Replace `cub::Traits` by numeric_limits and deprecate it #3384

bernhardmgruber · 2025-01-14T14:57:35Z

copy-pr-bot · 2025-01-14T14:57:39Z

Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually.

Contributors can view more details about this message here.

bernhardmgruber · 2025-01-14T14:57:46Z

/ok to test

bernhardmgruber · 2025-01-14T15:00:56Z

@miscco I would love to deprecate cub::Traits in favor of standard facilities in libcu++. As it currently stands, we would still need:

support for FP16, BF16 and FP8 types by cuda::std::is_floating_point
support for FP16, BF16 and FP8 types by cuda::std::numeric_limits (only min and lowest)

Do you think it's possible we can have this support soonish?

copy-pr-bot · 2025-01-22T15:50:19Z

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.

bernhardmgruber · 2025-01-22T15:51:06Z

/ok to test

bernhardmgruber · 2025-01-22T16:40:59Z

/ok to test

c2h/include/c2h/half.cuh

github-actions · 2025-01-23T04:53:13Z

🟨 CI finished in 4h 49m: Pass: 91%/78 | Total: 2d 06h | Avg: 41m 37s | Max: 1h 14m | Hits: 183%/11826

🟨 cub: Pass: 81%/38 | Total: 1d 08h | Avg: 51m 44s | Max: 1h 14m | Hits: 81%/2646

🔍 cpu: amd64 🔍
  🔍 amd64              Pass:  80%/36  | Total:  1d 06h | Avg: 50m 51s | Max:  1h 14m | Hits:  81%/2646  
  🟩 arm64              Pass: 100%/2   | Total:  2h 15m | Avg:  1h 07m | Max:  1h 09m
🔍 cudacxx_family: nvcc 🔍
  🟩 ClangCUDA          Pass: 100%/2   | Total:  2h 00m | Avg:  1h 00m | Max:  1h 01m
  🔍 nvcc               Pass:  80%/36  | Total:  1d 06h | Avg: 51m 16s | Max:  1h 14m | Hits:  81%/2646  
🔍 gpu: v100 🔍
  🟩 h100               Pass: 100%/2   | Total: 45m 20s | Avg: 22m 40s | Max: 25m 54s
  🔍 v100               Pass:  80%/36  | Total:  1d 08h | Avg: 53m 21s | Max:  1h 14m | Hits:  81%/2646  
🟨 ctk
  🟥 12.0               Pass:   0%/5   | Total:  3h 18m | Avg: 39m 43s | Max:  1h 01m
  🟩 12.5               Pass: 100%/2   | Total:  2h 27m | Avg:  1h 13m | Max:  1h 14m
  🟨 12.6               Pass:  93%/31  | Total:  1d 03h | Avg: 52m 15s | Max:  1h 11m | Hits:  81%/2646  
🟨 cudacxx
  🟩 ClangCUDA18        Pass: 100%/2   | Total:  2h 00m | Avg:  1h 00m | Max:  1h 01m
  🟥 nvcc12.0           Pass:   0%/5   | Total:  3h 18m | Avg: 39m 43s | Max:  1h 01m
  🟩 nvcc12.5           Pass: 100%/2   | Total:  2h 27m | Avg:  1h 13m | Max:  1h 14m
  🟨 nvcc12.6           Pass:  93%/29  | Total:  1d 00h | Avg: 51m 42s | Max:  1h 11m | Hits:  81%/2646  
🟨 cxx
  🟨 Clang14            Pass:  50%/4   | Total:  3h 10m | Avg: 47m 32s | Max:  1h 02m
  🟩 Clang15            Pass: 100%/1   | Total:  1h 02m | Avg:  1h 02m | Max:  1h 02m
  🟩 Clang16            Pass: 100%/1   | Total:  1h 01m | Avg:  1h 01m | Max:  1h 01m
  🟩 Clang17            Pass: 100%/1   | Total: 59m 06s | Avg: 59m 06s | Max: 59m 06s
  🟨 Clang18            Pass:  85%/7   | Total:  6h 20m | Avg: 54m 19s | Max:  1h 09m
  🟨 GCC7               Pass:  50%/2   | Total:  1h 35m | Avg: 47m 52s | Max:  1h 02m
  🟩 GCC8               Pass: 100%/1   | Total:  1h 02m | Avg:  1h 02m | Max:  1h 02m
  🟨 GCC9               Pass:  50%/2   | Total:  1h 35m | Avg: 47m 35s | Max:  1h 01m
  🟩 GCC10              Pass: 100%/1   | Total:  1h 03m | Avg:  1h 03m | Max:  1h 03m
  🟩 GCC11              Pass: 100%/1   | Total: 56m 34s | Avg: 56m 34s | Max: 56m 34s
  🟩 GCC12              Pass: 100%/3   | Total:  1h 47m | Avg: 35m 52s | Max:  1h 02m
  🟨 GCC13              Pass:  87%/8   | Total:  5h 12m | Avg: 39m 06s | Max:  1h 06m
  🟨 MSVC14.29          Pass:  50%/2   | Total:  2h 11m | Avg:  1h 05m | Max:  1h 10m | Hits:  84%/882   
  🟩 MSVC14.39          Pass: 100%/2   | Total:  2h 19m | Avg:  1h 09m | Max:  1h 11m | Hits:  80%/1764  
  🟩 NVHPC24.7          Pass: 100%/2   | Total:  2h 27m | Avg:  1h 13m | Max:  1h 14m
🟨 cxx_family
  🟨 Clang              Pass:  78%/14  | Total: 12h 33m | Avg: 53m 50s | Max:  1h 09m
  🟨 GCC                Pass:  83%/18  | Total: 13h 14m | Avg: 44m 07s | Max:  1h 06m
  🟨 MSVC               Pass:  75%/4   | Total:  4h 30m | Avg:  1h 07m | Max:  1h 11m | Hits:  81%/2646  
  🟩 NVHPC              Pass: 100%/2   | Total:  2h 27m | Avg:  1h 13m | Max:  1h 14m
🟨 jobs
  🟨 Build              Pass:  83%/31  | Total:  1d 05h | Avg: 57m 03s | Max:  1h 14m | Hits:  81%/2646  
  🟩 DeviceLaunch       Pass: 100%/1   | Total: 29m 04s | Avg: 29m 04s | Max: 29m 04s
  🟩 GraphCapture       Pass: 100%/1   | Total: 17m 49s | Avg: 17m 49s | Max: 17m 49s
  🟩 HostLaunch         Pass: 100%/3   | Total:  1h 18m | Avg: 26m 15s | Max: 30m 00s
  🟥 TestGPU            Pass:   0%/2   | Total:  1h 12m | Avg: 36m 02s | Max: 45m 58s
🟩 sm
  🟩 90                 Pass: 100%/2   | Total: 45m 20s | Avg: 22m 40s | Max: 25m 54s
  🟩 90a                Pass: 100%/1   | Total: 26m 26s | Avg: 26m 26s | Max: 26m 26s
🟨 std
  🟨 17                 Pass:  71%/14  | Total: 13h 18m | Avg: 57m 02s | Max:  1h 13m | Hits:  84%/1764  
  🟨 20                 Pass:  87%/24  | Total: 19h 27m | Avg: 48m 39s | Max:  1h 14m | Hits:  77%/882

🟩 thrust: Pass: 100%/37 | Total: 20h 24m | Avg: 33m 05s | Max: 1h 03m | Hits: 212%/9180

