Backport to 2.8: Implement `cuda::std::numeric_limits` for `half` and `nv_bfloat16` (#3361) #3490

bernhardmgruber · 2025-01-22T20:01:55Z

Merge first: Backport to 2.8: Refactor limits and climits (#3221) #3488
Merge first Fix typo in limits #3491, create backport and merge that first as well

* implement builtins for huge val, nan and nans * change `INFINITY` and `NAN` implementation for NVRTC

NVIDIA#3361) * implement `cuda::std::numeric_limits` for `__half` and `__nv_bfloat16`

github-actions · 2025-01-23T00:15:13Z

🟨 CI finished in 4h 10m: Pass: 96%/170 | Total: 4d 00h | Avg: 33m 56s | Max: 1h 17m | Hits: 242%/22584

🟨 libcudacxx: Pass: 95%/48 | Total: 19h 52m | Avg: 24m 50s | Max: 1h 17m | Hits: 408%/9880

🔍 cpu: amd64 🔍
  🔍 amd64              Pass:  95%/46  | Total: 19h 09m | Avg: 24m 59s | Max:  1h 17m | Hits: 408%/9880  
  🟩 arm64              Pass: 100%/2   | Total: 42m 03s | Avg: 21m 01s | Max: 21m 10s
🔍 ctk: 11.1 🔍
  🔍 11.1               Pass:  71%/7   | Total:  2h 21m | Avg: 20m 09s | Max: 34m 50s | Hits: 453%/2287  
  🟩 12.5               Pass: 100%/2   | Total:  1h 03m | Avg: 31m 47s | Max: 32m 45s
  🟩 12.6               Pass: 100%/39  | Total: 16h 27m | Avg: 25m 18s | Max:  1h 17m | Hits: 395%/7593  
🔍 cudacxx: nvcc11.1 🔍
  🟩 ClangCUDA18        Pass: 100%/4   | Total:  1h 08m | Avg: 17m 14s | Max: 22m 04s
  🔍 nvcc11.1           Pass:  71%/7   | Total:  2h 21m | Avg: 20m 09s | Max: 34m 50s | Hits: 453%/2287  
  🟩 nvcc12.5           Pass: 100%/2   | Total:  1h 03m | Avg: 31m 47s | Max: 32m 45s
  🟩 nvcc12.6           Pass: 100%/35  | Total: 15h 18m | Avg: 26m 14s | Max:  1h 17m | Hits: 395%/7593  
🔍 cudacxx_family: nvcc 🔍
  🟩 ClangCUDA          Pass: 100%/4   | Total:  1h 08m | Avg: 17m 14s | Max: 22m 04s
  🔍 nvcc               Pass:  95%/44  | Total: 18h 43m | Avg: 25m 31s | Max:  1h 17m | Hits: 408%/9880  
🔍 jobs: Build 🔍
  🔍 Build              Pass:  95%/41  | Total: 15h 36m | Avg: 22m 51s | Max: 43m 13s | Hits: 408%/9880  
  🟩 NVRTC              Pass: 100%/4   | Total:  1h 57m | Avg: 29m 19s | Max: 32m 44s
  🟩 Test               Pass: 100%/2   | Total:  2h 15m | Avg:  1h 07m | Max:  1h 17m
  🟩 VerifyCodegen      Pass: 100%/1   | Total:  2m 17s | Avg:  2m 17s | Max:  2m 17s
🔍 std: 17 🔍
  🟩 11                 Pass: 100%/6   | Total:  2h 06m | Avg: 21m 00s | Max: 24m 30s
  🟩 14                 Pass: 100%/5   | Total:  1h 58m | Avg: 23m 47s | Max: 34m 50s | Hits: 453%/2287  
  🔍 17                 Pass:  84%/13  | Total:  5h 21m | Avg: 24m 43s | Max: 37m 50s | Hits: 401%/4964  
  🟩 20                 Pass: 100%/23  | Total: 10h 23m | Avg: 27m 06s | Max:  1h 17m | Hits: 383%/2629  
🟨 cxx
  🟨 Clang9             Pass:  75%/4   | Total:  1h 14m | Avg: 18m 33s | Max: 24m 30s
