Drop `thrust::constant_iterator` in favor of `cuda::constant_iterator`
🟨 CI finished in 45m 50s: Pass: 5%/210 | Total: 20h 06m | Avg: 5m 44s | Max: 37m 18s | Hits: 90%/42
-
🟥 cub: Pass: 0%/50 | Total: 4h 38m | Avg: 5m 34s | Max: 35m 38s
🟥 cpu 🟥 amd64 Pass: 0%/48 | Total: 4h 34m | Avg: 5m 43s | Max: 35m 38s 🟥 arm64 Pass: 0%/2 | Total: 3m 57s | Avg: 1m 58s | Max: 2m 12s 🟥 ctk 🟥 12.0 Pass: 0%/5 | Total: 41m 31s | Avg: 8m 18s | Max: 33m 11s 🟥 12.9 Pass: 0%/45 | Total: 3h 56m | Avg: 5m 15s | Max: 35m 38s 🟥 cudacxx 🟥 ClangCUDA19 Pass: 0%/2 | Total: 4m 18s | Avg: 2m 09s | Max: 2m 09s 🟥 nvcc12.0 Pass: 0%/5 | Total: 41m 31s | Avg: 8m 18s | Max: 33m 11s 🟥 nvcc12.9 Pass: 0%/43 | Total: 3h 52m | Avg: 5m 24s | Max: 35m 38s 🟥 cudacxx_family 🟥 ClangCUDA Pass: 0%/2 | Total: 4m 18s | Avg: 2m 09s | Max: 2m 09s 🟥 nvcc Pass: 0%/48 | Total: 4h 34m | Avg: 5m 42s | Max: 35m 38s 🟥 cxx 🟥 Clang14 Pass: 0%/4 | Total: 9m 38s | Avg: 2m 24s | Max: 2m 29s 🟥 Clang15 Pass: 0%/2 | Total: 4m 55s | Avg: 2m 27s | Max: 2m 31s 🟥 Clang16 Pass: 0%/2 | Total: 4m 58s | Avg: 2m 29s | Max: 2m 30s 🟥 Clang17 Pass: 0%/2 | Total: 5m 01s | Avg: 2m 30s | Max: 2m 36s 🟥 Clang18 Pass: 0%/2 | Total: 5m 05s | Avg: 2m 32s | Max: 2m 39s 🟥 Clang19 Pass: 0%/7 | Total: 11m 11s | Avg: 1m 35s | Max: 2m 35s 🟥 GCC7 Pass: 0%/2 | Total: 3m 46s | Avg: 1m 53s | Max: 2m 01s 🟥 GCC8 Pass: 0%/1 | Total: 2m 26s | Avg: 2m 26s | Max: 2m 26s 🟥 GCC9 Pass: 0%/2 | Total: 3m 54s | Avg: 1m 57s | Max: 2m 03s 🟥 GCC10 Pass: 0%/2 | Total: 4m 01s | Avg: 2m 00s | Max: 2m 02s 🟥 GCC11 Pass: 0%/2 | Total: 4m 04s | Avg: 2m 02s | Max: 2m 02s 🟥 GCC12 Pass: 0%/2 | Total: 4m 09s | Avg: 2m 04s | Max: 2m 06s 🟥 GCC13 Pass: 0%/12 | Total: 12m 40s | Avg: 1m 03s | Max: 2m 15s 🟥 MSVC14.29 Pass: 0%/2 | Total: 1h 05m | Avg: 32m 48s | Max: 33m 11s 🟥 MSVC14.43 Pass: 0%/4 | Total: 2h 07m | Avg: 31m 59s | Max: 35m 38s 🟥 NVHPC25.5 Pass: 0%/2 | Total: 9m 09s | Avg: 4m 34s | Max: 4m 37s 🟥 cxx_family 🟥 Clang Pass: 0%/19 | Total: 40m 48s | Avg: 2m 08s | Max: 2m 39s 🟥 GCC Pass: 0%/23 | Total: 35m 00s | Avg: 1m 31s | Max: 2m 26s 🟥 MSVC Pass: 0%/6 | Total: 3h 13m | Avg: 32m 15s | Max: 35m 38s 🟥 NVHPC Pass: 0%/2 | Total: 9m 09s | Avg: 4m 34s | Max: 4m 37s 🟥 gpu 🟥 h100 Pass: 0%/3 | Total: 2m 06s | Avg: 0m 42s | Max: 2m 06s 🟥 rtx2080 Pass: 0%/39 | Total: 4h 32m | Avg: 6m 58s | Max: 35m 38s 🟥 rtxa6000 Pass: 0%/8 | Total: 4m 21s | Avg: 0m 32s | Max: 2m 15s 🟥 jobs 🟥 Build Pass: 0%/42 | Total: 4h 38m | Avg: 6m 37s | Max: 35m 38s 🟥 DeviceLaunch Pass: 0%/1 🟥 GraphCapture Pass: 0%/1 🟥 HostLaunch Pass: 0%/3 🟥 TestGPU Pass: 0%/3 🟥 sm 🟥 90 Pass: 0%/3 | Total: 2m 06s | Avg: 0m 42s | Max: 2m 06s 🟥 90;90a Pass: 0%/2 | Total: 30m 03s | Avg: 15m 01s | Max: 27m 51s 🟥 100;120 Pass: 0%/2 | Total: 32m 15s | Avg: 16m 07s | Max: 30m 02s 🟥 std 🟥 17 Pass: 0%/21 | Total: 2h 22m | Avg: 6m 47s | Max: 34m 25s 🟥 20 Pass: 0%/29 | Total: 2h 15m | Avg: 4m 40s | Max: 35m 38s -
🟥 thrust: Pass: 0%/50 | Total: 4h 56m | Avg: 5m 55s | Max: 37m 18s
🟥 cmake_options 🟥 -DTHRUST_DISPATCH_TYPE=Force32bit Pass: 0%/2 | Total: 2m 05s | Avg: 1m 02s | Max: 2m 05s 🟥 cpu 🟥 amd64 Pass: 0%/48 | Total: 4h 53m | Avg: 6m 06s | Max: 37m 18s 🟥 arm64 Pass: 0%/2 | Total: 3m 02s | Avg: 1m 31s | Max: 1m 31s 🟥 ctk 🟥 12.0 Pass: 0%/5 | Total: 44m 30s | Avg: 8m 54s | Max: 37m 18s 🟥 12.9 Pass: 0%/45 | Total: 4h 11m | Avg: 5m 35s | Max: 37m 16s 🟥 cudacxx 🟥 ClangCUDA19 Pass: 0%/2 | Total: 4m 03s | Avg: 2m 01s | Max: 2m 04s 🟥 nvcc12.0 Pass: 0%/5 | Total: 44m 30s | Avg: 8m 54s | Max: 37m 18s 🟥 nvcc12.9 Pass: 0%/43 | Total: 4h 07m | Avg: 5m 45s | Max: 37m 16s 🟥 cudacxx_family 🟥 ClangCUDA Pass: 0%/2 | Total: 4m 03s | Avg: 2m 01s | Max: 2m 04s 🟥 nvcc Pass: 0%/48 | Total: 4h 52m | Avg: 6m 05s | Max: 37m 18s 🟥 cxx 🟥 Clang14 Pass: 0%/4 | Total: 7m 43s | Avg: 1m 55s | Max: 1m 58s 🟥 Clang15 Pass: 0%/2 | Total: 4m 00s | Avg: 2m 00s | Max: 2m 01s 🟥 Clang16 Pass: 0%/2 | Total: 4m 07s | Avg: 2m 03s | Max: 2m 04s 🟥 Clang17 Pass: 0%/2 | Total: 4m 06s | Avg: 2m 03s | Max: 2m 04s 🟥 Clang18 Pass: 0%/2 | Total: 3m 49s | Avg: 1m 54s | Max: 1m 55s 🟥 Clang19 Pass: 0%/7 | Total: 9m 28s | Avg: 1m 21s | Max: 2m 04s 🟥 GCC7 Pass: 0%/2 | Total: 3m 36s | Avg: 1m 48s | Max: 1m 55s 🟥 GCC8 Pass: 0%/1 | Total: 1m 55s | Avg: 1m 55s | Max: 1m 55s 🟥 GCC9 Pass: 0%/2 | Total: 3m 39s | Avg: 1m 49s | Max: 1m 56s 🟥 GCC10 Pass: 0%/2 | Total: 3m 50s | Avg: 1m 55s | Max: 1m 57s 🟥 GCC11 Pass: 0%/2 | Total: 3m 39s | Avg: 1m 49s | Max: 1m 51s 🟥 GCC12 Pass: 0%/2 | Total: 3m 47s | Avg: 1m 53s | Max: 1m 56s 🟥 GCC13 Pass: 0%/11 | Total: 13m 33s | Avg: 1m 13s | Max: 2m 05s 🟥 MSVC14.29 Pass: 0%/2 | Total: 1h 14m | Avg: 37m 16s | Max: 37m 18s 🟥 MSVC14.43 Pass: 0%/5 | Total: 2h 25m | Avg: 29m 02s | Max: 37m 16s 🟥 NVHPC25.5 Pass: 0%/2 | Total: 9m 21s | Avg: 4m 40s | Max: 4m 42s 🟥 cxx_family 🟥 Clang Pass: 0%/19 | Total: 33m 13s | Avg: 1m 44s | Max: 2m 04s 🟥 GCC Pass: 0%/22 | Total: 33m 59s | Avg: 1m 32s | Max: 2m 05s 🟥 MSVC Pass: 0%/7 | Total: 3h 39m | Avg: 31m 23s | Max: 37m 18s 🟥 NVHPC Pass: 0%/2 | Total: 9m 21s | Avg: 4m 40s | Max: 4m 42s 🟥 gpu 🟥 h100 Pass: 0%/2 | Total: 1m 57s | Avg: 0m 58s | Max: 1m 57s 🟥 rtx2080 Pass: 0%/38 | Total: 4h 12m | Avg: 6m 38s | Max: 37m 18s 🟥 rtx4090 Pass: 0%/10 | Total: 41m 54s | Avg: 4m 11s | Max: 35m 52s 🟥 jobs 🟥 Build Pass: 0%/43 | Total: 4h 56m | Avg: 6m 53s | Max: 37m 18s 🟥 TestCPU Pass: 0%/3 🟥 TestGPU Pass: 0%/4 🟥 sm 🟥 90 Pass: 0%/2 | Total: 1m 57s | Avg: 0m 58s | Max: 1m 57s 🟥 90;90a Pass: 0%/2 | Total: 37m 57s | Avg: 18m 58s | Max: 36m 01s 🟥 100;120 Pass: 0%/2 | Total: 38m 04s | Avg: 19m 02s | Max: 36m 03s 🟥 std 🟥 17 Pass: 0%/21 | Total: 2h 29m | Avg: 7m 05s | Max: 37m 18s 🟥 20 Pass: 0%/27 | Total: 2h 25m | Avg: 5m 22s | Max: 36m 03s -
🟨 libcudacxx: Pass: 6%/48 | Total: 7h 40m | Avg: 9m 35s | Max: 35m 57s | Hits: 90%/42