🟩 cmake_options
  🟩 -DTHRUST_DISPATCH_TYPE=Force32bit Pass: 100%/2   | Total: 41m 31s | Avg: 20m 45s | Max: 27m 07s
🟩 cpu
  🟩 amd64              Pass: 100%/35  | Total: 19h 24m | Avg: 33m 16s | Max:  1h 03m | Hits: 212%/9180  
  🟩 arm64              Pass: 100%/2   | Total: 59m 22s | Avg: 29m 41s | Max: 31m 07s
🟩 ctk
  🟩 12.0               Pass: 100%/5   | Total:  3h 03m | Avg: 36m 46s | Max: 53m 41s | Hits: 173%/1836  
  🟩 12.5               Pass: 100%/2   | Total:  1h 56m | Avg: 58m 23s | Max: 59m 22s
  🟩 12.6               Pass: 100%/30  | Total: 15h 23m | Avg: 30m 47s | Max:  1h 03m | Hits: 221%/7344  
🟩 cudacxx
  🟩 ClangCUDA18        Pass: 100%/2   | Total: 52m 56s | Avg: 26m 28s | Max: 26m 57s
  🟩 nvcc12.0           Pass: 100%/5   | Total:  3h 03m | Avg: 36m 46s | Max: 53m 41s | Hits: 173%/1836  
  🟩 nvcc12.5           Pass: 100%/2   | Total:  1h 56m | Avg: 58m 23s | Max: 59m 22s
  🟩 nvcc12.6           Pass: 100%/28  | Total: 14h 30m | Avg: 31m 05s | Max:  1h 03m | Hits: 221%/7344  
🟩 cudacxx_family
  🟩 ClangCUDA          Pass: 100%/2   | Total: 52m 56s | Avg: 26m 28s | Max: 26m 57s
  🟩 nvcc               Pass: 100%/35  | Total: 19h 31m | Avg: 33m 28s | Max:  1h 03m | Hits: 212%/9180  
🟩 cxx
  🟩 Clang14            Pass: 100%/4   | Total:  2h 07m | Avg: 31m 55s | Max: 33m 27s
  🟩 Clang15            Pass: 100%/1   | Total: 32m 16s | Avg: 32m 16s | Max: 32m 16s
  🟩 Clang16            Pass: 100%/1   | Total: 31m 34s | Avg: 31m 34s | Max: 31m 34s
  🟩 Clang17            Pass: 100%/1   | Total: 29m 57s | Avg: 29m 57s | Max: 29m 57s
  🟩 Clang18            Pass: 100%/7   | Total:  2h 40m | Avg: 22m 52s | Max: 30m 12s
  🟩 GCC7               Pass: 100%/2   | Total:  1h 04m | Avg: 32m 07s | Max: 32m 36s
  🟩 GCC8               Pass: 100%/1   | Total: 32m 55s | Avg: 32m 55s | Max: 32m 55s
  🟩 GCC9               Pass: 100%/2   | Total:  1h 08m | Avg: 34m 19s | Max: 35m 45s
  🟩 GCC10              Pass: 100%/1   | Total: 35m 15s | Avg: 35m 15s | Max: 35m 15s
  🟩 GCC11              Pass: 100%/1   | Total: 36m 31s | Avg: 36m 31s | Max: 36m 31s
  🟩 GCC12              Pass: 100%/1   | Total: 36m 03s | Avg: 36m 03s | Max: 36m 03s
  🟩 GCC13              Pass: 100%/8   | Total:  3h 01m | Avg: 22m 42s | Max: 38m 59s
  🟩 MSVC14.29          Pass: 100%/2   | Total:  1h 47m | Avg: 53m 54s | Max: 54m 08s | Hits: 173%/3672  
  🟩 MSVC14.39          Pass: 100%/3   | Total:  2h 42m | Avg: 54m 17s | Max:  1h 03m | Hits: 237%/5508  
  🟩 NVHPC24.7          Pass: 100%/2   | Total:  1h 56m | Avg: 58m 23s | Max: 59m 22s
🟩 cxx_family
  🟩 Clang              Pass: 100%/14  | Total:  6h 21m | Avg: 27m 15s | Max: 33m 27s
  🟩 GCC                Pass: 100%/16  | Total:  7h 35m | Avg: 28m 27s | Max: 38m 59s
  🟩 MSVC               Pass: 100%/5   | Total:  4h 30m | Avg: 54m 08s | Max:  1h 03m | Hits: 212%/9180  
  🟩 NVHPC              Pass: 100%/2   | Total:  1h 56m | Avg: 58m 23s | Max: 59m 22s
🟩 gpu
  🟩 v100               Pass: 100%/37  | Total: 20h 24m | Avg: 33m 05s | Max:  1h 03m | Hits: 212%/9180  
🟩 jobs
  🟩 Build              Pass: 100%/31  | Total: 18h 54m | Avg: 36m 36s | Max:  1h 03m | Hits: 173%/7344  
  🟩 TestCPU            Pass: 100%/3   | Total: 50m 57s | Avg: 16m 59s | Max: 35m 42s | Hits: 365%/1836  
  🟩 TestGPU            Pass: 100%/3   | Total: 38m 20s | Avg: 12m 46s | Max: 14m 24s
🟩 sm
  🟩 90a                Pass: 100%/1   | Total: 18m 32s | Avg: 18m 32s | Max: 18m 32s
🟩 std
  🟩 17                 Pass: 100%/14  | Total:  9h 05m | Avg: 38m 59s | Max:  1h 03m | Hits: 173%/5508  
  🟩 20                 Pass: 100%/21  | Total: 10h 36m | Avg: 30m 19s | Max:  1h 03m | Hits: 269%/3672

🟩 cccl_c_parallel: Pass: 100%/2 | Total: 9m 40s | Avg: 4m 50s | Max: 7m 28s

🟩 cpu
  🟩 amd64              Pass: 100%/2   | Total:  9m 40s | Avg:  4m 50s | Max:  7m 28s
🟩 ctk
  🟩 12.6               Pass: 100%/2   | Total:  9m 40s | Avg:  4m 50s | Max:  7m 28s
🟩 cudacxx
  🟩 nvcc12.6           Pass: 100%/2   | Total:  9m 40s | Avg:  4m 50s | Max:  7m 28s
🟩 cudacxx_family
  🟩 nvcc               Pass: 100%/2   | Total:  9m 40s | Avg:  4m 50s | Max:  7m 28s
🟩 cxx
  🟩 GCC13              Pass: 100%/2   | Total:  9m 40s | Avg:  4m 50s | Max:  7m 28s
🟩 cxx_family
  🟩 GCC                Pass: 100%/2   | Total:  9m 40s | Avg:  4m 50s | Max:  7m 28s
🟩 gpu
  🟩 v100               Pass: 100%/2   | Total:  9m 40s | Avg:  4m 50s | Max:  7m 28s
🟩 jobs
  🟩 Build              Pass: 100%/1   | Total:  2m 12s | Avg:  2m 12s | Max:  2m 12s
  🟩 Test               Pass: 100%/1   | Total:  7m 28s | Avg:  7m 28s | Max:  7m 28s

🟩 python: Pass: 100%/1 | Total: 46m 17s | Avg: 46m 17s | Max: 46m 17s

🟩 cpu
  🟩 amd64              Pass: 100%/1   | Total: 46m 17s | Avg: 46m 17s | Max: 46m 17s
🟩 ctk
  🟩 12.6               Pass: 100%/1   | Total: 46m 17s | Avg: 46m 17s | Max: 46m 17s
🟩 cudacxx
  🟩 nvcc12.6           Pass: 100%/1   | Total: 46m 17s | Avg: 46m 17s | Max: 46m 17s
🟩 cudacxx_family
  🟩 nvcc               Pass: 100%/1   | Total: 46m 17s | Avg: 46m 17s | Max: 46m 17s
🟩 cxx
  🟩 GCC13              Pass: 100%/1   | Total: 46m 17s | Avg: 46m 17s | Max: 46m 17s
🟩 cxx_family
  🟩 GCC                Pass: 100%/1   | Total: 46m 17s | Avg: 46m 17s | Max: 46m 17s
🟩 gpu
  🟩 v100               Pass: 100%/1   | Total: 46m 17s | Avg: 46m 17s | Max: 46m 17s
🟩 jobs
  🟩 Test               Pass: 100%/1   | Total: 46m 17s | Avg: 46m 17s | Max: 46m 17s

👃 Inspect Changes

Modifications in project?

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

Modifications in project or dependencies?

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

🏃‍ Runner counts (total jobs: 78)

#	Runner
53	`linux-amd64-cpu16`
11	`linux-amd64-gpu-v100-latest-1`
9	`windows-amd64-cpu16`
4	`linux-arm64-cpu16`
1	`linux-amd64-gpu-h100-latest-1-testing`

bernhardmgruber · 2025-01-23T13:42:46Z

It increasingly seems that replacing cub::Traits will break a lot of behavior in CUB, since users need to move over to using and specializing numeric_limits. We should probably split this PR in the pure deprecation, which we backport to 2.8, and the replacement which should target 3.0.

github-actions · 2025-01-28T11:37:25Z

🟨 CI finished in 3h 42m: Pass: 87%/153 | Total: 2d 22h | Avg: 27m 44s | Max: 1h 16m | Hits: 188%/7898