  🟩 Clang10            Pass: 100%/1   | Total: 25m 11s | Avg: 25m 11s | Max: 25m 11s
  🟩 Clang11            Pass: 100%/1   | Total: 20m 44s | Avg: 20m 44s | Max: 20m 44s
  🟩 Clang12            Pass: 100%/1   | Total: 23m 00s | Avg: 23m 00s | Max: 23m 00s
  🟩 Clang13            Pass: 100%/1   | Total: 24m 10s | Avg: 24m 10s | Max: 24m 10s
  🟩 Clang14            Pass: 100%/1   | Total: 21m 28s | Avg: 21m 28s | Max: 21m 28s
  🟩 Clang15            Pass: 100%/1   | Total: 22m 57s | Avg: 22m 57s | Max: 22m 57s
  🟩 Clang16            Pass: 100%/1   | Total: 22m 30s | Avg: 22m 30s | Max: 22m 30s
  🟩 Clang17            Pass: 100%/1   | Total: 26m 38s | Avg: 26m 38s | Max: 26m 38s
  🟩 Clang18            Pass: 100%/8   | Total:  3h 16m | Avg: 24m 31s | Max: 57m 42s
  🟩 GCC6               Pass: 100%/2   | Total: 32m 50s | Avg: 16m 25s | Max: 20m 48s
  🟩 GCC7               Pass: 100%/2   | Total: 36m 09s | Avg: 18m 04s | Max: 19m 46s
  🟩 GCC8               Pass: 100%/1   | Total: 19m 29s | Avg: 19m 29s | Max: 19m 29s
  🟨 GCC9               Pass:  66%/3   | Total: 59m 03s | Avg: 19m 41s | Max: 23m 32s
  🟩 GCC10              Pass: 100%/1   | Total: 21m 46s | Avg: 21m 46s | Max: 21m 46s
  🟩 GCC11              Pass: 100%/1   | Total: 22m 39s | Avg: 22m 39s | Max: 22m 39s
  🟩 GCC12              Pass: 100%/1   | Total: 23m 50s | Avg: 23m 50s | Max: 23m 50s
  🟩 GCC13              Pass: 100%/10  | Total:  4h 37m | Avg: 27m 42s | Max:  1h 17m
  🟩 Intel2023.2.0      Pass: 100%/1   | Total: 24m 58s | Avg: 24m 58s | Max: 24m 58s
  🟩 MSVC14.16          Pass: 100%/1   | Total: 34m 50s | Avg: 34m 50s | Max: 34m 50s | Hits: 453%/2287  
  🟩 MSVC14.29          Pass: 100%/1   | Total: 37m 50s | Avg: 37m 50s | Max: 37m 50s | Hits: 353%/2482  
  🟩 MSVC14.39          Pass: 100%/2   | Total:  1h 20m | Avg: 40m 27s | Max: 43m 13s | Hits: 416%/5111  
  🟩 NVHPC24.7          Pass: 100%/2   | Total:  1h 03m | Avg: 31m 47s | Max: 32m 45s
🟨 cxx_family
  🟨 Clang              Pass:  95%/20  | Total:  7h 37m | Avg: 22m 51s | Max: 57m 42s
  🟨 GCC                Pass:  95%/21  | Total:  8h 12m | Avg: 23m 28s | Max:  1h 17m
  🟩 Intel              Pass: 100%/1   | Total: 24m 58s | Avg: 24m 58s | Max: 24m 58s
  🟩 MSVC               Pass: 100%/4   | Total:  2h 33m | Avg: 38m 23s | Max: 43m 13s | Hits: 408%/9880  
  🟩 NVHPC              Pass: 100%/2   | Total:  1h 03m | Avg: 31m 47s | Max: 32m 45s
🟨 gpu
  🟨 v100               Pass:  95%/48  | Total: 19h 52m | Avg: 24m 50s | Max:  1h 17m | Hits: 408%/9880  
🟩 sm
  🟩 90                 Pass: 100%/1   | Total: 12m 58s | Avg: 12m 58s | Max: 12m 58s
  🟩 90a                Pass: 100%/2   | Total: 28m 45s | Avg: 14m 22s | Max: 15m 54s

🟨 cub: Pass: 95%/47 | Total: 1d 15h | Avg: 50m 53s | Max: 1h 13m | Hits: 32%/3132