🟨 jobs 🟥 Build Pass: 0%/42 | Total: 6h 51m | Avg: 9m 48s | Max: 35m 57s 🟩 NVRTC Pass: 100%/2 | Total: 47m 09s | Avg: 23m 34s | Max: 24m 47s | Hits: 90%/42 🟥 Test Pass: 0%/3 🟩 VerifyCodegen Pass: 100%/1 | Total: 1m 45s | Avg: 1m 45s | Max: 1m 45s 🟨 sm 🟩 75 Pass: 100%/2 | Total: 47m 09s | Avg: 23m 34s | Max: 24m 47s | Hits: 90%/42 🟥 90 Pass: 0%/2 | Total: 5m 20s | Avg: 2m 40s | Max: 5m 20s 🟥 90;90a Pass: 0%/2 | Total: 39m 05s | Avg: 19m 32s | Max: 33m 43s 🟥 100;120 Pass: 0%/2 | Total: 38m 58s | Avg: 19m 29s | Max: 32m 22s 🟨 cpu 🟨 amd64 Pass: 6%/46 | Total: 7h 31m | Avg: 9m 49s | Max: 35m 57s | Hits: 90%/42 🟥 arm64 Pass: 0%/2 | Total: 9m 02s | Avg: 4m 31s | Max: 6m 17s 🟨 ctk 🟥 12.0 Pass: 0%/5 | Total: 50m 35s | Avg: 10m 07s | Max: 32m 54s 🟨 12.9 Pass: 6%/43 | Total: 6h 50m | Avg: 9m 32s | Max: 35m 57s | Hits: 90%/42 🟨 cudacxx 🟥 ClangCUDA19 Pass: 0%/2 | Total: 4m 07s | Avg: 2m 03s | Max: 2m 06s 🟥 nvcc12.0 Pass: 0%/5 | Total: 50m 35s | Avg: 10m 07s | Max: 32m 54s 🟨 nvcc12.9 Pass: 7%/41 | Total: 6h 46m | Avg: 9m 54s | Max: 35m 57s | Hits: 90%/42 🟨 cudacxx_family 🟥 ClangCUDA Pass: 0%/2 | Total: 4m 07s | Avg: 2m 03s | Max: 2m 06s 🟨 nvcc Pass: 6%/46 | Total: 7h 36m | Avg: 9m 55s | Max: 35m 57s | Hits: 90%/42 🟨 cxx 🟥 Clang14 Pass: 0%/4 | Total: 12m 51s | Avg: 3m 12s | Max: 3m 26s 🟥 Clang15 Pass: 0%/2 | Total: 6m 57s | Avg: 3m 28s | Max: 3m 33s 🟥 Clang16 Pass: 0%/2 | Total: 6m 50s | Avg: 3m 25s | Max: 3m 26s 🟥 Clang17 Pass: 0%/2 | Total: 6m 34s | Avg: 3m 17s | Max: 3m 21s 🟥 Clang18 Pass: 0%/2 | Total: 6m 36s | Avg: 3m 18s | Max: 3m 22s 🟥 Clang19 Pass: 0%/6 | Total: 13m 25s | Avg: 2m 14s | Max: 3m 22s 🟥 GCC7 Pass: 0%/2 | Total: 11m 31s | Avg: 5m 45s | Max: 6m 18s 🟥 GCC8 Pass: 0%/1 | Total: 6m 52s | Avg: 6m 52s | Max: 6m 52s 🟥 GCC9 Pass: 0%/2 | Total: 12m 09s | Avg: 6m 04s | Max: 6m 47s 🟥 GCC10 Pass: 0%/2 | Total: 13m 57s | Avg: 6m 58s | Max: 7m 22s 🟥 GCC11 Pass: 0%/2 | Total: 12m 15s | Avg: 6m 07s | Max: 6m 54s 🟥 GCC12 Pass: 0%/2 | Total: 22m 08s | Avg: 11m 04s | Max: 12m 29s 🟨 GCC13 Pass: 27%/11 | Total: 1h 53m | Avg: 10m 16s | Max: 34m 40s | Hits: 90%/42 🟥 MSVC14.29 Pass: 0%/2 | Total: 1h 05m | Avg: 32m 42s | Max: 32m 54s 🟥 MSVC14.43 Pass: 0%/4 | Total: 2h 16m | Avg: 34m 01s | Max: 35m 57s 🟥 NVHPC25.5 Pass: 0%/2 | Total: 14m 05s | Avg: 7m 02s | Max: 7m 13s 🟨 cxx_family 🟥 Clang Pass: 0%/18 | Total: 53m 13s | Avg: 2m 57s | Max: 3m 33s 🟨 GCC Pass: 13%/22 | Total: 3h 11m | Avg: 8m 43s | Max: 34m 40s | Hits: 90%/42 🟥 MSVC Pass: 0%/6 | Total: 3h 21m | Avg: 33m 35s | Max: 35m 57s 🟥 NVHPC Pass: 0%/2 | Total: 14m 05s | Avg: 7m 02s | Max: 7m 13s 🟨 gpu 🟥 h100 Pass: 0%/2 | Total: 5m 20s | Avg: 2m 40s | Max: 5m 20s 🟨 rtx2080 Pass: 6%/46 | Total: 7h 35m | Avg: 9m 53s | Max: 35m 57s | Hits: 90%/42 🟨 std 🟨 17 Pass: 4%/22 | Total: 4h 07m | Avg: 11m 16s | Max: 35m 57s | Hits: 90%/21 🟨 20 Pass: 4%/25 | Total: 3h 31m | Avg: 8m 26s | Max: 34m 04s | Hits: 90%/21 -
🟥 cudax: Pass: 0%/28 | Total: 1h 54m | Avg: 4m 06s | Max: 13m 44s
🟥 cudacxx_family 🟥 nvcc Pass: 0%/28 | Total: 1h 54m | Avg: 4m 06s | Max: 13m 44s 🟥 cpu 🟥 amd64 Pass: 0%/24 | Total: 1h 45m | Avg: 4m 24s | Max: 13m 44s 🟥 arm64 Pass: 0%/4 | Total: 9m 01s | Avg: 2m 15s | Max: 2m 33s 🟥 ctk 🟥 12.0 Pass: 0%/3 | Total: 17m 52s | Avg: 5m 57s | Max: 12m 57s 🟥 12.9 Pass: 0%/25 | Total: 1h 37m | Avg: 3m 53s | Max: 13m 44s 🟥 cudacxx 🟥 nvcc12.0 Pass: 0%/3 | Total: 17m 52s | Avg: 5m 57s | Max: 12m 57s 🟥 nvcc12.9 Pass: 0%/25 | Total: 1h 37m | Avg: 3m 53s | Max: 13m 44s 🟥 cxx 🟥 Clang14 Pass: 0%/2 | Total: 5m 00s | Avg: 2m 30s | Max: 2m 41s 🟥 Clang15 Pass: 0%/1 | Total: 2m 42s | Avg: 2m 42s | Max: 2m 42s 🟥 Clang16 Pass: 0%/1 | Total: 2m 38s | Avg: 2m 38s | Max: 2m 38s 🟥 Clang17 Pass: 0%/1 | Total: 2m 43s | Avg: 2m 43s | Max: 2m 43s 🟥 Clang18 Pass: 0%/1 | Total: 2m 37s | Avg: 2m 37s | Max: 2m 37s 🟥 Clang19 Pass: 0%/4 | Total: 6m 37s | Avg: 1m 39s | Max: 2m 35s 🟥 GCC10 Pass: 0%/2 | Total: 5m 36s | Avg: 2m 48s | Max: 3m 00s 🟥 GCC11 Pass: 0%/1 | Total: 2m 45s | Avg: 2m 45s | Max: 2m 45s 🟥 GCC12 Pass: 0%/1 | Total: 3m 08s | Avg: 3m 08s | Max: 3m 08s 🟥 GCC13 Pass: 0%/8 | Total: 16m 28s | Avg: 2m 03s | Max: 3m 02s 🟥 MSVC14.39 Pass: 0%/1 | Total: 12m 57s | Avg: 12m 57s | Max: 12m 57s 🟥 MSVC14.43 Pass: 0%/3 | Total: 40m 18s | Avg: 13m 26s | Max: 13m 44s 🟥 NVHPC25.5 Pass: 0%/2 | Total: 11m 30s | Avg: 5m 45s | Max: 5m 58s 🟥 cxx_family 🟥 Clang Pass: 0%/10 | Total: 22m 17s | Avg: 2m 13s | Max: 2m 43s 🟥 GCC Pass: 0%/12 | Total: 27m 57s | Avg: 2m 19s | Max: 3m 08s 🟥 MSVC Pass: 0%/4 | Total: 53m 15s | Avg: 13m 18s | Max: 13m 44s 🟥 NVHPC Pass: 0%/2 | Total: 11m 30s | Avg: 5m 45s | Max: 5m 58s 🟥 gpu 🟥 h100 Pass: 0%/2 | Total: 2m 36s | Avg: 1m 18s | Max: 2m 36s 🟥 rtx2080 Pass: 0%/26 | Total: 1h 52m | Avg: 4m 19s | Max: 13m 44s 🟥 jobs 🟥 Build Pass: 0%/25 | Total: 1h 54m | Avg: 4m 35s | Max: 13m 44s 🟥 Test Pass: 0%/3 🟥 sm 🟥 90 Pass: 0%/2 | Total: 2m 36s | Avg: 1m 18s | Max: 2m 36s 🟥 90;90a Pass: 0%/2 | Total: 15m 52s | Avg: 7m 56s | Max: 12m 53s 🟥 100;120 Pass: 0%/2 | Total: 16m 33s | Avg: 8m 16s | Max: 13m 41s 🟥 std 🟥 17 Pass: 0%/3 | Total: 10m 24s | Avg: 3m 28s | Max: 5m 58s 🟥 20 Pass: 0%/25 | Total: 1h 44m | Avg: 4m 11s | Max: 13m 44s -
🟥 python: Pass: 0%/22 | Total: 19m 51s | Avg: 0m 54s | Max: 10m 04s
🟥 cpu 🟥 amd64 Pass: 0%/22 | Total: 19m 51s | Avg: 0m 54s | Max: 10m 04s 🟥 cudacxx_family 🟥 nvcc Pass: 0%/22 | Total: 19m 51s | Avg: 0m 54s | Max: 10m 04s 🟥 cxx 🟥 GCC13 Pass: 0%/22 | Total: 19m 51s | Avg: 0m 54s | Max: 10m 04s 🟥 cxx_family 🟥 GCC Pass: 0%/22 | Total: 19m 51s | Avg: 0m 54s | Max: 10m 04s 🟥 ctk 🟥 12.5 Pass: 0%/6 🟥 12.8 Pass: 0%/2 🟥 12.9 Pass: 0%/14 | Total: 19m 51s | Avg: 1m 25s | Max: 10m 04s 🟥 cudacxx 🟥 nvcc12.5 Pass: 0%/6 🟥 nvcc12.8 Pass: 0%/2 🟥 nvcc12.9 Pass: 0%/14 | Total: 19m 51s | Avg: 1m 25s | Max: 10m 04s 🟥 gpu 🟥 h100 Pass: 0%/4 🟥 l4 Pass: 0%/18 | Total: 19m 51s | Avg: 1m 06s | Max: 10m 04s 🟥 jobs 🟥 Build cuda.cccl Pass: 0%/2 | Total: 19m 51s | Avg: 9m 55s | Max: 10m 04s 🟥 Test cuda.cccl.cooperative Pass: 0%/5 🟥 Test cuda.cccl.examples Pass: 0%/5 🟥 Test cuda.cccl.headers Pass: 0%/5 🟥 Test cuda.cccl.parallel Pass: 0%/5 🟥 py_version 🟥 3.10 Pass: 0%/9 | Total: 10m 04s | Avg: 1m 07s | Max: 10m 04s 🟥 3.13 Pass: 0%/13 | Total: 9m 47s | Avg: 0m 45s | Max: 9m 47s -