🟨 cub: Pass: 81%/44 | Total: 1d 13h | Avg: 50m 55s | Max: 1h 16m

🔍 cpu: amd64 🔍
  🔍 amd64              Pass:  80%/42  | Total:  1d 11h | Avg: 50m 35s | Max:  1h 16m
  🟩 arm64              Pass: 100%/2   | Total:  1h 55m | Avg: 57m 54s | Max:  1h 01m
🔍 cudacxx_family: nvcc 🔍
  🟩 ClangCUDA          Pass: 100%/2   | Total:  2h 04m | Avg:  1h 02m | Max:  1h 03m
  🔍 nvcc               Pass:  80%/42  | Total:  1d 11h | Avg: 50m 23s | Max:  1h 16m
🔍 gpu: v100 🔍
  🟩 h100               Pass: 100%/2   | Total: 43m 50s | Avg: 21m 55s | Max: 24m 28s
  🔍 v100               Pass:  80%/42  | Total:  1d 12h | Avg: 52m 18s | Max:  1h 16m
🔍 jobs: Build 🔍
  🔍 Build              Pass:  78%/37  | Total:  1d 10h | Avg: 55m 24s | Max:  1h 16m
  🟩 DeviceLaunch       Pass: 100%/1   | Total: 21m 08s | Avg: 21m 08s | Max: 21m 08s
  🟩 GraphCapture       Pass: 100%/1   | Total: 24m 13s | Avg: 24m 13s | Max: 24m 13s
  🟩 HostLaunch         Pass: 100%/3   | Total:  1h 20m | Avg: 26m 55s | Max: 33m 10s
  🟩 TestGPU            Pass: 100%/2   | Total:  1h 04m | Avg: 32m 25s | Max: 32m 37s
🟨 ctk
  🟥 12.0               Pass:   0%/5   | Total:  3h 12m | Avg: 38m 30s | Max:  1h 02m
  🟩 12.5               Pass: 100%/2   | Total:  2h 24m | Avg:  1h 12m | Max:  1h 16m
  🟨 12.6               Pass:  91%/37  | Total:  1d 07h | Avg: 51m 27s | Max:  1h 16m
🟨 cudacxx
  🟩 ClangCUDA18        Pass: 100%/2   | Total:  2h 04m | Avg:  1h 02m | Max:  1h 03m
  🟥 nvcc12.0           Pass:   0%/5   | Total:  3h 12m | Avg: 38m 30s | Max:  1h 02m
  🟩 nvcc12.5           Pass: 100%/2   | Total:  2h 24m | Avg:  1h 12m | Max:  1h 16m
  🟨 nvcc12.6           Pass:  91%/35  | Total:  1d 05h | Avg: 50m 51s | Max:  1h 16m
🟨 cxx
  🟨 Clang14            Pass:  50%/4   | Total:  3h 03m | Avg: 45m 52s | Max: 59m 22s
  🟩 Clang15            Pass: 100%/2   | Total:  1h 48m | Avg: 54m 20s | Max: 55m 01s
  🟩 Clang16            Pass: 100%/2   | Total:  1h 53m | Avg: 56m 41s | Max: 59m 21s
  🟩 Clang17            Pass: 100%/2   | Total:  1h 52m | Avg: 56m 08s | Max:  1h 00m
  🟩 Clang18            Pass: 100%/7   | Total:  5h 55m | Avg: 50m 50s | Max:  1h 03m
  🟨 GCC7               Pass:  50%/2   | Total:  1h 27m | Avg: 43m 57s | Max: 55m 25s
  🟩 GCC8               Pass: 100%/1   | Total: 53m 25s | Avg: 53m 25s | Max: 53m 25s
  🟨 GCC9               Pass:  50%/2   | Total:  1h 32m | Avg: 46m 21s | Max:  1h 00m
  🟩 GCC10              Pass: 100%/2   | Total:  1h 56m | Avg: 58m 25s | Max: 58m 32s
  🟩 GCC11              Pass: 100%/2   | Total:  1h 53m | Avg: 56m 39s | Max:  1h 00m
  🟩 GCC12              Pass: 100%/4   | Total:  2h 42m | Avg: 40m 38s | Max:  1h 01m
  🟩 GCC13              Pass: 100%/8   | Total:  5h 20m | Avg: 40m 00s | Max:  1h 01m
  🟥 MSVC14.29          Pass:   0%/2   | Total:  2h 11m | Avg:  1h 05m | Max:  1h 09m
  🟥 MSVC14.39          Pass:   0%/2   | Total:  2h 24m | Avg:  1h 12m | Max:  1h 16m
  🟩 NVHPC24.7          Pass: 100%/2   | Total:  2h 24m | Avg:  1h 12m | Max:  1h 16m
🟨 cxx_family
  🟨 Clang              Pass:  88%/17  | Total: 14h 33m | Avg: 51m 23s | Max:  1h 03m
  🟨 GCC                Pass:  90%/21  | Total: 15h 46m | Avg: 45m 05s | Max:  1h 01m
  🟥 MSVC               Pass:   0%/4   | Total:  4h 36m | Avg:  1h 09m | Max:  1h 16m
  🟩 NVHPC              Pass: 100%/2   | Total:  2h 24m | Avg:  1h 12m | Max:  1h 16m
🟩 sm
  🟩 90                 Pass: 100%/2   | Total: 43m 50s | Avg: 21m 55s | Max: 24m 28s
  🟩 90a                Pass: 100%/1   | Total: 26m 28s | Avg: 26m 28s | Max: 26m 28s
🟨 std
  🟨 17                 Pass:  70%/20  | Total: 18h 22m | Avg: 55m 07s | Max:  1h 09m
  🟨 20                 Pass:  91%/24  | Total: 18h 58m | Avg: 47m 25s | Max:  1h 16m

🟨 libcudacxx: Pass: 81%/43 | Total: 8h 12m | Avg: 11m 27s | Max: 31m 43s

🔍 cpu: amd64 🔍
  🔍 amd64              Pass:  80%/41  | Total:  8h 05m | Avg: 11m 50s | Max: 31m 43s
  🟩 arm64              Pass: 100%/2   | Total:  6m 58s | Avg:  3m 29s | Max:  3m 38s
🔍 cudacxx_family: nvcc 🔍
  🟩 ClangCUDA          Pass: 100%/4   | Total:  1h 08m | Avg: 17m 04s | Max: 20m 34s
  🔍 nvcc               Pass:  79%/39  | Total:  7h 04m | Avg: 10m 53s | Max: 31m 43s
🔍 jobs: Build 🔍
  🔍 Build              Pass:  78%/38  | Total:  6h 50m | Avg: 10m 48s | Max: 31m 43s
  🟩 NVRTC              Pass: 100%/2   | Total: 46m 11s | Avg: 23m 05s | Max: 23m 12s
  🟩 Test               Pass: 100%/2   | Total: 33m 53s | Avg: 16m 56s | Max: 17m 44s
  🟩 VerifyCodegen      Pass: 100%/1   | Total:  2m 02s | Avg:  2m 02s | Max:  2m 02s
🟨 ctk
  🟥 12.0               Pass:   0%/5   | Total:  1h 25m | Avg: 17m 00s | Max: 19m 40s
  🟩 12.5               Pass: 100%/2   | Total:  1h 02m | Avg: 31m 00s | Max: 31m 43s
  🟨 12.6               Pass:  91%/36  | Total:  5h 45m | Avg:  9m 36s | Max: 23m 12s
🟨 cudacxx
  🟩 ClangCUDA18        Pass: 100%/4   | Total:  1h 08m | Avg: 17m 04s | Max: 20m 34s
  🟥 nvcc12.0           Pass:   0%/5   | Total:  1h 25m | Avg: 17m 00s | Max: 19m 40s
  🟩 nvcc12.5           Pass: 100%/2   | Total:  1h 02m | Avg: 31m 00s | Max: 31m 43s
  🟨 nvcc12.6           Pass:  90%/32  | Total:  4h 37m | Avg:  8m 40s | Max: 23m 12s
🟨 cxx
  🟨 Clang14            Pass:  50%/4   | Total:  1h 01m | Avg: 15m 17s | Max: 20m 58s
  🟩 Clang15            Pass: 100%/2   | Total: 23m 06s | Avg: 11m 33s | Max: 18m 43s
  🟩 Clang16            Pass: 100%/2   | Total:  8m 22s | Avg:  4m 11s | Max:  4m 11s
  🟩 Clang17            Pass: 100%/2   | Total:  8m 29s | Avg:  4m 14s | Max:  4m 15s
  🟩 Clang18            Pass: 100%/8   | Total:  1h 38m | Avg: 12m 15s | Max: 20m 34s
  🟨 GCC7               Pass:  50%/2   | Total: 23m 10s | Avg: 11m 35s | Max: 19m 40s
  🟩 GCC8               Pass: 100%/1   | Total: 20m 34s | Avg: 20m 34s | Max: 20m 34s
  🟨 GCC9               Pass:  50%/2   | Total: 22m 48s | Avg: 11m 24s | Max: 18m 50s
  🟩 GCC10              Pass: 100%/2   | Total: 10m 25s | Avg:  5m 12s | Max:  6m 26s
  🟩 GCC11              Pass: 100%/2   | Total:  7m 54s | Avg:  3m 57s | Max:  4m 18s
  🟩 GCC12              Pass: 100%/2   | Total:  7m 45s | Avg:  3m 52s | Max:  4m 11s
  🟩 GCC13              Pass: 100%/8   | Total:  1h 18m | Avg:  9m 52s | Max: 23m 12s
  🟥 MSVC14.29          Pass:   0%/2   | Total: 24m 16s | Avg: 12m 08s | Max: 13m 51s
  🟥 MSVC14.39          Pass:   0%/2   | Total: 35m 42s | Avg: 17m 51s | Max: 18m 22s
  🟩 NVHPC24.7          Pass: 100%/2   | Total:  1h 02m | Avg: 31m 00s | Max: 31m 43s
🟨 cxx_family
  🟨 Clang              Pass:  88%/18  | Total:  3h 19m | Avg: 11m 04s | Max: 20m 58s
  🟨 GCC                Pass:  89%/19  | Total:  2h 51m | Avg:  9m 01s | Max: 23m 12s
  🟥 MSVC               Pass:   0%/4   | Total: 59m 58s | Avg: 14m 59s | Max: 18m 22s
  🟩 NVHPC              Pass: 100%/2   | Total:  1h 02m | Avg: 31m 00s | Max: 31m 43s
🟨 gpu
  🟨 v100               Pass:  81%/43  | Total:  8h 12m | Avg: 11m 27s | Max: 31m 43s
🟩 sm
  🟩 90                 Pass: 100%/1   | Total: 14m 05s | Avg: 14m 05s | Max: 14m 05s
  🟩 90a                Pass: 100%/2   | Total: 17m 02s | Avg:  8m 31s | Max: 13m 24s
🟨 std
  🟨 17                 Pass:  71%/21  | Total:  3h 54m | Avg: 11m 11s | Max: 30m 17s
  🟨 20                 Pass:  90%/21  | Total:  4h 15m | Avg: 12m 11s | Max: 31m 43s