🔍 cpu: amd64 🔍
  🔍 amd64              Pass:  95%/45  | Total:  1d 13h | Avg: 50m 29s | Max:  1h 13m | Hits:  32%/3132  
  🟩 arm64              Pass: 100%/2   | Total:  1h 59m | Avg: 59m 38s | Max:  1h 01m
🔍 ctk: 11.1 🔍
  🔍 11.1               Pass:  71%/7   | Total:  4h 25m | Avg: 37m 58s | Max:  1h 07m | Hits:  34%/783   
  🟩 12.5               Pass: 100%/2   | Total:  2h 23m | Avg:  1h 11m | Max:  1h 11m
  🟩 12.6               Pass: 100%/38  | Total:  1d 09h | Avg: 52m 10s | Max:  1h 13m | Hits:  32%/2349  
🔍 cudacxx: nvcc11.1 🔍
  🟩 ClangCUDA18        Pass: 100%/2   | Total:  2h 13m | Avg:  1h 06m | Max:  1h 11m
  🔍 nvcc11.1           Pass:  71%/7   | Total:  4h 25m | Avg: 37m 58s | Max:  1h 07m | Hits:  34%/783   
  🟩 nvcc12.5           Pass: 100%/2   | Total:  2h 23m | Avg:  1h 11m | Max:  1h 11m
  🟩 nvcc12.6           Pass: 100%/36  | Total:  1d 06h | Avg: 51m 21s | Max:  1h 13m | Hits:  32%/2349  
🔍 cudacxx_family: nvcc 🔍
  🟩 ClangCUDA          Pass: 100%/2   | Total:  2h 13m | Avg:  1h 06m | Max:  1h 11m
  🔍 nvcc               Pass:  95%/45  | Total:  1d 13h | Avg: 50m 10s | Max:  1h 13m | Hits:  32%/3132  
🔍 gpu: v100 🔍
  🟩 h100               Pass: 100%/2   | Total: 41m 41s | Avg: 20m 50s | Max: 25m 27s
  🔍 v100               Pass:  95%/45  | Total:  1d 15h | Avg: 52m 13s | Max:  1h 13m | Hits:  32%/3132  
🔍 jobs: Build 🔍
  🔍 Build              Pass:  95%/40  | Total:  1d 12h | Avg: 55m 24s | Max:  1h 13m | Hits:  32%/3132  
  🟩 DeviceLaunch       Pass: 100%/1   | Total: 17m 42s | Avg: 17m 42s | Max: 17m 42s
  🟩 GraphCapture       Pass: 100%/1   | Total: 22m 07s | Avg: 22m 07s | Max: 22m 07s
  🟩 HostLaunch         Pass: 100%/3   | Total: 52m 25s | Avg: 17m 28s | Max: 18m 18s
  🟩 TestGPU            Pass: 100%/2   | Total:  1h 23m | Avg: 41m 34s | Max:  1h 03m
🔍 std: 17 🔍
  🟩 11                 Pass: 100%/5   | Total:  4h 14m | Avg: 50m 52s | Max: 55m 57s
  🟩 14                 Pass: 100%/4   | Total:  3h 55m | Avg: 58m 55s | Max:  1h 07m | Hits:  34%/783   
  🔍 17                 Pass:  83%/12  | Total: 10h 38m | Avg: 53m 13s | Max:  1h 13m | Hits:  33%/1566  
  🟩 20                 Pass: 100%/26  | Total: 21h 02m | Avg: 48m 33s | Max:  1h 11m | Hits:  29%/783   
🟨 cxx
  🟨 Clang9             Pass:  75%/4   | Total:  2h 43m | Avg: 40m 53s | Max: 56m 55s
  🟩 Clang10            Pass: 100%/1   | Total: 56m 39s | Avg: 56m 39s | Max: 56m 39s
  🟩 Clang11            Pass: 100%/1   | Total:  1h 00m | Avg:  1h 00m | Max:  1h 00m
