Assignees
No one assigned
Labels
None yet
Projects
CCCL
Archived in project
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
4 participants
@ericniebler @Jorgeperezr @CaseyCarter @miscco