🟨 thrust: Pass: 93%/43 | Total: 22h 21m | Avg: 31m 12s | Max: 1h 09m | Hits: 174%/7376

🔍 ctk: 12.6 🔍
  🟩 12.0               Pass: 100%/5   | Total:  3h 03m | Avg: 36m 43s | Max: 55m 49s | Hits: 174%/1844  
  🟩 12.5               Pass: 100%/2   | Total:  1h 50m | Avg: 55m 08s | Max: 56m 19s
  🔍 12.6               Pass:  91%/36  | Total: 17h 27m | Avg: 29m 06s | Max:  1h 09m | Hits: 174%/5532  
🔍 cudacxx: nvcc12.6 🔍
  🟩 ClangCUDA18        Pass: 100%/2   | Total: 50m 44s | Avg: 25m 22s | Max: 25m 42s
  🟩 nvcc12.0           Pass: 100%/5   | Total:  3h 03m | Avg: 36m 43s | Max: 55m 49s | Hits: 174%/1844  
  🟩 nvcc12.5           Pass: 100%/2   | Total:  1h 50m | Avg: 55m 08s | Max: 56m 19s
  🔍 nvcc12.6           Pass:  91%/34  | Total: 16h 37m | Avg: 29m 19s | Max:  1h 09m | Hits: 174%/5532  
🔍 cudacxx_family: nvcc 🔍
  🟩 ClangCUDA          Pass: 100%/2   | Total: 50m 44s | Avg: 25m 22s | Max: 25m 42s
  🔍 nvcc               Pass:  92%/41  | Total: 21h 31m | Avg: 31m 29s | Max:  1h 09m | Hits: 174%/7376  
🔍 std: 20 🔍
  🟩 17                 Pass: 100%/20  | Total: 12h 05m | Avg: 36m 15s | Max:  1h 01m | Hits: 174%/5532  
  🔍 20                 Pass:  85%/21  | Total:  9h 34m | Avg: 27m 21s | Max:  1h 09m | Hits: 174%/1844  
🟨 cxx
  🟩 Clang14            Pass: 100%/4   | Total:  1h 59m | Avg: 29m 49s | Max: 30m 40s
  🟩 Clang15            Pass: 100%/2   | Total:  1h 01m | Avg: 30m 47s | Max: 32m 08s
  🟩 Clang16            Pass: 100%/2   | Total:  1h 00m | Avg: 30m 02s | Max: 30m 45s
  🟩 Clang17            Pass: 100%/2   | Total: 56m 32s | Avg: 28m 16s | Max: 28m 26s
  🟨 Clang18            Pass:  85%/7   | Total:  2h 25m | Avg: 20m 49s | Max: 31m 34s
  🟩 GCC7               Pass: 100%/2   | Total:  1h 05m | Avg: 32m 48s | Max: 32m 57s
  🟩 GCC8               Pass: 100%/1   | Total: 29m 51s | Avg: 29m 51s | Max: 29m 51s
  🟩 GCC9               Pass: 100%/2   | Total:  1h 06m | Avg: 33m 05s | Max: 34m 58s
  🟩 GCC10              Pass: 100%/2   | Total: 58m 33s | Avg: 29m 16s | Max: 29m 48s
  🟩 GCC11              Pass: 100%/2   | Total:  1h 06m | Avg: 33m 12s | Max: 33m 42s
  🟩 GCC12              Pass: 100%/2   | Total:  1h 03m | Avg: 31m 52s | Max: 33m 23s
  🟨 GCC13              Pass:  87%/8   | Total:  2h 37m | Avg: 19m 39s | Max: 33m 55s
  🟩 MSVC14.29          Pass: 100%/2   | Total:  1h 57m | Avg: 58m 34s | Max:  1h 01m | Hits: 174%/3688  
  🟨 MSVC14.39          Pass:  66%/3   | Total:  2h 43m | Avg: 54m 31s | Max:  1h 09m | Hits: 174%/3688  
  🟩 NVHPC24.7          Pass: 100%/2   | Total:  1h 50m | Avg: 55m 08s | Max: 56m 19s
🟨 cxx_family
  🟨 Clang              Pass:  94%/17  | Total:  7h 23m | Avg: 26m 04s | Max: 32m 08s
  🟨 GCC                Pass:  94%/19  | Total:  8h 27m | Avg: 26m 42s | Max: 34m 58s
  🟨 MSVC               Pass:  80%/5   | Total:  4h 40m | Avg: 56m 08s | Max:  1h 09m | Hits: 174%/7376  
  🟩 NVHPC              Pass: 100%/2   | Total:  1h 50m | Avg: 55m 08s | Max: 56m 19s
🟨 jobs
  🟨 Build              Pass:  94%/37  | Total: 20h 37m | Avg: 33m 27s | Max:  1h 09m | Hits: 174%/7376  
  🟨 TestCPU            Pass:  66%/3   | Total: 48m 45s | Avg: 16m 15s | Max: 32m 50s
  🟩 TestGPU            Pass: 100%/3   | Total: 55m 16s | Avg: 18m 25s | Max: 18m 57s
🟨 gpu
  🟨 v100               Pass:  93%/43  | Total: 22h 21m | Avg: 31m 12s | Max:  1h 09m | Hits: 174%/7376  
🟩 cmake_options
  🟩 -DTHRUST_DISPATCH_TYPE=Force32bit Pass: 100%/2   | Total: 42m 27s | Avg: 21m 13s | Max: 25m 04s
🟨 cpu
  🟨 amd64              Pass:  97%/41  | Total: 22h 11m | Avg: 32m 28s | Max:  1h 09m | Hits: 174%/7376  
  🟥 arm64              Pass:   0%/2   | Total: 10m 20s | Avg:  5m 10s | Max:  5m 14s
🟩 sm
  🟩 90a                Pass: 100%/1   | Total: 17m 14s | Avg: 17m 14s | Max: 17m 14s

🟩 cudax: Pass: 100%/20 | Total: 1h 51m | Avg: 5m 35s | Max: 16m 29s | Hits: 383%/522

🟩 cpu
  🟩 amd64              Pass: 100%/16  | Total:  1h 40m | Avg:  6m 17s | Max: 16m 29s | Hits: 383%/522   
  🟩 arm64              Pass: 100%/4   | Total: 11m 12s | Avg:  2m 48s | Max:  3m 24s
🟩 ctk
  🟩 12.0               Pass: 100%/1   | Total: 10m 21s | Avg: 10m 21s | Max: 10m 21s | Hits: 383%/261   
  🟩 12.5               Pass: 100%/2   | Total: 13m 04s | Avg:  6m 32s | Max:  6m 38s
  🟩 12.6               Pass: 100%/17  | Total:  1h 28m | Avg:  5m 12s | Max: 16m 29s | Hits: 383%/261   
🟩 cudacxx
  🟩 nvcc12.0           Pass: 100%/1   | Total: 10m 21s | Avg: 10m 21s | Max: 10m 21s | Hits: 383%/261   
  🟩 nvcc12.5           Pass: 100%/2   | Total: 13m 04s | Avg:  6m 32s | Max:  6m 38s
  🟩 nvcc12.6           Pass: 100%/17  | Total:  1h 28m | Avg:  5m 12s | Max: 16m 29s | Hits: 383%/261   
🟩 cudacxx_family
  🟩 nvcc               Pass: 100%/20  | Total:  1h 51m | Avg:  5m 35s | Max: 16m 29s | Hits: 383%/522   
🟩 cxx
  🟩 Clang14            Pass: 100%/1   | Total:  3m 31s | Avg:  3m 31s | Max:  3m 31s
  🟩 Clang15            Pass: 100%/1   | Total:  3m 15s | Avg:  3m 15s | Max:  3m 15s
  🟩 Clang16            Pass: 100%/1   | Total:  3m 11s | Avg:  3m 11s | Max:  3m 11s
  🟩 Clang17            Pass: 100%/1   | Total:  3m 22s | Avg:  3m 22s | Max:  3m 22s
  🟩 Clang18            Pass: 100%/4   | Total: 25m 15s | Avg:  6m 18s | Max: 16m 18s
  🟩 GCC10              Pass: 100%/1   | Total:  3m 03s | Avg:  3m 03s | Max:  3m 03s
  🟩 GCC11              Pass: 100%/1   | Total:  3m 12s | Avg:  3m 12s | Max:  3m 12s
  🟩 GCC12              Pass: 100%/2   | Total: 20m 31s | Avg: 10m 15s | Max: 16m 29s
  🟩 GCC13              Pass: 100%/4   | Total: 11m 32s | Avg:  2m 53s | Max:  3m 24s
  🟩 MSVC14.36          Pass: 100%/1   | Total: 10m 21s | Avg: 10m 21s | Max: 10m 21s | Hits: 383%/261   
  🟩 MSVC14.39          Pass: 100%/1   | Total: 11m 34s | Avg: 11m 34s | Max: 11m 34s | Hits: 383%/261   
  🟩 NVHPC24.7          Pass: 100%/2   | Total: 13m 04s | Avg:  6m 32s | Max:  6m 38s
🟩 cxx_family
  🟩 Clang              Pass: 100%/8   | Total: 38m 34s | Avg:  4m 49s | Max: 16m 18s
  🟩 GCC                Pass: 100%/8   | Total: 38m 18s | Avg:  4m 47s | Max: 16m 29s
  🟩 MSVC               Pass: 100%/2   | Total: 21m 55s | Avg: 10m 57s | Max: 11m 34s | Hits: 383%/522   
  🟩 NVHPC              Pass: 100%/2   | Total: 13m 04s | Avg:  6m 32s | Max:  6m 38s
🟩 gpu
  🟩 v100               Pass: 100%/20  | Total:  1h 51m | Avg:  5m 35s | Max: 16m 29s | Hits: 383%/522   
🟩 jobs
  🟩 Build              Pass: 100%/18  | Total:  1h 19m | Avg:  4m 23s | Max: 11m 34s | Hits: 383%/522   
  🟩 Test               Pass: 100%/2   | Total: 32m 47s | Avg: 16m 23s | Max: 16m 29s
🟩 sm
  🟩 90                 Pass: 100%/1   | Total:  2m 52s | Avg:  2m 52s | Max:  2m 52s
  🟩 90a                Pass: 100%/1   | Total:  2m 48s | Avg:  2m 48s | Max:  2m 48s
🟩 std
  🟩 17                 Pass: 100%/4   | Total: 14m 35s | Avg:  3m 38s | Max:  6m 38s
  🟩 20                 Pass: 100%/16  | Total:  1h 37m | Avg:  6m 04s | Max: 16m 29s | Hits: 383%/522