  🟩 Clang12            Pass: 100%/1   | Total:  1h 01m | Avg:  1h 01m | Max:  1h 01m
  🟩 Clang13            Pass: 100%/1   | Total: 54m 18s | Avg: 54m 18s | Max: 54m 18s
  🟩 Clang14            Pass: 100%/1   | Total: 55m 00s | Avg: 55m 00s | Max: 55m 00s
  🟩 Clang15            Pass: 100%/1   | Total: 57m 58s | Avg: 57m 58s | Max: 57m 58s
  🟩 Clang16            Pass: 100%/1   | Total:  1h 01m | Avg:  1h 01m | Max:  1h 01m
  🟩 Clang17            Pass: 100%/1   | Total:  1h 00m | Avg:  1h 00m | Max:  1h 00m
  🟩 Clang18            Pass: 100%/7   | Total:  6h 30m | Avg: 55m 42s | Max:  1h 11m
  🟩 GCC6               Pass: 100%/2   | Total:  1h 38m | Avg: 49m 24s | Max: 49m 26s
  🟩 GCC7               Pass: 100%/2   | Total:  1h 56m | Avg: 58m 17s | Max:  1h 01m
  🟩 GCC8               Pass: 100%/1   | Total: 59m 46s | Avg: 59m 46s | Max: 59m 46s
  🟨 GCC9               Pass:  66%/3   | Total:  1h 48m | Avg: 36m 12s | Max: 59m 49s
  🟩 GCC10              Pass: 100%/1   | Total: 59m 34s | Avg: 59m 34s | Max: 59m 34s
  🟩 GCC11              Pass: 100%/1   | Total: 56m 46s | Avg: 56m 46s | Max: 56m 46s
  🟩 GCC12              Pass: 100%/3   | Total:  1h 43m | Avg: 34m 22s | Max:  1h 01m
  🟩 GCC13              Pass: 100%/8   | Total:  4h 41m | Avg: 35m 14s | Max:  1h 01m
  🟩 Intel2023.2.0      Pass: 100%/1   | Total:  1h 03m | Avg:  1h 03m | Max:  1h 03m
  🟩 MSVC14.16          Pass: 100%/1   | Total:  1h 07m | Avg:  1h 07m | Max:  1h 07m | Hits:  34%/783   
  🟩 MSVC14.29          Pass: 100%/1   | Total:  1h 08m | Avg:  1h 08m | Max:  1h 08m | Hits:  33%/783   
  🟩 MSVC14.39          Pass: 100%/2   | Total:  2h 23m | Avg:  1h 11m | Max:  1h 13m | Hits:  31%/1566  
  🟩 NVHPC24.7          Pass: 100%/2   | Total:  2h 23m | Avg:  1h 11m | Max:  1h 11m
🟨 cxx_family
  🟨 Clang              Pass:  94%/19  | Total: 16h 59m | Avg: 53m 40s | Max:  1h 11m
  🟨 GCC                Pass:  95%/21  | Total: 14h 45m | Avg: 42m 08s | Max:  1h 01m
  🟩 Intel              Pass: 100%/1   | Total:  1h 03m | Avg:  1h 03m | Max:  1h 03m
  🟩 MSVC               Pass: 100%/4   | Total:  4h 39m | Avg:  1h 09m | Max:  1h 13m | Hits:  32%/3132  
  🟩 NVHPC              Pass: 100%/2   | Total:  2h 23m | Avg:  1h 11m | Max:  1h 11m
🟩 sm
  🟩 90                 Pass: 100%/2   | Total: 41m 41s | Avg: 20m 50s | Max: 25m 27s
  🟩 90a                Pass: 100%/1   | Total: 26m 52s | Avg: 26m 52s | Max: 26m 52s