🟥 cccl_c_parallel: Pass: 0%/4 | Total: 2m 27s | Avg: 0m 36s | Max: 2m 27s
🟥 cpu 🟥 amd64 Pass: 0%/4 | Total: 2m 27s | Avg: 0m 36s | Max: 2m 27s 🟥 ctk 🟥 12.9 Pass: 0%/4 | Total: 2m 27s | Avg: 0m 36s | Max: 2m 27s 🟥 cudacxx 🟥 nvcc12.9 Pass: 0%/4 | Total: 2m 27s | Avg: 0m 36s | Max: 2m 27s 🟥 cudacxx_family 🟥 nvcc Pass: 0%/4 | Total: 2m 27s | Avg: 0m 36s | Max: 2m 27s 🟥 cxx 🟥 GCC13 Pass: 0%/4 | Total: 2m 27s | Avg: 0m 36s | Max: 2m 27s 🟥 cxx_family 🟥 GCC Pass: 0%/4 | Total: 2m 27s | Avg: 0m 36s | Max: 2m 27s 🟥 gpu 🟥 h100 Pass: 0%/1 🟥 l4 Pass: 0%/1 🟥 rtx2080 Pass: 0%/2 | Total: 2m 27s | Avg: 1m 13s | Max: 2m 27s 🟥 jobs 🟥 Build Pass: 0%/1 | Total: 2m 27s | Avg: 2m 27s | Max: 2m 27s 🟥 Test Pass: 0%/3 -
🟩 packaging: Pass: 100%/4 | Total: 18m 18s | Avg: 4m 34s | Max: 5m 28s
🟩 cpu 🟩 amd64 Pass: 100%/4 | Total: 18m 18s | Avg: 4m 34s | Max: 5m 28s 🟩 ctk 🟩 12.0 Pass: 100%/2 | Total: 8m 58s | Avg: 4m 29s | Max: 5m 12s 🟩 12.9 Pass: 100%/2 | Total: 9m 20s | Avg: 4m 40s | Max: 5m 28s 🟩 cudacxx 🟩 nvcc12.0 Pass: 100%/2 | Total: 8m 58s | Avg: 4m 29s | Max: 5m 12s 🟩 nvcc12.9 Pass: 100%/2 | Total: 9m 20s | Avg: 4m 40s | Max: 5m 28s 🟩 cudacxx_family 🟩 nvcc Pass: 100%/4 | Total: 18m 18s | Avg: 4m 34s | Max: 5m 28s 🟩 cxx 🟩 Clang14 Pass: 100%/1 | Total: 3m 46s | Avg: 3m 46s | Max: 3m 46s 🟩 Clang19 Pass: 100%/1 | Total: 3m 52s | Avg: 3m 52s | Max: 3m 52s 🟩 GCC12 Pass: 100%/1 | Total: 5m 12s | Avg: 5m 12s | Max: 5m 12s 🟩 GCC13 Pass: 100%/1 | Total: 5m 28s | Avg: 5m 28s | Max: 5m 28s 🟩 cxx_family 🟩 Clang Pass: 100%/2 | Total: 7m 38s | Avg: 3m 49s | Max: 3m 52s 🟩 GCC Pass: 100%/2 | Total: 10m 40s | Avg: 5m 20s | Max: 5m 28s 🟩 gpu 🟩 rtx2080 Pass: 100%/4 | Total: 18m 18s | Avg: 4m 34s | Max: 5m 28s 🟩 jobs 🟩 Test Pass: 100%/4 | Total: 18m 18s | Avg: 4m 34s | Max: 5m 28s -
🟩 stdpar: Pass: 100%/4 | Total: 15m 56s | Avg: 3m 59s | Max: 4m 07s
🟩 cpu 🟩 amd64 Pass: 100%/2 | Total: 8m 09s | Avg: 4m 04s | Max: 4m 07s 🟩 arm64 Pass: 100%/2 | Total: 7m 47s | Avg: 3m 53s | Max: 3m 57s 🟩 ctk 🟩 12.9 Pass: 100%/4 | Total: 15m 56s | Avg: 3m 59s | Max: 4m 07s 🟩 cudacxx 🟩 nvcc12.9 Pass: 100%/4 | Total: 15m 56s | Avg: 3m 59s | Max: 4m 07s 🟩 cudacxx_family 🟩 nvcc Pass: 100%/4 | Total: 15m 56s | Avg: 3m 59s | Max: 4m 07s 🟩 cxx 🟩 NVHPC25.5 Pass: 100%/4 | Total: 15m 56s | Avg: 3m 59s | Max: 4m 07s 🟩 cxx_family 🟩 NVHPC Pass: 100%/4 | Total: 15m 56s | Avg: 3m 59s | Max: 4m 07s 🟩 gpu 🟩 rtx2080 Pass: 100%/4 | Total: 15m 56s | Avg: 3m 59s | Max: 4m 07s 🟩 jobs 🟩 Build Pass: 100%/4 | Total: 15m 56s | Avg: 3m 59s | Max: 4m 07s 🟩 std 🟩 17 Pass: 100%/2 | Total: 7m 52s | Avg: 3m 56s | Max: 4m 02s 🟩 20 Pass: 100%/2 | Total: 8m 04s | Avg: 4m 02s | Max: 4m 07s
👃 Inspect Changes
Modifications in project?
| Project | |
|---|---|
| CCCL Infrastructure | |
| CCCL Packaging | |
| +/- | libcu++ |
| +/- | CUB |
| +/- | Thrust |
| CUDA Experimental | |
| stdpar | |
| python | |
| CCCL C Parallel Library | |
| +/- | Catch2Helper |
Modifications in project or dependencies?
| Project | |
|---|---|
| CCCL Infrastructure | |
| +/- | CCCL Packaging |
| +/- | libcu++ |
| +/- | CUB |
| +/- | Thrust |
| +/- | CUDA Experimental |
| +/- | stdpar |
| +/- | python |
| +/- | CCCL C Parallel Library |
| +/- | Catch2Helper |
🏃 Runner counts (total jobs: 210)
| # | Runner |
|---|---|
| 128 | linux-amd64-cpu16 |
| 23 | windows-amd64-cpu16 |
| 17 | linux-amd64-gpu-l4-latest-1 |
| 12 | linux-arm64-cpu16 |
| 11 | linux-amd64-gpu-rtx2080-latest-1 |
| 10 | linux-amd64-gpu-h100-latest-1 |
| 6 | linux-amd64-gpu-rtxa6000-latest-1 |
| 3 | linux-amd64-gpu-rtx4090-latest-1 |
😬 CI Workflow Results
🟥 Finished in 8h 50m: Pass: 78%/236 | Total: 7d 12h | Max: 6h 00m | Hits: 70%/281569
See results here.
😬 CI Workflow Results
🟥 Finished in 3h 39m: Pass: 80%/240 | Total: 7d 07h | Max: 3h 38m | Hits: 67%/209876
See results here.
😬 CI Workflow Results
🟥 Finished in 6h 05m: Pass: 92%/240 | Total: 7d 03h | Max: 3h 29m | Hits: 63%/288146
See results here.
😬 CI Workflow Results
🟥 Finished in 6h 01m: Pass: 1%/171 | Total: 2d 15h | Max: 6h 00m | Hits: 97%/7895
See results here.
😬 CI Workflow Results
🟥 Finished in 6h 01m: Pass: 1%/171 | Total: 2d 14h | Max: 6h 00m | Hits: 97%/7821
See results here.
😬 CI Workflow Results
🟥 Finished in 1h 51m: Pass: 43%/176 | Total: 2d 18h | Max: 1h 51m | Hits: 94%/199680
See results here.
😬 CI Workflow Results
🟥 Finished in 1h 46m: Pass: 43%/176 | Total: 2d 11h | Max: 1h 46m | Hits: 93%/200210
See results here.
😬 CI Workflow Results
🟥 Finished in 4h 51m: Pass: 71%/176 | Total: 7d 06h | Max: 4h 49m | Hits: 78%/268455
See results here.
I reviewed the SASS diff of this PR for the benchmark cub.bench.run_length_encode.encode.base on sm_86 and sm_120 and here is what I found out:
- there are a LOT of small difference
- most changes are completely benign, just instructions appearing in different places
- registers are sometimes differently allocated, leading to large diffs, but should not have an effect on performance
- sm_86 shows almost no changes, most changes happen for sm_120
- the kernel
batch_memcpy::MultiBlockBatchMemcpyKernelwithi128on sm120 has the largest changes and it looks like the compiler made different unrolling decisions (it previously unrolled and does not do that now, the new kernel is half the size) - the kernel
DeviceReduceByKeyKernelis affected the most, since it uses RLE internally.