🟩 cccl_c_parallel: Pass: 100%/2 | Total: 10m 24s | Avg: 5m 12s | Max: 8m 20s

🟩 cpu
  🟩 amd64              Pass: 100%/2   | Total: 10m 24s | Avg:  5m 12s | Max:  8m 20s
🟩 ctk
  🟩 12.6               Pass: 100%/2   | Total: 10m 24s | Avg:  5m 12s | Max:  8m 20s
🟩 cudacxx
  🟩 nvcc12.6           Pass: 100%/2   | Total: 10m 24s | Avg:  5m 12s | Max:  8m 20s
🟩 cudacxx_family
  🟩 nvcc               Pass: 100%/2   | Total: 10m 24s | Avg:  5m 12s | Max:  8m 20s
🟩 cxx
  🟩 GCC13              Pass: 100%/2   | Total: 10m 24s | Avg:  5m 12s | Max:  8m 20s
🟩 cxx_family
  🟩 GCC                Pass: 100%/2   | Total: 10m 24s | Avg:  5m 12s | Max:  8m 20s
🟩 gpu
  🟩 v100               Pass: 100%/2   | Total: 10m 24s | Avg:  5m 12s | Max:  8m 20s
🟩 jobs
  🟩 Build              Pass: 100%/1   | Total:  2m 04s | Avg:  2m 04s | Max:  2m 04s
  🟩 Test               Pass: 100%/1   | Total:  8m 20s | Avg:  8m 20s | Max:  8m 20s

🟩 python: Pass: 100%/1 | Total: 45m 32s | Avg: 45m 32s | Max: 45m 32s

🟩 cpu
  🟩 amd64              Pass: 100%/1   | Total: 45m 32s | Avg: 45m 32s | Max: 45m 32s
🟩 ctk
  🟩 12.6               Pass: 100%/1   | Total: 45m 32s | Avg: 45m 32s | Max: 45m 32s
🟩 cudacxx
  🟩 nvcc12.6           Pass: 100%/1   | Total: 45m 32s | Avg: 45m 32s | Max: 45m 32s
🟩 cudacxx_family
  🟩 nvcc               Pass: 100%/1   | Total: 45m 32s | Avg: 45m 32s | Max: 45m 32s
🟩 cxx
  🟩 GCC13              Pass: 100%/1   | Total: 45m 32s | Avg: 45m 32s | Max: 45m 32s
🟩 cxx_family
  🟩 GCC                Pass: 100%/1   | Total: 45m 32s | Avg: 45m 32s | Max: 45m 32s
🟩 gpu
  🟩 v100               Pass: 100%/1   | Total: 45m 32s | Avg: 45m 32s | Max: 45m 32s
🟩 jobs
  🟩 Test               Pass: 100%/1   | Total: 45m 32s | Avg: 45m 32s | Max: 45m 32s

👃 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: 153)

#	Runner
110	`linux-amd64-cpu16`
17	`linux-amd64-gpu-v100-latest-1`
15	`windows-amd64-cpu16`
10	`linux-arm64-cpu16`
1	`linux-amd64-gpu-h100-latest-1-testing`

github-actions · 2025-01-28T13:32:43Z

🟨 CI finished in 1h 28m: Pass: 87%/153 | Total: 1d 06h | Avg: 11m 47s | Max: 47m 24s | Hits: 366%/7898

🟨 cub: Pass: 81%/44 | Total: 12h 05m | Avg: 16m 29s | Max: 34m 04s

🔍 cpu: amd64 🔍
  🔍 amd64              Pass:  80%/42  | Total: 11h 47m | Avg: 16m 50s | Max: 34m 04s
  🟩 arm64              Pass: 100%/2   | Total: 18m 28s | Avg:  9m 14s | Max:  9m 35s
🔍 cudacxx_family: nvcc 🔍
  🟩 ClangCUDA          Pass: 100%/2   | Total: 14m 28s | Avg:  7m 14s | Max:  7m 26s
  🔍 nvcc               Pass:  80%/42  | Total: 11h 51m | Avg: 16m 56s | Max: 34m 04s
🔍 gpu: v100 🔍
  🟩 h100               Pass: 100%/2   | Total: 25m 47s | Avg: 12m 53s | Max: 19m 26s
  🔍 v100               Pass:  80%/42  | Total: 11h 40m | Avg: 16m 40s | Max: 34m 04s
🔍 jobs: Build 🔍
  🔍 Build              Pass:  78%/37  | Total:  9h 16m | Avg: 15m 03s | Max: 34m 04s
  🟩 DeviceLaunch       Pass: 100%/1   | Total: 24m 37s | Avg: 24m 37s | Max: 24m 37s
  🟩 GraphCapture       Pass: 100%/1   | Total: 26m 29s | Avg: 26m 29s | Max: 26m 29s
  🟩 HostLaunch         Pass: 100%/3   | Total:  1h 06m | Avg: 22m 08s | Max: 23m 36s
  🟩 TestGPU            Pass: 100%/2   | Total: 51m 19s | Avg: 25m 39s | Max: 27m 31s
🟨 ctk
  🟥 12.0               Pass:   0%/5   | Total:  2h 39m | Avg: 31m 49s | Max: 34m 04s
  🟩 12.5               Pass: 100%/2   | Total: 33m 24s | Avg: 16m 42s | Max: 16m 43s
  🟨 12.6               Pass:  91%/37  | Total:  8h 53m | Avg: 14m 24s | Max: 32m 58s
🟨 cudacxx
  🟩 ClangCUDA18        Pass: 100%/2   | Total: 14m 28s | Avg:  7m 14s | Max:  7m 26s
  🟥 nvcc12.0           Pass:   0%/5   | Total:  2h 39m | Avg: 31m 49s | Max: 34m 04s
  🟩 nvcc12.5           Pass: 100%/2   | Total: 33m 24s | Avg: 16m 42s | Max: 16m 43s
  🟨 nvcc12.6           Pass:  91%/35  | Total:  8h 38m | Avg: 14m 49s | Max: 32m 58s
🟨 cxx
  🟨 Clang14            Pass:  50%/4   | Total:  1h 28m | Avg: 22m 01s | Max: 34m 04s
  🟩 Clang15            Pass: 100%/2   | Total: 21m 44s | Avg: 10m 52s | Max: 10m 58s
  🟩 Clang16            Pass: 100%/2   | Total: 19m 35s | Avg:  9m 47s | Max: 10m 02s
  🟩 Clang17            Pass: 100%/2   | Total: 20m 06s | Avg: 10m 03s | Max: 10m 10s
  🟩 Clang18            Pass: 100%/7   | Total:  1h 31m | Avg: 13m 03s | Max: 23m 48s
  🟨 GCC7               Pass:  50%/2   | Total: 43m 03s | Avg: 21m 31s | Max: 32m 24s
  🟩 GCC8               Pass: 100%/1   | Total: 11m 28s | Avg: 11m 28s | Max: 11m 28s
  🟨 GCC9               Pass:  50%/2   | Total: 44m 35s | Avg: 22m 17s | Max: 33m 26s
  🟩 GCC10              Pass: 100%/2   | Total: 21m 02s | Avg: 10m 31s | Max: 10m 35s
  🟩 GCC11              Pass: 100%/2   | Total: 21m 34s | Avg: 10m 47s | Max: 10m 49s
  🟩 GCC12              Pass: 100%/4   | Total: 47m 13s | Avg: 11m 48s | Max: 19m 26s
  🟩 GCC13              Pass: 100%/8   | Total:  2h 20m | Avg: 17m 36s | Max: 27m 31s
  🟥 MSVC14.29          Pass:   0%/2   | Total: 56m 01s | Avg: 28m 00s | Max: 29m 55s
  🟥 MSVC14.39          Pass:   0%/2   | Total:  1h 05m | Avg: 32m 49s | Max: 32m 58s
  🟩 NVHPC24.7          Pass: 100%/2   | Total: 33m 24s | Avg: 16m 42s | Max: 16m 43s
🟨 cxx_family
  🟨 Clang              Pass:  88%/17  | Total:  4h 00m | Avg: 14m 10s | Max: 34m 04s
  🟨 GCC                Pass:  90%/21  | Total:  5h 29m | Avg: 15m 42s | Max: 33m 26s
  🟥 MSVC               Pass:   0%/4   | Total:  2h 01m | Avg: 30m 24s | Max: 32m 58s
  🟩 NVHPC              Pass: 100%/2   | Total: 33m 24s | Avg: 16m 42s | Max: 16m 43s
🟩 sm
  🟩 90                 Pass: 100%/2   | Total: 25m 47s | Avg: 12m 53s | Max: 19m 26s
  🟩 90a                Pass: 100%/1   | Total:  6m 59s | Avg:  6m 59s | Max:  6m 59s
🟨 std
  🟨 17                 Pass:  70%/20  | Total:  5h 38m | Avg: 16m 55s | Max: 33m 26s
  🟨 20                 Pass:  91%/24  | Total:  6h 27m | Avg: 16m 08s | Max: 34m 04s