🟨 thrust: Pass: 95%/46 | Total: 1d 05h | Avg: 38m 11s | Max: 1h 16m | Hits: 135%/9260

🔍 cpu: amd64 🔍
  🔍 amd64              Pass:  95%/44  | Total:  1d 03h | Avg: 38m 08s | Max:  1h 16m | Hits: 135%/9260  
  🟩 arm64              Pass: 100%/2   | Total:  1h 19m | Avg: 39m 31s | Max: 42m 44s
🔍 ctk: 11.1 🔍
  🔍 11.1               Pass:  71%/7   | Total:  3h 14m | Avg: 27m 47s | Max:  1h 11m | Hits:  92%/1852  
  🟩 12.5               Pass: 100%/2   | Total:  2h 23m | Avg:  1h 11m | Max:  1h 16m
  🟩 12.6               Pass: 100%/37  | Total: 23h 38m | Avg: 38m 21s | Max:  1h 13m | Hits: 146%/7408  
🔍 cudacxx: nvcc11.1 🔍
  🟩 ClangCUDA18        Pass: 100%/2   | Total:  1h 17m | Avg: 38m 48s | Max: 39m 28s
  🔍 nvcc11.1           Pass:  71%/7   | Total:  3h 14m | Avg: 27m 47s | Max:  1h 11m | Hits:  92%/1852  
  🟩 nvcc12.5           Pass: 100%/2   | Total:  2h 23m | Avg:  1h 11m | Max:  1h 16m
  🟩 nvcc12.6           Pass: 100%/35  | Total: 22h 21m | Avg: 38m 19s | Max:  1h 13m | Hits: 146%/7408  
🔍 cudacxx_family: nvcc 🔍
  🟩 ClangCUDA          Pass: 100%/2   | Total:  1h 17m | Avg: 38m 48s | Max: 39m 28s
  🔍 nvcc               Pass:  95%/44  | Total:  1d 03h | Avg: 38m 10s | Max:  1h 16m | Hits: 135%/9260  
🔍 jobs: Build 🔍
  🔍 Build              Pass:  95%/40  | Total:  1d 03h | Avg: 41m 22s | Max:  1h 16m | Hits:  78%/7408  
  🟩 TestCPU            Pass: 100%/3   | Total: 53m 24s | Avg: 17m 48s | Max: 37m 34s | Hits: 365%/1852  
  🟩 TestGPU            Pass: 100%/3   | Total: 48m 31s | Avg: 16m 10s | Max: 16m 55s
🔍 std: 17 🔍
  🟩 11                 Pass: 100%/5   | Total:  2h 24m | Avg: 28m 51s | Max: 30m 44s
  🟩 14                 Pass: 100%/4   | Total:  3h 05m | Avg: 46m 15s | Max:  1h 11m | Hits:  92%/1852  
  🔍 17                 Pass:  83%/12  | Total:  8h 19m | Avg: 41m 36s | Max:  1h 13m | Hits:  71%/3704  
  🟩 20                 Pass: 100%/23  | Total: 14h 37m | Avg: 38m 07s | Max:  1h 16m | Hits: 221%/3704  
🟨 cxx
  🟨 Clang9             Pass:  75%/4   | Total:  1h 41m | Avg: 25m 17s | Max: 41m 53s
  🟩 Clang10            Pass: 100%/1   | Total: 38m 28s | Avg: 38m 28s | Max: 38m 28s
  🟩 Clang11            Pass: 100%/1   | Total: 42m 12s | Avg: 42m 12s | Max: 42m 12s
  🟩 Clang12            Pass: 100%/1   | Total: 39m 06s | Avg: 39m 06s | Max: 39m 06s
  🟩 Clang13            Pass: 100%/1   | Total: 37m 06s | Avg: 37m 06s | Max: 37m 06s
  🟩 Clang14            Pass: 100%/1   | Total: 37m 12s | Avg: 37m 12s | Max: 37m 12s
  🟩 Clang15            Pass: 100%/1   | Total: 40m 11s | Avg: 40m 11s | Max: 40m 11s
  🟩 Clang16            Pass: 100%/1   | Total: 39m 45s | Avg: 39m 45s | Max: 39m 45s
  🟩 Clang17            Pass: 100%/1   | Total: 40m 12s | Avg: 40m 12s | Max: 40m 12s
  🟩 Clang18            Pass: 100%/7   | Total:  3h 35m | Avg: 30m 44s | Max: 39m 28s
  🟩 GCC6               Pass: 100%/2   | Total:  1h 06m | Avg: 33m 01s | Max: 36m 10s
  🟩 GCC7               Pass: 100%/2   | Total:  1h 05m | Avg: 32m 59s | Max: 35m 19s
  🟩 GCC8               Pass: 100%/1   | Total: 36m 50s | Avg: 36m 50s | Max: 36m 50s
  🟨 GCC9               Pass:  66%/3   | Total:  1h 08m | Avg: 22m 50s | Max: 40m 15s
  🟩 GCC10              Pass: 100%/1   | Total: 48m 27s | Avg: 48m 27s | Max: 48m 27s
  🟩 GCC11              Pass: 100%/1   | Total: 45m 17s | Avg: 45m 17s | Max: 45m 17s
  🟩 GCC12              Pass: 100%/1   | Total: 44m 08s | Avg: 44m 08s | Max: 44m 08s
  🟩 GCC13              Pass: 100%/8   | Total:  3h 49m | Avg: 28m 43s | Max: 46m 05s
  🟩 Intel2023.2.0      Pass: 100%/1   | Total: 55m 31s | Avg: 55m 31s | Max: 55m 31s
  🟩 MSVC14.16          Pass: 100%/1   | Total:  1h 11m | Avg:  1h 11m | Max:  1h 11m | Hits:  92%/1852  
  🟩 MSVC14.29          Pass: 100%/1   | Total:  1h 08m | Avg:  1h 08m | Max:  1h 08m | Hits:  78%/1852  
  🟩 MSVC14.39          Pass: 100%/3   | Total:  3h 01m | Avg:  1h 00m | Max:  1h 13m | Hits: 169%/5556  
  🟩 NVHPC24.7          Pass: 100%/2   | Total:  2h 23m | Avg:  1h 11m | Max:  1h 16m
🟨 cxx_family
  🟨 Clang              Pass:  94%/19  | Total: 10h 30m | Avg: 33m 11s | Max: 42m 12s
  🟨 GCC                Pass:  94%/19  | Total: 10h 04m | Avg: 31m 50s | Max: 48m 27s
  🟩 Intel              Pass: 100%/1   | Total: 55m 31s | Avg: 55m 31s | Max: 55m 31s
  🟩 MSVC               Pass: 100%/5   | Total:  5h 22m | Avg:  1h 04m | Max:  1h 13m | Hits: 135%/9260  
  🟩 NVHPC              Pass: 100%/2   | Total:  2h 23m | Avg:  1h 11m | Max:  1h 16m
🟨 gpu
  🟨 v100               Pass:  95%/46  | Total:  1d 05h | Avg: 38m 11s | Max:  1h 16m | Hits: 135%/9260  
🟩 cmake_options
  🟩 -DTHRUST_DISPATCH_TYPE=Force32bit Pass: 100%/2   | Total: 51m 30s | Avg: 25m 45s | Max: 36m 41s
🟩 sm
  🟩 90a                Pass: 100%/1   | Total: 26m 07s | Avg: 26m 07s | Max: 26m 07s