In summary the changes don't look problematic to me. I could not see any worrying codegen. So I guess we should do a proper benchmark on the cluster. I tried doing one locally on a RTX 5090, but it was too shaky. We should definitely pick a high SM version, since the compiler shows the largest SASS impact there.
cub.bench.run_length_encode.encode.base on B200:
## [0] NVIDIA B200
| T{ct} | OffsetT{ct} | RunLengthT{ct} | Elements{io} | MaxSegSize | Ref Time | Ref Noise | Cmp Time | Cmp Noise | Diff | %Diff | Status |
|---------|---------------|------------------|----------------|--------------|------------|-------------|------------|-------------|-------------|---------|----------|
| I8 | I32 | I32 | 2^16 | 2^1 | 11.563 us | 6.18% | 11.552 us | 6.25% | -0.011 us | -0.10% | SAME |
| I8 | I32 | I32 | 2^20 | 2^1 | 13.735 us | 6.15% | 13.909 us | 6.87% | 0.174 us | 1.27% | SAME |
| I8 | I32 | I32 | 2^24 | 2^1 | 70.537 us | 1.21% | 63.205 us | 1.75% | -7.332 us | -10.39% | FAST |
| I8 | I32 | I32 | 2^28 | 2^1 | 960.569 us | 0.20% | 840.719 us | 0.20% | -119.850 us | -12.48% | FAST |
| I8 | I32 | I32 | 2^16 | 2^4 | 11.363 us | 3.68% | 11.366 us | 3.18% | 0.004 us | 0.03% | SAME |
| I8 | I32 | I32 | 2^20 | 2^4 | 13.421 us | 3.41% | 13.478 us | 3.75% | 0.057 us | 0.42% | SAME |
| I8 | I32 | I32 | 2^24 | 2^4 | 55.936 us | 1.83% | 50.233 us | 1.61% | -5.703 us | -10.19% | FAST |
| I8 | I32 | I32 | 2^28 | 2^4 | 734.661 us | 0.26% | 631.863 us | 0.28% | -102.798 us | -13.99% | FAST |
| I8 | I32 | I32 | 2^16 | 2^8 | 11.431 us | 5.00% | 11.400 us | 3.95% | -0.031 us | -0.27% | SAME |
| I8 | I32 | I32 | 2^20 | 2^8 | 13.412 us | 3.15% | 13.544 us | 3.97% | 0.132 us | 0.98% | SAME |
| I8 | I32 | I32 | 2^24 | 2^8 | 54.474 us | 1.31% | 48.263 us | 1.04% | -6.211 us | -11.40% | FAST |
| I8 | I32 | I32 | 2^28 | 2^8 | 710.287 us | 0.27% | 605.339 us | 0.31% | -104.948 us | -14.78% | FAST |
| I8 | I32 | I64 | 2^16 | 2^1 | 14.090 us | 7.49% | 13.511 us | 4.29% | -0.579 us | -4.11% | SAME |
| I8 | I32 | I64 | 2^20 | 2^1 | 18.300 us | 6.59% | 17.961 us | 4.83% | -0.338 us | -1.85% | SAME |
| I8 | I32 | I64 | 2^24 | 2^1 | 110.951 us | 0.92% | 110.205 us | 0.89% | -0.746 us | -0.67% | SAME |
| I8 | I32 | I64 | 2^28 | 2^1 | 1.580 ms | 0.08% | 1.588 ms | 0.09% | 8.192 us | 0.52% | SLOW |
| I8 | I32 | I64 | 2^16 | 2^4 | 13.942 us | 7.32% | 13.372 us | 3.69% | -0.570 us | -4.09% | FAST |
| I8 | I32 | I64 | 2^20 | 2^4 | 17.897 us | 5.88% | 17.465 us | 3.57% | -0.432 us | -2.41% | SAME |
| I8 | I32 | I64 | 2^24 | 2^4 | 93.895 us | 1.52% | 93.119 us | 1.05% | -0.776 us | -0.83% | SAME |
| I8 | I32 | I64 | 2^28 | 2^4 | 1.304 ms | 0.16% | 1.311 ms | 0.17% | 7.538 us | 0.58% | SLOW |
| I8 | I32 | I64 | 2^16 | 2^8 | 13.917 us | 6.83% | 13.360 us | 3.99% | -0.557 us | -4.00% | FAST |
| I8 | I32 | I64 | 2^20 | 2^8 | 17.810 us | 7.34% | 17.378 us | 4.75% | -0.432 us | -2.43% | SAME |
| I8 | I32 | I64 | 2^24 | 2^8 | 91.669 us | 1.60% | 91.141 us | 1.18% | -0.528 us | -0.58% | SAME |
| I8 | I32 | I64 | 2^28 | 2^8 | 1.267 ms | 0.18% | 1.275 ms | 0.18% | 7.476 us | 0.59% | SLOW |
| I8 | I64 | I32 | 2^16 | 2^1 | 11.729 us | 7.27% | 11.461 us | 4.37% | -0.268 us | -2.29% | SAME |
| I8 | I64 | I32 | 2^20 | 2^1 | 14.312 us | 10.88% | 13.740 us | 6.33% | -0.572 us | -3.99% | SAME |
| I8 | I64 | I32 | 2^24 | 2^1 | 69.735 us | 1.51% | 68.701 us | 0.95% | -1.034 us | -1.48% | FAST |
| I8 | I64 | I32 | 2^28 | 2^1 | 936.946 us | 0.18% | 937.209 us | 0.18% | 0.263 us | 0.03% | SAME |
| I8 | I64 | I32 | 2^16 | 2^4 | 11.683 us | 6.99% | 11.418 us | 3.78% | -0.265 us | -2.27% | SAME |
| I8 | I64 | I32 | 2^20 | 2^4 | 14.036 us | 6.96% | 13.484 us | 3.55% | -0.551 us | -3.93% | FAST |
| I8 | I64 | I32 | 2^24 | 2^4 | 55.488 us | 2.04% | 54.425 us | 1.36% | -1.063 us | -1.92% | FAST |
| I8 | I64 | I32 | 2^28 | 2^4 | 712.085 us | 0.21% | 711.024 us | 0.27% | -1.061 us | -0.15% | SAME |
| I8 | I64 | I32 | 2^16 | 2^8 | 11.661 us | 6.92% | 11.391 us | 3.69% | -0.270 us | -2.32% | SAME |
| I8 | I64 | I32 | 2^20 | 2^8 | 13.946 us | 6.79% | 13.416 us | 3.20% | -0.529 us | -3.80% | FAST |
| I8 | I64 | I32 | 2^24 | 2^8 | 53.260 us | 2.28% | 52.177 us | 1.64% | -1.082 us | -2.03% | FAST |
| I8 | I64 | I32 | 2^28 | 2^8 | 673.695 us | 0.29% | 673.708 us | 0.28% | 0.013 us | 0.00% | SAME |
| I8 | I64 | I64 | 2^16 | 2^1 | 13.843 us | 8.91% | 13.300 us | 6.62% | -0.544 us | -3.93% | SAME |
| I8 | I64 | I64 | 2^20 | 2^1 | 18.136 us | 5.89% | 17.695 us | 3.88% | -0.441 us | -2.43% | SAME |
| I8 | I64 | I64 | 2^24 | 2^1 | 114.873 us | 0.99% | 113.095 us | 1.06% | -1.779 us | -1.55% | FAST |
| I8 | I64 | I64 | 2^28 | 2^1 | 1.647 ms | 0.09% | 1.637 ms | 0.09% | -9.749 us | -0.59% | FAST |
| I8 | I64 | I64 | 2^16 | 2^4 | 13.610 us | 9.17% | 13.046 us | 7.00% | -0.564 us | -4.14% | SAME |
| I8 | I64 | I64 | 2^20 | 2^4 | 17.886 us | 6.44% | 17.275 us | 5.15% | -0.611 us | -3.42% | SAME |
| I8 | I64 | I64 | 2^24 | 2^4 | 97.437 us | 1.57% | 95.733 us | 1.03% | -1.703 us | -1.75% | FAST |
| I8 | I64 | I64 | 2^28 | 2^4 | 1.367 ms | 0.17% | 1.359 ms | 0.18% | -7.938 us | -0.58% | FAST |
| I8 | I64 | I64 | 2^16 | 2^8 | 13.768 us | 9.40% | 13.026 us | 7.24% | -0.742 us | -5.39% | SAME |
| I8 | I64 | I64 | 2^20 | 2^8 | 17.503 us | 8.36% | 16.905 us | 6.96% | -0.598 us | -3.42% | SAME |
| I8 | I64 | I64 | 2^24 | 2^8 | 94.823 us | 1.35% | 93.468 us | 1.18% | -1.355 us | -1.43% | FAST |
| I8 | I64 | I64 | 2^28 | 2^8 | 1.324 ms | 0.19% | 1.316 ms | 0.19% | -7.620 us | -0.58% | FAST |
| I16 | I32 | I32 | 2^16 | 2^1 | 11.499 us | 5.74% | 11.322 us | 2.63% | -0.177 us | -1.54% | SAME |
| I16 | I32 | I32 | 2^20 | 2^1 | 16.022 us | 5.96% | 15.461 us | 2.62% | -0.561 us | -3.50% | FAST |
| I16 | I32 | I32 | 2^24 | 2^1 | 69.291 us | 2.06% | 68.242 us | 1.65% | -1.049 us | -1.51% | SAME |
| I16 | I32 | I32 | 2^28 | 2^1 | 915.073 us | 0.20% | 914.271 us | 0.23% | -0.802 us | -0.09% | SAME |
| I16 | I32 | I32 | 2^16 | 2^4 | 11.650 us | 6.86% | 11.348 us | 3.14% | -0.303 us | -2.60% | SAME |
| I16 | I32 | I32 | 2^20 | 2^4 | 13.992 us | 8.02% | 13.632 us | 5.20% | -0.360 us | -2.57% | SAME |
| I16 | I32 | I32 | 2^24 | 2^4 | 56.103 us | 2.56% | 55.119 us | 2.02% | -0.984 us | -1.75% | SAME |
| I16 | I32 | I32 | 2^28 | 2^4 | 712.815 us | 0.23% | 712.320 us | 0.27% | -0.495 us | -0.07% | SAME |
| I16 | I32 | I32 | 2^16 | 2^8 | 11.528 us | 5.86% | 11.351 us | 3.02% | -0.178 us | -1.54% | SAME |
| I16 | I32 | I32 | 2^20 | 2^8 | 14.066 us | 7.37% | 13.574 us | 4.75% | -0.491 us | -3.49% | SAME |