🟨 libcudacxx: Pass: 81%/43 | Total: 7h 18m | Avg: 10m 11s | Max: 37m 08s

🔍 cpu: amd64 🔍
  🔍 amd64              Pass:  80%/41  | Total:  6h 54m | Avg: 10m 06s | Max: 37m 08s
  🟩 arm64              Pass: 100%/2   | Total: 24m 07s | Avg: 12m 03s | Max: 20m 28s
🔍 cudacxx_family: nvcc 🔍
  🟩 ClangCUDA          Pass: 100%/4   | Total:  1h 06m | Avg: 16m 40s | Max: 20m 23s
  🔍 nvcc               Pass:  79%/39  | Total:  6h 11m | Avg:  9m 32s | Max: 37m 08s
🔍 jobs: Build 🔍
  🔍 Build              Pass:  78%/38  | Total:  5h 22m | Avg:  8m 28s | Max: 27m 44s
  🟩 NVRTC              Pass: 100%/2   | Total:  1h 14m | Avg: 37m 00s | Max: 37m 08s
  🟩 Test               Pass: 100%/2   | Total: 40m 24s | Avg: 20m 12s | Max: 20m 29s
  🟩 VerifyCodegen      Pass: 100%/1   | Total:  1m 58s | Avg:  1m 58s | Max:  1m 58s
🟨 ctk
  🟥 12.0               Pass:   0%/5   | Total: 39m 27s | Avg:  7m 53s | Max: 17m 28s
  🟩 12.5               Pass: 100%/2   | Total: 37m 05s | Avg: 18m 32s | Max: 27m 44s
  🟨 12.6               Pass:  91%/36  | Total:  6h 02m | Avg: 10m 03s | Max: 37m 08s
🟨 cudacxx
  🟩 ClangCUDA18        Pass: 100%/4   | Total:  1h 06m | Avg: 16m 40s | Max: 20m 23s
  🟥 nvcc12.0           Pass:   0%/5   | Total: 39m 27s | Avg:  7m 53s | Max: 17m 28s
  🟩 nvcc12.5           Pass: 100%/2   | Total: 37m 05s | Avg: 18m 32s | Max: 27m 44s
  🟨 nvcc12.6           Pass:  90%/32  | Total:  4h 55m | Avg:  9m 13s | Max: 37m 08s
🟨 cxx
  🟨 Clang14            Pass:  50%/4   | Total: 30m 12s | Avg:  7m 33s | Max: 17m 28s
  🟩 Clang15            Pass: 100%/2   | Total:  8m 39s | Avg:  4m 19s | Max:  4m 29s
  🟩 Clang16            Pass: 100%/2   | Total: 11m 40s | Avg:  5m 50s | Max:  7m 21s
  🟩 Clang17            Pass: 100%/2   | Total:  8m 55s | Avg:  4m 27s | Max:  4m 31s
  🟩 Clang18            Pass: 100%/8   | Total:  1h 39m | Avg: 12m 27s | Max: 20m 29s
  🟨 GCC7               Pass:  50%/2   | Total:  6m 59s | Avg:  3m 29s | Max:  3m 33s
  🟩 GCC8               Pass: 100%/1   | Total:  3m 56s | Avg:  3m 56s | Max:  3m 56s
  🟨 GCC9               Pass:  50%/2   | Total: 11m 31s | Avg:  5m 45s | Max:  7m 50s
  🟩 GCC10              Pass: 100%/2   | Total: 22m 06s | Avg: 11m 03s | Max: 18m 06s
  🟩 GCC11              Pass: 100%/2   | Total:  7m 57s | Avg:  3m 58s | Max:  4m 02s
  🟩 GCC12              Pass: 100%/2   | Total:  8m 04s | Avg:  4m 02s | Max:  4m 10s
  🟩 GCC13              Pass: 100%/8   | Total:  2h 07m | Avg: 15m 59s | Max: 37m 08s
  🟥 MSVC14.29          Pass:   0%/2   | Total: 24m 26s | Avg: 12m 13s | Max: 13m 44s
  🟥 MSVC14.39          Pass:   0%/2   | Total: 29m 30s | Avg: 14m 45s | Max: 14m 53s
  🟩 NVHPC24.7          Pass: 100%/2   | Total: 37m 05s | Avg: 18m 32s | Max: 27m 44s
🟨 cxx_family
  🟨 Clang              Pass:  88%/18  | Total:  2h 39m | Avg:  8m 50s | Max: 20m 29s
  🟨 GCC                Pass:  89%/19  | Total:  3h 08m | Avg:  9m 55s | Max: 37m 08s
  🟥 MSVC               Pass:   0%/4   | Total: 53m 56s | Avg: 13m 29s | Max: 14m 53s
  🟩 NVHPC              Pass: 100%/2   | Total: 37m 05s | Avg: 18m 32s | Max: 27m 44s
🟨 gpu
  🟨 v100               Pass:  81%/43  | Total:  7h 18m | Avg: 10m 11s | Max: 37m 08s
🟩 sm
  🟩 90                 Pass: 100%/1   | Total: 12m 41s | Avg: 12m 41s | Max: 12m 41s
  🟩 90a                Pass: 100%/2   | Total: 17m 35s | Avg:  8m 47s | Max: 14m 05s
🟨 std
  🟨 17                 Pass:  71%/21  | Total:  3h 10m | Avg:  9m 03s | Max: 37m 08s
  🟨 20                 Pass:  90%/21  | Total:  4h 06m | Avg: 11m 44s | Max: 36m 53s

🟨 thrust: Pass: 93%/43 | Total: 7h 36m | Avg: 10m 37s | Max: 37m 00s | Hits: 365%/7376