🟩 cudax: Pass: 100%/26 | Total: 6h 27m | Avg: 14m 55s | Max: 21m 12s | Hits: 223%/312

🟩 cpu
  🟩 amd64              Pass: 100%/22  | Total:  5h 35m | Avg: 15m 13s | Max: 21m 12s | Hits: 223%/312   
  🟩 arm64              Pass: 100%/4   | Total: 52m 57s | Avg: 13m 14s | Max: 14m 06s
🟩 ctk
  🟩 12.0               Pass: 100%/3   | Total: 42m 14s | Avg: 14m 04s | Max: 14m 49s | Hits: 223%/156   
  🟩 12.5               Pass: 100%/2   | Total: 19m 00s | Avg:  9m 30s | Max:  9m 45s
  🟩 12.6               Pass: 100%/21  | Total:  5h 26m | Avg: 15m 33s | Max: 21m 12s | Hits: 223%/156   
🟩 cudacxx
  🟩 nvcc12.0           Pass: 100%/3   | Total: 42m 14s | Avg: 14m 04s | Max: 14m 49s | Hits: 223%/156   
  🟩 nvcc12.5           Pass: 100%/2   | Total: 19m 00s | Avg:  9m 30s | Max:  9m 45s
  🟩 nvcc12.6           Pass: 100%/21  | Total:  5h 26m | Avg: 15m 33s | Max: 21m 12s | Hits: 223%/156   
🟩 cudacxx_family
  🟩 nvcc               Pass: 100%/26  | Total:  6h 27m | Avg: 14m 55s | Max: 21m 12s | Hits: 223%/312   
🟩 cxx
  🟩 Clang9             Pass: 100%/1   | Total: 14m 49s | Avg: 14m 49s | Max: 14m 49s
  🟩 Clang10            Pass: 100%/1   | Total: 17m 52s | Avg: 17m 52s | Max: 17m 52s
  🟩 Clang11            Pass: 100%/1   | Total: 15m 12s | Avg: 15m 12s | Max: 15m 12s
  🟩 Clang12            Pass: 100%/1   | Total: 15m 20s | Avg: 15m 20s | Max: 15m 20s
  🟩 Clang13            Pass: 100%/1   | Total: 13m 59s | Avg: 13m 59s | Max: 13m 59s
  🟩 Clang14            Pass: 100%/1   | Total: 15m 15s | Avg: 15m 15s | Max: 15m 15s
  🟩 Clang15            Pass: 100%/1   | Total: 18m 46s | Avg: 18m 46s | Max: 18m 46s
  🟩 Clang16            Pass: 100%/1   | Total: 14m 47s | Avg: 14m 47s | Max: 14m 47s
  🟩 Clang17            Pass: 100%/1   | Total: 14m 48s | Avg: 14m 48s | Max: 14m 48s
  🟩 Clang18            Pass: 100%/4   | Total:  1h 02m | Avg: 15m 40s | Max: 20m 18s
  🟩 GCC9               Pass: 100%/1   | Total: 13m 42s | Avg: 13m 42s | Max: 13m 42s
  🟩 GCC10              Pass: 100%/1   | Total: 16m 58s | Avg: 16m 58s | Max: 16m 58s
  🟩 GCC11              Pass: 100%/1   | Total: 16m 55s | Avg: 16m 55s | Max: 16m 55s
  🟩 GCC12              Pass: 100%/2   | Total: 39m 15s | Avg: 19m 37s | Max: 21m 12s
  🟩 GCC13              Pass: 100%/4   | Total: 50m 48s | Avg: 12m 42s | Max: 14m 06s
  🟩 MSVC14.36          Pass: 100%/1   | Total: 13m 43s | Avg: 13m 43s | Max: 13m 43s | Hits: 223%/156   
  🟩 MSVC14.39          Pass: 100%/1   | Total: 14m 08s | Avg: 14m 08s | Max: 14m 08s | Hits: 223%/156   
  🟩 NVHPC24.7          Pass: 100%/2   | Total: 19m 00s | Avg:  9m 30s | Max:  9m 45s
🟩 cxx_family
  🟩 Clang              Pass: 100%/13  | Total:  3h 23m | Avg: 15m 39s | Max: 20m 18s
  🟩 GCC                Pass: 100%/9   | Total:  2h 17m | Avg: 15m 17s | Max: 21m 12s
  🟩 MSVC               Pass: 100%/2   | Total: 27m 51s | Avg: 13m 55s | Max: 14m 08s | Hits: 223%/312   
  🟩 NVHPC              Pass: 100%/2   | Total: 19m 00s | Avg:  9m 30s | Max:  9m 45s
🟩 gpu
  🟩 v100               Pass: 100%/26  | Total:  6h 27m | Avg: 14m 55s | Max: 21m 12s | Hits: 223%/312   
🟩 jobs
  🟩 Build              Pass: 100%/24  | Total:  5h 46m | Avg: 14m 26s | Max: 18m 46s | Hits: 223%/312   
  🟩 Test               Pass: 100%/2   | Total: 41m 30s | Avg: 20m 45s | Max: 21m 12s
🟩 sm
  🟩 90                 Pass: 100%/1   | Total: 11m 31s | Avg: 11m 31s | Max: 11m 31s
  🟩 90a                Pass: 100%/1   | Total: 12m 22s | Avg: 12m 22s | Max: 12m 22s
🟩 std
  🟩 17                 Pass: 100%/6   | Total:  1h 14m | Avg: 12m 26s | Max: 14m 49s
  🟩 20                 Pass: 100%/20  | Total:  5h 13m | Avg: 15m 40s | Max: 21m 12s | Hits: 223%/312