| I16 | I32 | I32 | 2^24 | 2^8 | 54.878 us | 2.47% | 53.793 us | 2.05% | -1.085 us | -1.98% | SAME |
| I16 | I32 | I32 | 2^28 | 2^8 | 686.441 us | 0.26% | 686.144 us | 0.26% | -0.297 us | -0.04% | SAME |
| I16 | I32 | I64 | 2^16 | 2^1 | 13.684 us | 6.18% | 13.398 us | 4.73% | -0.286 us | -2.09% | SAME |
| I16 | I32 | I64 | 2^20 | 2^1 | 18.372 us | 6.94% | 17.847 us | 4.48% | -0.525 us | -2.86% | SAME |
| I16 | I32 | I64 | 2^24 | 2^1 | 113.095 us | 1.01% | 110.414 us | 1.02% | -2.681 us | -2.37% | FAST |
| I16 | I32 | I64 | 2^28 | 2^1 | 1.613 ms | 0.09% | 1.587 ms | 0.10% | -25.627 us | -1.59% | FAST |
| I16 | I32 | I64 | 2^16 | 2^4 | 13.838 us | 8.04% | 13.204 us | 5.32% | -0.634 us | -4.58% | SAME |
| I16 | I32 | I64 | 2^20 | 2^4 | 17.416 us | 4.07% | 17.203 us | 5.23% | -0.213 us | -1.22% | SAME |
| I16 | I32 | I64 | 2^24 | 2^4 | 96.225 us | 1.26% | 94.120 us | 1.35% | -2.105 us | -2.19% | FAST |
| I16 | I32 | I64 | 2^28 | 2^4 | 1.354 ms | 0.19% | 1.321 ms | 0.22% | -33.534 us | -2.48% | FAST |
| I16 | I32 | I64 | 2^16 | 2^8 | 13.159 us | 6.31% | 13.117 us | 6.29% | -0.042 us | -0.32% | SAME |
| I16 | I32 | I64 | 2^20 | 2^8 | 17.423 us | 3.95% | 17.250 us | 5.31% | -0.173 us | -0.99% | SAME |
| I16 | I32 | I64 | 2^24 | 2^8 | 93.660 us | 1.19% | 91.529 us | 1.28% | -2.131 us | -2.28% | FAST |
| I16 | I32 | I64 | 2^28 | 2^8 | 1.309 ms | 0.19% | 1.275 ms | 0.23% | -34.319 us | -2.62% | FAST |
| I16 | I64 | I32 | 2^16 | 2^1 | 11.405 us | 3.89% | 11.411 us | 4.07% | 0.005 us | 0.05% | SAME |
| I16 | I64 | I32 | 2^20 | 2^1 | 15.401 us | 3.65% | 15.447 us | 4.07% | 0.046 us | 0.30% | SAME |
| I16 | I64 | I32 | 2^24 | 2^1 | 74.955 us | 1.09% | 74.714 us | 1.15% | -0.241 us | -0.32% | SAME |
| I16 | I64 | I32 | 2^28 | 2^1 | 1.040 ms | 0.23% | 1.035 ms | 0.24% | -4.622 us | -0.44% | FAST |
| I16 | I64 | I32 | 2^16 | 2^4 | 11.519 us | 5.96% | 11.434 us | 4.25% | -0.084 us | -0.73% | SAME |
| I16 | I64 | I32 | 2^20 | 2^4 | 13.651 us | 6.07% | 13.606 us | 4.92% | -0.045 us | -0.33% | SAME |
| I16 | I64 | I32 | 2^24 | 2^4 | 61.315 us | 1.90% | 61.227 us | 1.81% | -0.087 us | -0.14% | SAME |
| I16 | I64 | I32 | 2^28 | 2^4 | 821.175 us | 0.23% | 821.314 us | 0.26% | 0.140 us | 0.02% | SAME |
| I16 | I64 | I32 | 2^16 | 2^8 | 11.832 us | 7.74% | 11.410 us | 4.15% | -0.422 us | -3.57% | SAME |
| I16 | I64 | I32 | 2^20 | 2^8 | 13.853 us | 7.48% | 13.408 us | 2.88% | -0.445 us | -3.21% | FAST |
| I16 | I64 | I32 | 2^24 | 2^8 | 59.830 us | 2.30% | 59.031 us | 1.80% | -0.799 us | -1.33% | SAME |
| I16 | I64 | I32 | 2^28 | 2^8 | 780.664 us | 0.24% | 783.496 us | 0.26% | 2.832 us | 0.36% | SLOW |
| I16 | I64 | I64 | 2^16 | 2^1 | 13.535 us | 9.56% | 12.951 us | 7.85% | -0.584 us | -4.32% | SAME |
| I16 | I64 | I64 | 2^20 | 2^1 | 18.061 us | 5.34% | 17.664 us | 3.46% | -0.397 us | -2.20% | SAME |
| I16 | I64 | I64 | 2^24 | 2^1 | 115.032 us | 1.02% | 114.100 us | 0.77% | -0.932 us | -0.81% | FAST |
| I16 | I64 | I64 | 2^28 | 2^1 | 1.649 ms | 0.11% | 1.655 ms | 0.12% | 5.183 us | 0.31% | SLOW |
| I16 | I64 | I64 | 2^16 | 2^4 | 12.590 us | 11.02% | 12.195 us | 9.09% | -0.394 us | -3.13% | SAME |
| I16 | I64 | I64 | 2^20 | 2^4 | 17.483 us | 7.98% | 17.142 us | 5.58% | -0.341 us | -1.95% | SAME |
| I16 | I64 | I64 | 2^24 | 2^4 | 98.317 us | 1.48% | 97.739 us | 1.14% | -0.579 us | -0.59% | SAME |
| I16 | I64 | I64 | 2^28 | 2^4 | 1.380 ms | 0.22% | 1.388 ms | 0.22% | 7.610 us | 0.55% | SLOW |
| I16 | I64 | I64 | 2^16 | 2^8 | 12.799 us | 11.01% | 12.181 us | 8.91% | -0.618 us | -4.83% | SAME |
| I16 | I64 | I64 | 2^20 | 2^8 | 17.210 us | 8.35% | 16.824 us | 6.71% | -0.386 us | -2.25% | SAME |
| I16 | I64 | I64 | 2^24 | 2^8 | 95.180 us | 1.50% | 94.581 us | 1.43% | -0.600 us | -0.63% | SAME |
| I16 | I64 | I64 | 2^28 | 2^8 | 1.325 ms | 0.23% | 1.333 ms | 0.24% | 8.243 us | 0.62% | SLOW |
| I32 | I32 | I32 | 2^16 | 2^1 | 11.518 us | 5.80% | 11.310 us | 2.53% | -0.208 us | -1.81% | SAME |
| I32 | I32 | I32 | 2^20 | 2^1 | 16.012 us | 5.93% | 15.486 us | 3.10% | -0.526 us | -3.28% | FAST |
| I32 | I32 | I32 | 2^24 | 2^1 | 67.852 us | 1.80% | 66.778 us | 1.02% | -1.075 us | -1.58% | FAST |
| I32 | I32 | I32 | 2^28 | 2^1 | 893.816 us | 0.23% | 893.440 us | 0.22% | -0.376 us | -0.04% | SAME |
| I32 | I32 | I32 | 2^16 | 2^4 | 11.645 us | 6.85% | 11.376 us | 3.42% | -0.269 us | -2.31% | SAME |
| I32 | I32 | I32 | 2^20 | 2^4 | 15.132 us | 8.73% | 14.949 us | 6.77% | -0.183 us | -1.21% | SAME |
| I32 | I32 | I32 | 2^24 | 2^4 | 55.815 us | 2.39% | 54.820 us | 1.77% | -0.996 us | -1.78% | FAST |
| I32 | I32 | I32 | 2^28 | 2^4 | 696.998 us | 0.21% | 696.333 us | 0.27% | -0.665 us | -0.10% | SAME |
| I32 | I32 | I32 | 2^16 | 2^8 | 11.854 us | 7.82% | 11.391 us | 3.88% | -0.463 us | -3.90% | FAST |
| I32 | I32 | I32 | 2^20 | 2^8 | 14.994 us | 10.17% | 14.732 us | 7.75% | -0.262 us | -1.75% | SAME |
| I32 | I32 | I32 | 2^24 | 2^8 | 54.843 us | 2.77% | 53.688 us | 2.08% | -1.155 us | -2.11% | FAST |
| I32 | I32 | I32 | 2^28 | 2^8 | 667.165 us | 0.28% | 667.357 us | 0.28% | 0.193 us | 0.03% | SAME |
| I32 | I32 | I64 | 2^16 | 2^1 | 14.055 us | 7.58% | 13.579 us | 4.79% | -0.476 us | -3.39% | SAME |
| I32 | I32 | I64 | 2^20 | 2^1 | 19.140 us | 8.02% | 18.902 us | 5.97% | -0.237 us | -1.24% | SAME |
| I32 | I32 | I64 | 2^24 | 2^1 | 114.935 us | 0.94% | 114.493 us | 1.01% | -0.442 us | -0.38% | SAME |
| I32 | I32 | I64 | 2^28 | 2^1 | 1.631 ms | 0.11% | 1.643 ms | 0.10% | 12.180 us | 0.75% | SLOW |
| I32 | I32 | I64 | 2^16 | 2^4 | 13.668 us | 5.95% | 13.405 us | 3.90% | -0.263 us | -1.92% | SAME |
| I32 | I32 | I64 | 2^20 | 2^4 | 18.115 us | 5.69% | 17.623 us | 3.58% | -0.492 us | -2.72% | SAME |
| I32 | I32 | I64 | 2^24 | 2^4 | 98.015 us | 1.66% | 97.622 us | 1.17% | -0.393 us | -0.40% | SAME |
| I32 | I32 | I64 | 2^28 | 2^4 | 1.350 ms | 0.25% | 1.365 ms | 0.25% | 15.179 us | 1.12% | SLOW |
| I32 | I32 | I64 | 2^16 | 2^8 | 13.932 us | 7.01% | 13.478 us | 3.72% | -0.454 us | -3.26% | SAME |
| I32 | I32 | I64 | 2^20 | 2^8 | 18.097 us | 5.55% | 17.693 us | 3.92% | -0.403 us | -2.23% | SAME |
| I32 | I32 | I64 | 2^24 | 2^8 | 95.006 us | 1.59% | 94.843 us | 1.27% | -0.163 us | -0.17% | SAME |
| I32 | I32 | I64 | 2^28 | 2^8 | 1.297 ms | 0.25% | 1.313 ms | 0.24% | 16.074 us | 1.24% | SLOW |
| I32 | I64 | I32 | 2^16 | 2^1 | 11.555 us | 6.07% | 11.427 us | 3.99% | -0.128 us | -1.11% | SAME |
| I32 | I64 | I32 | 2^20 | 2^1 | 15.873 us | 6.66% | 15.487 us | 3.95% | -0.386 us | -2.43% | SAME |