🔍 ctk: 12.6 🔍
  🟩 12.0               Pass: 100%/5   | Total: 45m 55s | Avg:  9m 11s | Max: 24m 56s | Hits: 365%/1844  
  🟩 12.5               Pass: 100%/2   | Total: 31m 16s | Avg: 15m 38s | Max: 15m 44s
  🔍 12.6               Pass:  91%/36  | Total:  6h 19m | Avg: 10m 32s | Max: 37m 00s | Hits: 365%/5532  
🔍 cudacxx: nvcc12.6 🔍
  🟩 ClangCUDA18        Pass: 100%/2   | Total: 10m 11s | Avg:  5m 05s | Max:  5m 18s
  🟩 nvcc12.0           Pass: 100%/5   | Total: 45m 55s | Avg:  9m 11s | Max: 24m 56s | Hits: 365%/1844  
  🟩 nvcc12.5           Pass: 100%/2   | Total: 31m 16s | Avg: 15m 38s | Max: 15m 44s
  🔍 nvcc12.6           Pass:  91%/34  | Total:  6h 09m | Avg: 10m 51s | Max: 37m 00s | Hits: 365%/5532  
🔍 cudacxx_family: nvcc 🔍
  🟩 ClangCUDA          Pass: 100%/2   | Total: 10m 11s | Avg:  5m 05s | Max:  5m 18s
  🔍 nvcc               Pass:  92%/41  | Total:  7h 26m | Avg: 10m 53s | Max: 37m 00s | Hits: 365%/7376  
🔍 std: 20 🔍
  🟩 17                 Pass: 100%/20  | Total:  3h 07m | Avg:  9m 22s | Max: 29m 23s | Hits: 365%/5532  
  🔍 20                 Pass:  85%/21  | Total:  3h 49m | Avg: 10m 55s | Max: 37m 00s | Hits: 365%/1844  
🟨 cxx
  🟩 Clang14            Pass: 100%/4   | Total: 21m 29s | Avg:  5m 22s | Max:  5m 53s
  🟩 Clang15            Pass: 100%/2   | Total: 11m 03s | Avg:  5m 31s | Max:  5m 43s
  🟩 Clang16            Pass: 100%/2   | Total: 11m 12s | Avg:  5m 36s | Max:  5m 53s
  🟩 Clang17            Pass: 100%/2   | Total: 11m 37s | Avg:  5m 48s | Max:  5m 59s
  🟨 Clang18            Pass:  85%/7   | Total:  1h 00m | Avg:  8m 34s | Max: 27m 34s
  🟩 GCC7               Pass: 100%/2   | Total: 11m 01s | Avg:  5m 30s | Max:  5m 34s
  🟩 GCC8               Pass: 100%/1   | Total:  5m 15s | Avg:  5m 15s | Max:  5m 15s
  🟩 GCC9               Pass: 100%/2   | Total: 10m 54s | Avg:  5m 27s | Max:  5m 41s
  🟩 GCC10              Pass: 100%/2   | Total: 11m 03s | Avg:  5m 31s | Max:  5m 33s
  🟩 GCC11              Pass: 100%/2   | Total: 12m 05s | Avg:  6m 02s | Max:  6m 07s
  🟩 GCC12              Pass: 100%/2   | Total: 13m 04s | Avg:  6m 32s | Max:  6m 32s
  🟨 GCC13              Pass:  87%/8   | Total:  1h 33m | Avg: 11m 44s | Max: 33m 45s
  🟩 MSVC14.29          Pass: 100%/2   | Total: 54m 19s | Avg: 27m 09s | Max: 29m 23s | Hits: 365%/3688  
  🟨 MSVC14.39          Pass:  66%/3   | Total:  1h 38m | Avg: 32m 47s | Max: 37m 00s | Hits: 365%/3688  
  🟩 NVHPC24.7          Pass: 100%/2   | Total: 31m 16s | Avg: 15m 38s | Max: 15m 44s
🟨 cxx_family
  🟨 Clang              Pass:  94%/17  | Total:  1h 55m | Avg:  6m 47s | Max: 27m 34s
  🟨 GCC                Pass:  94%/19  | Total:  2h 37m | Avg:  8m 16s | Max: 33m 45s
  🟨 MSVC               Pass:  80%/5   | Total:  2h 32m | Avg: 30m 32s | Max: 37m 00s | Hits: 365%/7376  
  🟩 NVHPC              Pass: 100%/2   | Total: 31m 16s | Avg: 15m 38s | Max: 15m 44s
🟨 jobs
  🟨 Build              Pass:  94%/37  | Total:  5h 19m | Avg:  8m 38s | Max: 37m 00s | Hits: 365%/7376  
  🟨 TestCPU            Pass:  66%/3   | Total: 49m 47s | Avg: 16m 35s | Max: 33m 52s
  🟩 TestGPU            Pass: 100%/3   | Total:  1h 27m | Avg: 29m 07s | Max: 33m 45s
🟨 gpu
  🟨 v100               Pass:  93%/43  | Total:  7h 36m | Avg: 10m 37s | Max: 37m 00s | Hits: 365%/7376  
🟩 cmake_options
  🟩 -DTHRUST_DISPATCH_TYPE=Force32bit Pass: 100%/2   | Total: 39m 48s | Avg: 19m 54s | Max: 33m 45s
🟨 cpu
  🟨 amd64              Pass:  97%/41  | Total:  7h 31m | Avg: 11m 01s | Max: 37m 00s | Hits: 365%/7376  
  🟥 arm64              Pass:   0%/2   | Total:  4m 52s | Avg:  2m 26s | Max:  2m 29s
🟩 sm
  🟩 90a                Pass: 100%/1   | Total:  4m 51s | Avg:  4m 51s | Max:  4m 51s

🟩 cudax: Pass: 100%/20 | Total: 2h 03m | Avg: 6m 11s | Max: 31m 18s | Hits: 388%/522

🟩 cpu
  🟩 amd64              Pass: 100%/16  | Total:  1h 53m | Avg:  7m 04s | Max: 31m 18s | Hits: 388%/522   
  🟩 arm64              Pass: 100%/4   | Total: 10m 31s | Avg:  2m 37s | Max:  2m 40s
🟩 ctk
  🟩 12.0               Pass: 100%/1   | Total:  8m 45s | Avg:  8m 45s | Max:  8m 45s | Hits: 388%/261   
  🟩 12.5               Pass: 100%/2   | Total: 10m 31s | Avg:  5m 15s | Max:  5m 19s
  🟩 12.6               Pass: 100%/17  | Total:  1h 44m | Avg:  6m 08s | Max: 31m 18s | Hits: 388%/261   
🟩 cudacxx
  🟩 nvcc12.0           Pass: 100%/1   | Total:  8m 45s | Avg:  8m 45s | Max:  8m 45s | Hits: 388%/261   
  🟩 nvcc12.5           Pass: 100%/2   | Total: 10m 31s | Avg:  5m 15s | Max:  5m 19s
  🟩 nvcc12.6           Pass: 100%/17  | Total:  1h 44m | Avg:  6m 08s | Max: 31m 18s | Hits: 388%/261   
🟩 cudacxx_family
  🟩 nvcc               Pass: 100%/20  | Total:  2h 03m | Avg:  6m 11s | Max: 31m 18s | Hits: 388%/522   
🟩 cxx
  🟩 Clang14            Pass: 100%/1   | Total:  3m 22s | Avg:  3m 22s | Max:  3m 22s
  🟩 Clang15            Pass: 100%/1   | Total:  3m 24s | Avg:  3m 24s | Max:  3m 24s
  🟩 Clang16            Pass: 100%/1   | Total:  3m 19s | Avg:  3m 19s | Max:  3m 19s
  🟩 Clang17            Pass: 100%/1   | Total:  3m 12s | Avg:  3m 12s | Max:  3m 12s
  🟩 Clang18            Pass: 100%/4   | Total: 28m 14s | Avg:  7m 03s | Max: 19m 37s
  🟩 GCC10              Pass: 100%/1   | Total:  3m 05s | Avg:  3m 05s | Max:  3m 05s
  🟩 GCC11              Pass: 100%/1   | Total:  2m 57s | Avg:  2m 57s | Max:  2m 57s
  🟩 GCC12              Pass: 100%/2   | Total: 34m 37s | Avg: 17m 18s | Max: 31m 18s
  🟩 GCC13              Pass: 100%/4   | Total: 10m 55s | Avg:  2m 43s | Max:  2m 50s
  🟩 MSVC14.36          Pass: 100%/1   | Total:  8m 45s | Avg:  8m 45s | Max:  8m 45s | Hits: 388%/261   
  🟩 MSVC14.39          Pass: 100%/1   | Total: 11m 20s | Avg: 11m 20s | Max: 11m 20s | Hits: 388%/261   
  🟩 NVHPC24.7          Pass: 100%/2   | Total: 10m 31s | Avg:  5m 15s | Max:  5m 19s
🟩 cxx_family
  🟩 Clang              Pass: 100%/8   | Total: 41m 31s | Avg:  5m 11s | Max: 19m 37s
  🟩 GCC                Pass: 100%/8   | Total: 51m 34s | Avg:  6m 26s | Max: 31m 18s
  🟩 MSVC               Pass: 100%/2   | Total: 20m 05s | Avg: 10m 02s | Max: 11m 20s | Hits: 388%/522   
  🟩 NVHPC              Pass: 100%/2   | Total: 10m 31s | Avg:  5m 15s | Max:  5m 19s
🟩 gpu
  🟩 v100               Pass: 100%/20  | Total:  2h 03m | Avg:  6m 11s | Max: 31m 18s | Hits: 388%/522   
🟩 jobs
  🟩 Build              Pass: 100%/18  | Total:  1h 12m | Avg:  4m 02s | Max: 11m 20s | Hits: 388%/522   
  🟩 Test               Pass: 100%/2   | Total: 50m 55s | Avg: 25m 27s | Max: 31m 18s
🟩 sm
  🟩 90                 Pass: 100%/1   | Total:  2m 50s | Avg:  2m 50s | Max:  2m 50s
  🟩 90a                Pass: 100%/1   | Total:  2m 50s | Avg:  2m 50s | Max:  2m 50s
🟩 std
  🟩 17                 Pass: 100%/4   | Total: 13m 22s | Avg:  3m 20s | Max:  5m 12s
  🟩 20                 Pass: 100%/16  | Total:  1h 50m | Avg:  6m 53s | Max: 31m 18s | Hits: 388%/522

🟩 cccl_c_parallel: Pass: 100%/2 | Total: 11m 04s | Avg: 5m 32s | Max: 8m 58s

🟩 cpu
  🟩 amd64              Pass: 100%/2   | Total: 11m 04s | Avg:  5m 32s | Max:  8m 58s
🟩 ctk
  🟩 12.6               Pass: 100%/2   | Total: 11m 04s | Avg:  5m 32s | Max:  8m 58s
🟩 cudacxx
  🟩 nvcc12.6           Pass: 100%/2   | Total: 11m 04s | Avg:  5m 32s | Max:  8m 58s
🟩 cudacxx_family
  🟩 nvcc               Pass: 100%/2   | Total: 11m 04s | Avg:  5m 32s | Max:  8m 58s
🟩 cxx
  🟩 GCC13              Pass: 100%/2   | Total: 11m 04s | Avg:  5m 32s | Max:  8m 58s
🟩 cxx_family
  🟩 GCC                Pass: 100%/2   | Total: 11m 04s | Avg:  5m 32s | Max:  8m 58s
🟩 gpu
  🟩 v100               Pass: 100%/2   | Total: 11m 04s | Avg:  5m 32s | Max:  8m 58s
🟩 jobs
  🟩 Build              Pass: 100%/1   | Total:  2m 06s | Avg:  2m 06s | Max:  2m 06s
  🟩 Test               Pass: 100%/1   | Total:  8m 58s | Avg:  8m 58s | Max:  8m 58s