🟩 cccl_c_parallel: Pass: 100%/2 | Total: 11m 04s | Avg: 5m 32s | Max: 8m 55s

🟩 cpu
  🟩 amd64              Pass: 100%/2   | Total: 11m 04s | Avg:  5m 32s | Max:  8m 55s
🟩 ctk
  🟩 12.6               Pass: 100%/2   | Total: 11m 04s | Avg:  5m 32s | Max:  8m 55s
🟩 cudacxx
  🟩 nvcc12.6           Pass: 100%/2   | Total: 11m 04s | Avg:  5m 32s | Max:  8m 55s
🟩 cudacxx_family
  🟩 nvcc               Pass: 100%/2   | Total: 11m 04s | Avg:  5m 32s | Max:  8m 55s
🟩 cxx
  🟩 GCC13              Pass: 100%/2   | Total: 11m 04s | Avg:  5m 32s | Max:  8m 55s
🟩 cxx_family
  🟩 GCC                Pass: 100%/2   | Total: 11m 04s | Avg:  5m 32s | Max:  8m 55s
🟩 gpu
  🟩 v100               Pass: 100%/2   | Total: 11m 04s | Avg:  5m 32s | Max:  8m 55s
🟩 jobs
  🟩 Build              Pass: 100%/1   | Total:  2m 09s | Avg:  2m 09s | Max:  2m 09s
  🟩 Test               Pass: 100%/1   | Total:  8m 55s | Avg:  8m 55s | Max:  8m 55s

🟩 python: Pass: 100%/1 | Total: 31m 16s | Avg: 31m 16s | Max: 31m 16s

🟩 cpu
  🟩 amd64              Pass: 100%/1   | Total: 31m 16s | Avg: 31m 16s | Max: 31m 16s
🟩 ctk
  🟩 12.6               Pass: 100%/1   | Total: 31m 16s | Avg: 31m 16s | Max: 31m 16s
🟩 cudacxx
  🟩 nvcc12.6           Pass: 100%/1   | Total: 31m 16s | Avg: 31m 16s | Max: 31m 16s
🟩 cudacxx_family
  🟩 nvcc               Pass: 100%/1   | Total: 31m 16s | Avg: 31m 16s | Max: 31m 16s
🟩 cxx
  🟩 GCC13              Pass: 100%/1   | Total: 31m 16s | Avg: 31m 16s | Max: 31m 16s
🟩 cxx_family
  🟩 GCC                Pass: 100%/1   | Total: 31m 16s | Avg: 31m 16s | Max: 31m 16s
🟩 gpu
  🟩 v100               Pass: 100%/1   | Total: 31m 16s | Avg: 31m 16s | Max: 31m 16s
🟩 jobs
  🟩 Test               Pass: 100%/1   | Total: 31m 16s | Avg: 31m 16s | Max: 31m 16s