| I32 | I64 | I32 | 2^24 | 2^1 | 73.329 us | 1.86% | 72.549 us | 1.42% | -0.780 us | -1.06% | SAME |
| I32 | I64 | I32 | 2^28 | 2^1 | 990.122 us | 0.23% | 994.033 us | 0.23% | 3.911 us | 0.39% | SLOW |
| I32 | I64 | I32 | 2^16 | 2^4 | 11.597 us | 6.50% | 11.431 us | 4.22% | -0.166 us | -1.43% | SAME |
| I32 | I64 | I32 | 2^20 | 2^4 | 14.935 us | 8.96% | 14.675 us | 7.64% | -0.260 us | -1.74% | SAME |
| I32 | I64 | I32 | 2^24 | 2^4 | 60.266 us | 2.37% | 59.718 us | 1.96% | -0.548 us | -0.91% | SAME |
| I32 | I64 | I32 | 2^28 | 2^4 | 781.092 us | 0.22% | 784.682 us | 0.25% | 3.590 us | 0.46% | SLOW |
| I32 | I64 | I32 | 2^16 | 2^8 | 11.875 us | 7.80% | 11.359 us | 3.42% | -0.517 us | -4.35% | FAST |
| I32 | I64 | I32 | 2^20 | 2^8 | 15.094 us | 9.00% | 14.378 us | 7.43% | -0.715 us | -4.74% | SAME |
| I32 | I64 | I32 | 2^24 | 2^8 | 58.159 us | 2.57% | 57.465 us | 2.05% | -0.695 us | -1.19% | SAME |
| I32 | I64 | I32 | 2^28 | 2^8 | 743.407 us | 0.26% | 746.304 us | 0.28% | 2.897 us | 0.39% | SLOW |
| I32 | I64 | I64 | 2^16 | 2^1 | 13.669 us | 6.78% | 13.372 us | 4.45% | -0.297 us | -2.17% | SAME |
| I32 | I64 | I64 | 2^20 | 2^1 | 18.456 us | 7.77% | 18.428 us | 6.19% | -0.028 us | -0.15% | SAME |
| I32 | I64 | I64 | 2^24 | 2^1 | 118.827 us | 1.10% | 119.542 us | 0.96% | 0.715 us | 0.60% | SAME |
| I32 | I64 | I64 | 2^28 | 2^1 | 1.701 ms | 0.13% | 1.729 ms | 0.12% | 28.490 us | 1.68% | SLOW |
| I32 | I64 | I64 | 2^16 | 2^4 | 13.503 us | 7.59% | 13.290 us | 5.87% | -0.213 us | -1.58% | SAME |
| I32 | I64 | I64 | 2^20 | 2^4 | 18.016 us | 5.59% | 17.588 us | 3.28% | -0.428 us | -2.37% | SAME |
| I32 | I64 | I64 | 2^24 | 2^4 | 101.608 us | 1.44% | 102.405 us | 1.21% | 0.797 us | 0.78% | SAME |
| I32 | I64 | I64 | 2^28 | 2^4 | 1.419 ms | 0.25% | 1.448 ms | 0.22% | 29.574 us | 2.08% | SLOW |
| I32 | I64 | I64 | 2^16 | 2^8 | 13.864 us | 8.23% | 13.391 us | 5.85% | -0.473 us | -3.41% | SAME |
| I32 | I64 | I64 | 2^20 | 2^8 | 17.933 us | 6.93% | 17.539 us | 4.62% | -0.394 us | -2.20% | SAME |
| I32 | I64 | I64 | 2^24 | 2^8 | 98.340 us | 1.62% | 99.024 us | 1.36% | 0.685 us | 0.70% | SAME |
| I32 | I64 | I64 | 2^28 | 2^8 | 1.358 ms | 0.25% | 1.386 ms | 0.24% | 28.418 us | 2.09% | SLOW |
| I64 | I32 | I32 | 2^16 | 2^1 | 11.887 us | 7.86% | 11.344 us | 3.38% | -0.543 us | -4.57% | FAST |
| I64 | I32 | I32 | 2^20 | 2^1 | 17.941 us | 5.12% | 17.541 us | 2.78% | -0.400 us | -2.23% | SAME |
| I64 | I32 | I32 | 2^24 | 2^1 | 104.278 us | 1.48% | 103.211 us | 1.11% | -1.067 us | -1.02% | SAME |
| I64 | I32 | I32 | 2^28 | 2^1 | 1.459 ms | 0.19% | 1.458 ms | 0.20% | -0.119 us | -0.01% | SAME |
| I64 | I32 | I32 | 2^16 | 2^4 | 11.448 us | 5.12% | 11.435 us | 4.62% | -0.013 us | -0.11% | SAME |
| I64 | I32 | I32 | 2^20 | 2^4 | 16.583 us | 7.47% | 16.605 us | 6.76% | 0.022 us | 0.13% | SAME |
| I64 | I32 | I32 | 2^24 | 2^4 | 81.196 us | 1.46% | 80.787 us | 1.19% | -0.409 us | -0.50% | SAME |
| I64 | I32 | I32 | 2^28 | 2^4 | 1.091 ms | 0.17% | 1.091 ms | 0.17% | 0.006 us | 0.00% | SAME |
| I64 | I32 | I32 | 2^16 | 2^8 | 11.877 us | 7.84% | 11.393 us | 3.96% | -0.484 us | -4.08% | FAST |
| I64 | I32 | I32 | 2^20 | 2^8 | 16.170 us | 7.47% | 15.792 us | 5.16% | -0.378 us | -2.34% | SAME |
| I64 | I32 | I32 | 2^24 | 2^8 | 79.204 us | 2.02% | 78.020 us | 1.48% | -1.184 us | -1.49% | FAST |
| I64 | I32 | I32 | 2^28 | 2^8 | 1.042 ms | 0.15% | 1.041 ms | 0.18% | -0.346 us | -0.03% | SAME |
| I64 | I32 | I64 | 2^16 | 2^1 | 13.952 us | 8.09% | 13.433 us | 5.51% | -0.518 us | -3.71% | SAME |
| I64 | I32 | I64 | 2^20 | 2^1 | 20.127 us | 5.49% | 19.797 us | 3.50% | -0.330 us | -1.64% | SAME |
| I64 | I32 | I64 | 2^24 | 2^1 | 121.090 us | 1.14% | 120.097 us | 0.89% | -0.993 us | -0.82% | SAME |
| I64 | I32 | I64 | 2^28 | 2^1 | 1.717 ms | 0.17% | 1.717 ms | 0.16% | 0.002 us | 0.00% | SAME |
| I64 | I32 | I64 | 2^16 | 2^4 | 13.350 us | 8.26% | 12.851 us | 7.49% | -0.499 us | -3.74% | SAME |
| I64 | I32 | I64 | 2^20 | 2^4 | 19.494 us | 7.24% | 18.987 us | 5.60% | -0.507 us | -2.60% | SAME |
| I64 | I32 | I64 | 2^24 | 2^4 | 107.303 us | 1.63% | 106.213 us | 1.34% | -1.090 us | -1.02% | SAME |
| I64 | I32 | I64 | 2^28 | 2^4 | 1.479 ms | 0.28% | 1.478 ms | 0.28% | -0.156 us | -0.01% | SAME |
| I64 | I32 | I64 | 2^16 | 2^8 | 13.584 us | 10.14% | 13.025 us | 7.17% | -0.560 us | -4.12% | SAME |
| I64 | I32 | I64 | 2^20 | 2^8 | 19.152 us | 7.49% | 18.702 us | 6.17% | -0.450 us | -2.35% | SAME |
| I64 | I32 | I64 | 2^24 | 2^8 | 103.452 us | 1.48% | 102.512 us | 1.31% | -0.940 us | -0.91% | SAME |
| I64 | I32 | I64 | 2^28 | 2^8 | 1.410 ms | 0.24% | 1.410 ms | 0.25% | -0.144 us | -0.01% | SAME |
| I64 | I64 | I32 | 2^16 | 2^1 | 11.565 us | 6.23% | 11.380 us | 3.35% | -0.185 us | -1.60% | SAME |
| I64 | I64 | I32 | 2^20 | 2^1 | 18.047 us | 5.28% | 17.582 us | 3.25% | -0.465 us | -2.58% | SAME |
| I64 | I64 | I32 | 2^24 | 2^1 | 104.948 us | 1.34% | 103.788 us | 0.89% | -1.160 us | -1.11% | FAST |
| I64 | I64 | I32 | 2^28 | 2^1 | 1.472 ms | 0.17% | 1.472 ms | 0.19% | 0.059 us | 0.00% | SAME |
| I64 | I64 | I32 | 2^16 | 2^4 | 11.884 us | 7.79% | 11.447 us | 4.26% | -0.438 us | -3.68% | SAME |
| I64 | I64 | I32 | 2^20 | 2^4 | 16.651 us | 8.83% | 16.302 us | 7.04% | -0.350 us | -2.10% | SAME |
| I64 | I64 | I32 | 2^24 | 2^4 | 82.197 us | 1.44% | 81.301 us | 0.96% | -0.896 us | -1.09% | FAST |
| I64 | I64 | I32 | 2^28 | 2^4 | 1.103 ms | 0.16% | 1.103 ms | 0.15% | 0.007 us | 0.00% | SAME |
| I64 | I64 | I32 | 2^16 | 2^8 | 11.869 us | 7.83% | 11.395 us | 4.02% | -0.474 us | -3.99% | SAME |
| I64 | I64 | I32 | 2^20 | 2^8 | 16.178 us | 7.24% | 15.709 us | 4.98% | -0.469 us | -2.90% | SAME |
| I64 | I64 | I32 | 2^24 | 2^8 | 78.772 us | 1.94% | 77.683 us | 1.44% | -1.089 us | -1.38% | SAME |
| I64 | I64 | I32 | 2^28 | 2^8 | 1.039 ms | 0.16% | 1.039 ms | 0.18% | -0.532 us | -0.05% | SAME |
| I64 | I64 | I64 | 2^16 | 2^1 | 13.744 us | 9.68% | 13.189 us | 7.19% | -0.555 us | -4.04% | SAME |
| I64 | I64 | I64 | 2^20 | 2^1 | 20.105 us | 4.89% | 19.560 us | 2.35% | -0.545 us | -2.71% | FAST |
| I64 | I64 | I64 | 2^24 | 2^1 | 131.443 us | 1.14% | 130.470 us | 0.93% | -0.973 us | -0.74% | SAME |
| I64 | I64 | I64 | 2^28 | 2^1 | 1.894 ms | 0.20% | 1.893 ms | 0.19% | -0.482 us | -0.03% | SAME |
| I64 | I64 | I64 | 2^16 | 2^4 | 13.392 us | 10.84% | 12.808 us | 9.30% | -0.584 us | -4.36% | SAME |
| I64 | I64 | I64 | 2^20 | 2^4 | 18.751 us | 7.63% | 18.294 us | 5.85% | -0.457 us | -2.44% | SAME |