🟩 python: Pass: 100%/1 | Total: 47m 24s | Avg: 47m 24s | Max: 47m 24s

🟩 cpu
  🟩 amd64              Pass: 100%/1   | Total: 47m 24s | Avg: 47m 24s | Max: 47m 24s
🟩 ctk
  🟩 12.6               Pass: 100%/1   | Total: 47m 24s | Avg: 47m 24s | Max: 47m 24s
🟩 cudacxx
  🟩 nvcc12.6           Pass: 100%/1   | Total: 47m 24s | Avg: 47m 24s | Max: 47m 24s
🟩 cudacxx_family
  🟩 nvcc               Pass: 100%/1   | Total: 47m 24s | Avg: 47m 24s | Max: 47m 24s
🟩 cxx
  🟩 GCC13              Pass: 100%/1   | Total: 47m 24s | Avg: 47m 24s | Max: 47m 24s
🟩 cxx_family
  🟩 GCC                Pass: 100%/1   | Total: 47m 24s | Avg: 47m 24s | Max: 47m 24s
🟩 gpu
  🟩 v100               Pass: 100%/1   | Total: 47m 24s | Avg: 47m 24s | Max: 47m 24s
🟩 jobs
  🟩 Test               Pass: 100%/1   | Total: 47m 24s | Avg: 47m 24s | Max: 47m 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: 153)

#	Runner
110	`linux-amd64-cpu16`
17	`linux-amd64-gpu-v100-latest-1`
15	`windows-amd64-cpu16`
10	`linux-arm64-cpu16`
1	`linux-amd64-gpu-h100-latest-1-testing`

Fixes: NVIDIA#3381

bernhardmgruber added the cub For all items related to CUB label Jan 14, 2025

bernhardmgruber force-pushed the depr_cub_traits branch 7 times, most recently from cdf13ed to ac81fd5 Compare January 22, 2025 15:50

bernhardmgruber changed the title ~~Deprecate cub::Traits~~ Replace cub::Traits by numeric_limits and deprecate it Jan 22, 2025

bernhardmgruber changed the title ~~Replace cub::Traits by numeric_limits and deprecate it~~ Replace cub::Traits by numeric_limits and deprecate it Jan 22, 2025

bernhardmgruber added the backport branch/2.8.x label Jan 22, 2025

bernhardmgruber force-pushed the depr_cub_traits branch from ac81fd5 to a976d6a Compare January 22, 2025 16:39

bernhardmgruber force-pushed the depr_cub_traits branch from a976d6a to d61de28 Compare January 22, 2025 19:27

bernhardmgruber marked this pull request as ready for review January 22, 2025 19:28

bernhardmgruber requested review from a team as code owners January 22, 2025 19:28

bernhardmgruber requested a review from jrhemstad January 22, 2025 19:28

bernhardmgruber commented Jan 22, 2025

View reviewed changes

c2h/include/c2h/half.cuh Outdated Show resolved Hide resolved

bernhardmgruber force-pushed the depr_cub_traits branch from f837505 to dcb51ad Compare January 23, 2025 00:00

bernhardmgruber force-pushed the depr_cub_traits branch from dcb51ad to d95d64c Compare January 23, 2025 13:40

bernhardmgruber force-pushed the depr_cub_traits branch from b6c760c to e5689e0 Compare January 24, 2025 10:55

bernhardmgruber force-pushed the depr_cub_traits branch from 9dc04fd to e1356f4 Compare January 28, 2025 07:45

bernhardmgruber force-pushed the depr_cub_traits branch from a709c82 to 0f34a15 Compare January 28, 2025 13:46

bernhardmgruber requested a review from a team as a code owner January 28, 2025 13:46

bernhardmgruber force-pushed the depr_cub_traits branch 2 times, most recently from ef74c94 to b3e36c2 Compare January 28, 2025 17:40

bernhardmgruber added 21 commits January 29, 2025 11:21

Support FP16 traits on CTK 12.0

3c3b207

Only enable constexpr limits when supported

ca5a2c5

Support float_eq on CTK < 12.2

9fd0772

Replace cub::Traits by numeric_limits and deprecate

bf21207

Fixes: NVIDIA#3381

Fix test errors

4a0bb59

Remove dead code

de55d07

Fix histogram

d6979d1

Remove use of ::CATEGORY

74d69ad

Better deprecation messages

b2d0a7e

Fix

ac39184

Safety checks

cd2118f

Consistently use ::cuda::std::numeric_limits in CUB

2a13f54

Fix vec traits

ebb55d2

Fix FP16 traits

e88575b

Only test half/bfloat16 when libcu++ supports them

e44c054

Fix

1df087e

Namespace fix

6788dba

CI test

d2e4cd6

fix

060803c

Handling unsigned char

57c8e91

Remove some checks

7bf26e5

bernhardmgruber force-pushed the depr_cub_traits branch from b3e36c2 to 7bf26e5 Compare January 29, 2025 10:21

fix generator macro

5e3afbf

Replace cub::Traits by numeric_limits and deprecate it #3384

Are you sure you want to change the base?

Replace cub::Traits by numeric_limits and deprecate it #3384

Conversation

bernhardmgruber commented Jan 14, 2025 • edited Loading

copy-pr-bot bot commented Jan 14, 2025

bernhardmgruber commented Jan 14, 2025

bernhardmgruber commented Jan 14, 2025 • edited Loading

copy-pr-bot bot commented Jan 22, 2025

bernhardmgruber commented Jan 22, 2025

bernhardmgruber commented Jan 22, 2025

github-actions bot commented Jan 23, 2025

🟨 cub: Pass: 81%/38 | Total: 1d 08h | Avg: 51m 44s | Max: 1h 14m | Hits: 81%/2646

🟩 thrust: Pass: 100%/37 | Total: 20h 24m | Avg: 33m 05s | Max: 1h 03m | Hits: 212%/9180

🟩 cccl_c_parallel: Pass: 100%/2 | Total: 9m 40s | Avg: 4m 50s | Max: 7m 28s

🟩 python: Pass: 100%/1 | Total: 46m 17s | Avg: 46m 17s | Max: 46m 17s

👃 Inspect Changes

Modifications in project?

Modifications in project or dependencies?

🏃‍ Runner counts (total jobs: 78)

bernhardmgruber commented Jan 23, 2025

github-actions bot commented Jan 28, 2025

🟨 cub: Pass: 81%/44 | Total: 1d 13h | Avg: 50m 55s | Max: 1h 16m

🟨 libcudacxx: Pass: 81%/43 | Total: 8h 12m | Avg: 11m 27s | Max: 31m 43s

🟨 thrust: Pass: 93%/43 | Total: 22h 21m | Avg: 31m 12s | Max: 1h 09m | Hits: 174%/7376

🟩 cudax: Pass: 100%/20 | Total: 1h 51m | Avg: 5m 35s | Max: 16m 29s | Hits: 383%/522

🟩 cccl_c_parallel: Pass: 100%/2 | Total: 10m 24s | Avg: 5m 12s | Max: 8m 20s

🟩 python: Pass: 100%/1 | Total: 45m 32s | Avg: 45m 32s | Max: 45m 32s

👃 Inspect Changes

Modifications in project?

Modifications in project or dependencies?

🏃‍ Runner counts (total jobs: 153)

github-actions bot commented Jan 28, 2025

🟨 cub: Pass: 81%/44 | Total: 12h 05m | Avg: 16m 29s | Max: 34m 04s

🟨 libcudacxx: Pass: 81%/43 | Total: 7h 18m | Avg: 10m 11s | Max: 37m 08s

🟨 thrust: Pass: 93%/43 | Total: 7h 36m | Avg: 10m 37s | Max: 37m 00s | Hits: 365%/7376

🟩 cudax: Pass: 100%/20 | Total: 2h 03m | Avg: 6m 11s | Max: 31m 18s | Hits: 388%/522

🟩 cccl_c_parallel: Pass: 100%/2 | Total: 11m 04s | Avg: 5m 32s | Max: 8m 58s

🟩 python: Pass: 100%/1 | Total: 47m 24s | Avg: 47m 24s | Max: 47m 24s

👃 Inspect Changes

Modifications in project?

Modifications in project or dependencies?

🏃‍ Runner counts (total jobs: 153)

Replace `cub::Traits` by numeric_limits and deprecate it #3384

Replace `cub::Traits` by numeric_limits and deprecate it #3384

bernhardmgruber commented Jan 14, 2025 •

edited

Loading

bernhardmgruber commented Jan 14, 2025 •

edited

Loading