👃 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`

davebayer added 2 commits January 22, 2025 20:59

Refactor limits and climits (NVIDIA#3221)

972076f

* implement builtins for huge val, nan and nans * change `INFINITY` and `NAN` implementation for NVRTC

Implement cuda::std::numeric_limits for __half and __nv_bfloat16 (

42ac268

NVIDIA#3361) * implement `cuda::std::numeric_limits` for `__half` and `__nv_bfloat16`

bernhardmgruber requested review from a team as code owners January 22, 2025 20:01

bernhardmgruber requested review from alliepiper, gonidelis and griwes January 22, 2025 20:01

bernhardmgruber mentioned this pull request Jan 22, 2025

Backport to 2.8: Specialize cuda::std::numeric_limits for FP8 types (#3478) #3492

Open

1 task

miscco approved these changes Jan 22, 2025

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Backport to 2.8: Implement `cuda::std::numeric_limits` for `half` and `nv_bfloat16` (#3361) #3490

Backport to 2.8: Implement `cuda::std::numeric_limits` for `half` and `nv_bfloat16` (#3361) #3490

bernhardmgruber commented Jan 22, 2025 •

edited

Loading

github-actions bot commented Jan 23, 2025

🟨 libcudacxx: Pass: 95%/48 | Total: 19h 52m | Avg: 24m 50s | Max: 1h 17m | Hits: 408%/9880

🟨 cub: Pass: 95%/47 | Total: 1d 15h | Avg: 50m 53s | Max: 1h 13m | Hits: 32%/3132

🟨 thrust: Pass: 95%/46 | Total: 1d 05h | Avg: 38m 11s | Max: 1h 16m | Hits: 135%/9260

🟩 cudax: Pass: 100%/26 | Total: 6h 27m | Avg: 14m 55s | Max: 21m 12s | Hits: 223%/312

🟩 cccl_c_parallel: Pass: 100%/2 | Total: 11m 04s | Avg: 5m 32s | Max: 8m 55s

🟩 python: Pass: 100%/1 | Total: 31m 16s | Avg: 31m 16s | Max: 31m 16s

👃 Inspect Changes

Modifications in project?

Modifications in project or dependencies?

🏃‍ Runner counts (total jobs: 170)

Backport to 2.8: Implement cuda::std::numeric_limits for __half and __nv_bfloat16 (#3361) #3490

Are you sure you want to change the base?

Backport to 2.8: Implement cuda::std::numeric_limits for __half and __nv_bfloat16 (#3361) #3490

Conversation

bernhardmgruber commented Jan 22, 2025 • edited Loading

github-actions bot commented Jan 23, 2025

🟨 libcudacxx: Pass: 95%/48 | Total: 19h 52m | Avg: 24m 50s | Max: 1h 17m | Hits: 408%/9880

🟨 cub: Pass: 95%/47 | Total: 1d 15h | Avg: 50m 53s | Max: 1h 13m | Hits: 32%/3132

🟨 thrust: Pass: 95%/46 | Total: 1d 05h | Avg: 38m 11s | Max: 1h 16m | Hits: 135%/9260

🟩 cudax: Pass: 100%/26 | Total: 6h 27m | Avg: 14m 55s | Max: 21m 12s | Hits: 223%/312

🟩 cccl_c_parallel: Pass: 100%/2 | Total: 11m 04s | Avg: 5m 32s | Max: 8m 55s

🟩 python: Pass: 100%/1 | Total: 31m 16s | Avg: 31m 16s | Max: 31m 16s

👃 Inspect Changes

Modifications in project?

Modifications in project or dependencies?

🏃‍ Runner counts (total jobs: 170)

Backport to 2.8: Implement `cuda::std::numeric_limits` for `half` and `nv_bfloat16` (#3361) #3490

Backport to 2.8: Implement `cuda::std::numeric_limits` for `half` and `nv_bfloat16` (#3361) #3490

bernhardmgruber commented Jan 22, 2025 •

edited

Loading