| I64 | I64 | I64 | 2^24 | 2^4 | 116.530 us | 1.49% | 115.544 us | 1.28% | -0.986 us | -0.85% | SAME |
| I64 | I64 | I64 | 2^28 | 2^4 | 1.641 ms | 0.24% | 1.641 ms | 0.27% | -0.129 us | -0.01% | SAME |
| I64 | I64 | I64 | 2^16 | 2^8 | 13.526 us | 10.11% | 12.821 us | 8.48% | -0.705 us | -5.21% | SAME |
| I64 | I64 | I64 | 2^20 | 2^8 | 18.348 us | 6.73% | 18.045 us | 5.12% | -0.303 us | -1.65% | SAME |
| I64 | I64 | I64 | 2^24 | 2^8 | 111.825 us | 1.53% | 110.584 us | 1.24% | -1.241 us | -1.11% | SAME |
| I64 | I64 | I64 | 2^28 | 2^8 | 1.556 ms | 0.27% | 1.556 ms | 0.28% | -0.492 us | -0.03% | SAME |
| I128 | I32 | I32 | 2^16 | 2^1 | 14.408 us | 9.37% | 13.844 us | 6.52% | -0.564 us | -3.91% | SAME |
| I128 | I32 | I32 | 2^20 | 2^1 | 23.757 us | 5.65% | 23.288 us | 3.65% | -0.469 us | -1.98% | SAME |
| I128 | I32 | I32 | 2^24 | 2^1 | 164.236 us | 0.78% | 163.093 us | 0.51% | -1.143 us | -0.70% | FAST |
| I128 | I32 | I32 | 2^28 | 2^1 | 2.402 ms | 0.07% | 2.402 ms | 0.08% | 0.101 us | 0.00% | SAME |
| I128 | I32 | I32 | 2^16 | 2^4 | 13.958 us | 7.13% | 13.543 us | 4.71% | -0.416 us | -2.98% | SAME |
| I128 | I32 | I32 | 2^20 | 2^4 | 20.545 us | 6.64% | 19.969 us | 4.68% | -0.576 us | -2.81% | SAME |
| I128 | I32 | I32 | 2^24 | 2^4 | 121.223 us | 1.05% | 120.246 us | 0.74% | -0.976 us | -0.81% | FAST |
| I128 | I32 | I32 | 2^28 | 2^4 | 1.707 ms | 0.10% | 1.707 ms | 0.11% | 0.046 us | 0.00% | SAME |
| I128 | I32 | I32 | 2^16 | 2^8 | 13.918 us | 7.16% | 13.495 us | 4.69% | -0.423 us | -3.04% | SAME |
| I128 | I32 | I32 | 2^20 | 2^8 | 20.172 us | 4.92% | 19.636 us | 2.76% | -0.536 us | -2.66% | SAME |
| I128 | I32 | I32 | 2^24 | 2^8 | 115.667 us | 1.43% | 114.470 us | 1.00% | -1.197 us | -1.03% | FAST |
| I128 | I32 | I32 | 2^28 | 2^8 | 1.611 ms | 0.10% | 1.611 ms | 0.11% | 0.088 us | 0.01% | SAME |
| I128 | I32 | I64 | 2^16 | 2^1 | 15.031 us | 9.79% | 14.562 us | 8.41% | -0.469 us | -3.12% | SAME |
| I128 | I32 | I64 | 2^20 | 2^1 | 30.460 us | 3.53% | 30.027 us | 2.11% | -0.433 us | -1.42% | SAME |
| I128 | I32 | I64 | 2^24 | 2^1 | 286.711 us | 0.44% | 285.607 us | 0.58% | -1.104 us | -0.38% | SAME |
| I128 | I32 | I64 | 2^28 | 2^1 | 4.363 ms | 0.10% | 4.362 ms | 0.10% | -1.279 us | -0.03% | SAME |
| I128 | I32 | I64 | 2^16 | 2^4 | 14.753 us | 9.84% | 14.210 us | 7.89% | -0.543 us | -3.68% | SAME |
| I128 | I32 | I64 | 2^20 | 2^4 | 28.489 us | 4.17% | 27.995 us | 2.71% | -0.495 us | -1.74% | SAME |
| I128 | I32 | I64 | 2^24 | 2^4 | 260.310 us | 0.69% | 259.336 us | 0.78% | -0.974 us | -0.37% | SAME |
| I128 | I32 | I64 | 2^28 | 2^4 | 3.929 ms | 0.16% | 3.929 ms | 0.16% | -0.365 us | -0.01% | SAME |
| I128 | I32 | I64 | 2^16 | 2^8 | 14.042 us | 7.48% | 14.077 us | 7.60% | 0.035 us | 0.25% | SAME |
| I128 | I32 | I64 | 2^20 | 2^8 | 27.879 us | 2.78% | 27.882 us | 2.61% | 0.003 us | 0.01% | SAME |
| I128 | I32 | I64 | 2^24 | 2^8 | 254.133 us | 0.71% | 253.946 us | 0.70% | -0.187 us | -0.07% | SAME |
| I128 | I32 | I64 | 2^28 | 2^8 | 3.845 ms | 0.15% | 3.845 ms | 0.15% | -0.238 us | -0.01% | SAME |
| I128 | I64 | I32 | 2^16 | 2^1 | 13.639 us | 5.65% | 13.794 us | 6.28% | 0.155 us | 1.14% | SAME |
| I128 | I64 | I32 | 2^20 | 2^1 | 23.414 us | 3.66% | 23.446 us | 3.41% | 0.032 us | 0.14% | SAME |
| I128 | I64 | I32 | 2^24 | 2^1 | 163.992 us | 0.68% | 164.175 us | 0.73% | 0.183 us | 0.11% | SAME |
| I128 | I64 | I32 | 2^28 | 2^1 | 2.420 ms | 0.08% | 2.419 ms | 0.08% | -0.040 us | -0.00% | SAME |
| I128 | I64 | I32 | 2^16 | 2^4 | 13.579 us | 4.60% | 13.551 us | 4.66% | -0.028 us | -0.21% | SAME |
| I128 | I64 | I32 | 2^20 | 2^4 | 19.794 us | 3.28% | 19.769 us | 3.17% | -0.025 us | -0.13% | SAME |
| I128 | I64 | I32 | 2^24 | 2^4 | 121.403 us | 0.93% | 121.337 us | 0.97% | -0.066 us | -0.05% | SAME |
| I128 | I64 | I32 | 2^28 | 2^4 | 1.725 ms | 0.11% | 1.725 ms | 0.11% | 0.191 us | 0.01% | SAME |
| I128 | I64 | I32 | 2^16 | 2^8 | 13.400 us | 4.00% | 13.470 us | 4.60% | 0.070 us | 0.52% | SAME |
| I128 | I64 | I32 | 2^20 | 2^8 | 19.529 us | 1.63% | 19.512 us | 1.43% | -0.018 us | -0.09% | SAME |
| I128 | I64 | I32 | 2^24 | 2^8 | 113.787 us | 0.72% | 113.811 us | 0.63% | 0.024 us | 0.02% | SAME |
| I128 | I64 | I32 | 2^28 | 2^8 | 1.602 ms | 0.10% | 1.602 ms | 0.11% | -0.134 us | -0.01% | SAME |
| I128 | I64 | I64 | 2^16 | 2^1 | 14.214 us | 8.00% | 14.143 us | 7.38% | -0.070 us | -0.50% | SAME |
| I128 | I64 | I64 | 2^20 | 2^1 | 29.963 us | 2.23% | 30.005 us | 2.52% | 0.042 us | 0.14% | SAME |
| I128 | I64 | I64 | 2^24 | 2^1 | 297.619 us | 0.62% | 297.654 us | 0.58% | 0.035 us | 0.01% | SAME |
| I128 | I64 | I64 | 2^28 | 2^1 | 4.571 ms | 0.12% | 4.571 ms | 0.11% | 0.266 us | 0.01% | SAME |
| I128 | I64 | I64 | 2^16 | 2^4 | 13.989 us | 7.24% | 13.920 us | 7.05% | -0.070 us | -0.50% | SAME |
| I128 | I64 | I64 | 2^20 | 2^4 | 28.373 us | 3.56% | 28.388 us | 3.56% | 0.015 us | 0.05% | SAME |
| I128 | I64 | I64 | 2^24 | 2^4 | 269.877 us | 0.70% | 270.098 us | 0.71% | 0.221 us | 0.08% | SAME |
| I128 | I64 | I64 | 2^28 | 2^4 | 4.126 ms | 0.15% | 4.127 ms | 0.16% | 0.939 us | 0.02% | SAME |
| I128 | I64 | I64 | 2^16 | 2^8 | 14.057 us | 7.67% | 13.898 us | 6.74% | -0.159 us | -1.13% | SAME |
| I128 | I64 | I64 | 2^20 | 2^8 | 28.059 us | 3.29% | 28.132 us | 3.46% | 0.073 us | 0.26% | SAME |
| I128 | I64 | I64 | 2^24 | 2^8 | 264.801 us | 0.75% | 264.808 us | 0.79% | 0.007 us | 0.00% | SAME |
| I128 | I64 | I64 | 2^28 | 2^8 | 4.030 ms | 0.16% | 4.030 ms | 0.15% | -0.294 us | -0.01% | SAME |
| F32 | I32 | I32 | 2^16 | 2^1 | 11.321 us | 2.77% | 11.352 us | 3.67% | 0.031 us | 0.27% | SAME |
| F32 | I32 | I32 | 2^20 | 2^1 | 15.437 us | 2.25% | 15.470 us | 2.59% | 0.033 us | 0.21% | SAME |
| F32 | I32 | I32 | 2^24 | 2^1 | 66.540 us | 1.27% | 66.529 us | 1.28% | -0.011 us | -0.02% | SAME |
| F32 | I32 | I32 | 2^28 | 2^1 | 713.251 us | 0.26% | 713.464 us | 0.27% | 0.212 us | 0.03% | SAME |
| F32 | I32 | I32 | 2^16 | 2^4 | 11.371 us | 3.52% | 11.374 us | 3.49% | 0.002 us | 0.02% | SAME |
| F32 | I32 | I32 | 2^20 | 2^4 | 14.557 us | 7.73% | 14.681 us | 7.56% | 0.124 us | 0.85% | SAME |
| F32 | I32 | I32 | 2^24 | 2^4 | 54.541 us | 1.36% | 54.530 us | 1.35% | -0.011 us | -0.02% | SAME |
| F32 | I32 | I32 | 2^28 | 2^4 | 690.972 us | 0.30% | 691.193 us | 0.28% | 0.222 us | 0.03% | SAME |
| F32 | I32 | I32 | 2^16 | 2^8 | 11.373 us | 3.31% | 11.347 us | 3.06% | -0.026 us | -0.23% | SAME |
| F32 | I32 | I32 | 2^20 | 2^8 | 14.708 us | 7.29% | 14.818 us | 7.33% | 0.110 us | 0.75% | SAME |
| F32 | I32 | I32 | 2^24 | 2^8 | 52.878 us | 1.87% | 53.093 us | 1.98% | 0.216 us | 0.41% | SAME |
| F32 | I32 | I32 | 2^28 | 2^8 | 662.018 us | 0.27% | 662.066 us | 0.28% | 0.048 us | 0.01% | SAME |
| F32 | I32 | I64 | 2^16 | 2^1 | 13.593 us | 5.01% | 13.626 us | 5.72% | 0.033 us | 0.24% | SAME |
| F32 | I32 | I64 | 2^20 | 2^1 | 19.187 us | 5.68% | 18.524 us | 6.29% | -0.663 us | -3.46% | SAME |
| F32 | I32 | I64 | 2^24 | 2^1 | 115.529 us | 0.92% | 112.191 us | 0.86% | -3.338 us | -2.89% | FAST |
| F32 | I32 | I64 | 2^28 | 2^1 | 1.399 ms | 0.19% | 1.353 ms | 0.24% | -46.090 us | -3.29% | FAST |
| F32 | I32 | I64 | 2^16 | 2^4 | 13.388 us | 3.04% | 13.450 us | 3.60% | 0.062 us | 0.46% | SAME |
| F32 | I32 | I64 | 2^20 | 2^4 | 17.581 us | 3.15% | 17.549 us | 2.93% | -0.031 us | -0.18% | SAME |
| F32 | I32 | I64 | 2^24 | 2^4 | 97.948 us | 1.22% | 95.279 us | 1.45% | -2.668 us | -2.72% | FAST |
| F32 | I32 | I64 | 2^28 | 2^4 | 1.365 ms | 0.23% | 1.319 ms | 0.28% | -45.545 us | -3.34% | FAST |
| F32 | I32 | I64 | 2^16 | 2^8 | 13.390 us | 4.86% | 13.453 us | 5.47% | 0.063 us | 0.47% | SAME |
| F32 | I32 | I64 | 2^20 | 2^8 | 17.663 us | 3.69% | 17.649 us | 3.58% | -0.014 us | -0.08% | SAME |
| F32 | I32 | I64 | 2^24 | 2^8 | 95.266 us | 1.28% | 92.389 us | 1.44% | -2.876 us | -3.02% | FAST |
| F32 | I32 | I64 | 2^28 | 2^8 | 1.323 ms | 0.23% | 1.276 ms | 0.27% | -47.480 us | -3.59% | FAST |
| F32 | I64 | I32 | 2^16 | 2^1 | 11.385 us | 3.58% | 11.392 us | 3.46% | 0.007 us | 0.06% | SAME |
| F32 | I64 | I32 | 2^20 | 2^1 | 15.201 us | 5.26% | 15.239 us | 4.81% | 0.038 us | 0.25% | SAME |
| F32 | I64 | I32 | 2^24 | 2^1 | 72.412 us | 1.46% | 72.495 us | 1.50% | 0.084 us | 0.12% | SAME |
| F32 | I64 | I32 | 2^28 | 2^1 | 809.182 us | 0.25% | 809.123 us | 0.25% | -0.059 us | -0.01% | SAME |
| F32 | I64 | I32 | 2^16 | 2^4 | 11.389 us | 3.50% | 11.428 us | 4.15% | 0.039 us | 0.34% | SAME |
| F32 | I64 | I32 | 2^20 | 2^4 | 14.957 us | 7.25% | 14.890 us | 7.73% | -0.067 us | -0.45% | SAME |
| F32 | I64 | I32 | 2^24 | 2^4 | 59.630 us | 2.03% | 59.596 us | 2.01% | -0.034 us | -0.06% | SAME |
| F32 | I64 | I32 | 2^28 | 2^4 | 784.961 us | 0.26% | 784.920 us | 0.26% | -0.041 us | -0.01% | SAME |
| F32 | I64 | I32 | 2^16 | 2^8 | 11.353 us | 3.31% | 11.376 us | 3.60% | 0.023 us | 0.20% | SAME |
| F32 | I64 | I32 | 2^20 | 2^8 | 14.522 us | 7.71% | 14.582 us | 7.42% | 0.059 us | 0.41% | SAME |
| F32 | I64 | I32 | 2^24 | 2^8 | 57.358 us | 1.91% | 57.422 us | 1.97% | 0.064 us | 0.11% | SAME |
| F32 | I64 | I32 | 2^28 | 2^8 | 744.046 us | 0.27% | 744.091 us | 0.27% | 0.045 us | 0.01% | SAME |
| F32 | I64 | I64 | 2^16 | 2^1 | 13.209 us | 6.24% | 13.471 us | 4.79% | 0.262 us | 1.98% | SAME |
| F32 | I64 | I64 | 2^20 | 2^1 | 17.991 us | 5.39% | 18.093 us | 5.71% | 0.102 us | 0.57% | SAME |
| F32 | I64 | I64 | 2^24 | 2^1 | 117.384 us | 0.90% | 117.643 us | 0.92% | 0.259 us | 0.22% | SAME |
| F32 | I64 | I64 | 2^28 | 2^1 | 1.439 ms | 0.22% | 1.441 ms | 0.23% | 2.391 us | 0.17% | SAME |
| F32 | I64 | I64 | 2^16 | 2^4 | 13.193 us | 6.28% | 13.207 us | 6.41% | 0.015 us | 0.11% | SAME |
| F32 | I64 | I64 | 2^20 | 2^4 | 17.497 us | 3.64% | 17.502 us | 3.86% | 0.005 us | 0.03% | SAME |
| F32 | I64 | I64 | 2^24 | 2^4 | 100.236 us | 1.24% | 100.432 us | 1.32% | 0.196 us | 0.20% | SAME |
| F32 | I64 | I64 | 2^28 | 2^4 | 1.406 ms | 0.25% | 1.409 ms | 0.26% | 2.145 us | 0.15% | SAME |
| F32 | I64 | I64 | 2^16 | 2^8 | 13.033 us | 7.29% | 13.177 us | 6.93% | 0.144 us | 1.11% | SAME |
| F32 | I64 | I64 | 2^20 | 2^8 | 17.490 us | 4.41% | 17.628 us | 4.29% | 0.138 us | 0.79% | SAME |
| F32 | I64 | I64 | 2^24 | 2^8 | 96.805 us | 1.47% | 97.136 us | 1.31% | 0.331 us | 0.34% | SAME |
| F32 | I64 | I64 | 2^28 | 2^8 | 1.352 ms | 0.24% | 1.355 ms | 0.25% | 3.350 us | 0.25% | SLOW |
| F64 | I32 | I32 | 2^16 | 2^1 | 11.316 us | 2.76% | 11.316 us | 3.03% | 0.000 us | 0.00% | SAME |
| F64 | I32 | I32 | 2^20 | 2^1 | 17.563 us | 3.29% | 17.553 us | 3.02% | -0.010 us | -0.06% | SAME |
| F64 | I32 | I32 | 2^24 | 2^1 | 102.777 us | 1.20% | 102.825 us | 1.21% | 0.048 us | 0.05% | SAME |
| F64 | I32 | I32 | 2^28 | 2^1 | 1.452 ms | 0.21% | 1.451 ms | 0.21% | -0.220 us | -0.02% | SAME |
| F64 | I32 | I32 | 2^16 | 2^4 | 11.383 us | 3.77% | 11.402 us | 4.13% | 0.019 us | 0.17% | SAME |
| F64 | I32 | I32 | 2^20 | 2^4 | 16.110 us | 6.57% | 16.164 us | 6.62% | 0.054 us | 0.33% | SAME |
| F64 | I32 | I32 | 2^24 | 2^4 | 80.305 us | 1.53% | 80.305 us | 1.42% | -0.000 us | -0.00% | SAME |
| F64 | I32 | I32 | 2^28 | 2^4 | 1.084 ms | 0.17% | 1.084 ms | 0.18% | 0.138 us | 0.01% | SAME |
| F64 | I32 | I32 | 2^16 | 2^8 | 11.359 us | 3.15% | 11.395 us | 3.87% | 0.036 us | 0.31% | SAME |
| F64 | I32 | I32 | 2^20 | 2^8 | 15.799 us | 5.46% | 15.883 us | 5.76% | 0.084 us | 0.53% | SAME |
| F64 | I32 | I32 | 2^24 | 2^8 | 77.566 us | 1.31% | 77.717 us | 1.42% | 0.151 us | 0.19% | SAME |
| F64 | I32 | I32 | 2^28 | 2^8 | 1.037 ms | 0.17% | 1.036 ms | 0.17% | -0.286 us | -0.03% | SAME |
| F64 | I32 | I64 | 2^16 | 2^1 | 13.392 us | 4.18% | 13.454 us | 4.73% | 0.062 us | 0.46% | SAME |
| F64 | I32 | I64 | 2^20 | 2^1 | 19.648 us | 3.53% | 19.566 us | 3.76% | -0.082 us | -0.42% | SAME |
| F64 | I32 | I64 | 2^24 | 2^1 | 123.576 us | 1.06% | 123.436 us | 1.00% | -0.140 us | -0.11% | SAME |
| F64 | I32 | I64 | 2^28 | 2^1 | 1.784 ms | 0.22% | 1.783 ms | 0.22% | -1.033 us | -0.06% | SAME |
| F64 | I32 | I64 | 2^16 | 2^4 | 13.111 us | 6.78% | 13.141 us | 6.21% | 0.030 us | 0.23% | SAME |
| F64 | I32 | I64 | 2^20 | 2^4 | 18.237 us | 5.81% | 18.199 us | 5.80% | -0.038 us | -0.21% | SAME |
| F64 | I32 | I64 | 2^24 | 2^4 | 109.000 us | 1.40% | 109.056 us | 1.28% | 0.056 us | 0.05% | SAME |
| F64 | I32 | I64 | 2^28 | 2^4 | 1.537 ms | 0.31% | 1.540 ms | 0.34% | 2.626 us | 0.17% | SAME |
| F64 | I32 | I64 | 2^16 | 2^8 | 13.152 us | 6.99% | 13.132 us | 6.76% | -0.020 us | -0.15% | SAME |
| F64 | I32 | I64 | 2^20 | 2^8 | 18.064 us | 5.53% | 17.944 us | 4.98% | -0.120 us | -0.66% | SAME |
| F64 | I32 | I64 | 2^24 | 2^8 | 105.020 us | 1.44% | 105.071 us | 1.37% | 0.051 us | 0.05% | SAME |
| F64 | I32 | I64 | 2^28 | 2^8 | 1.454 ms | 0.30% | 1.458 ms | 0.29% | 4.038 us | 0.28% | SAME |
| F64 | I64 | I32 | 2^16 | 2^1 | 11.415 us | 3.95% | 11.428 us | 4.22% | 0.013 us | 0.12% | SAME |
| F64 | I64 | I32 | 2^20 | 2^1 | 17.500 us | 2.28% | 17.547 us | 2.73% | 0.047 us | 0.27% | SAME |
| F64 | I64 | I32 | 2^24 | 2^1 | 103.457 us | 1.08% | 103.407 us | 1.06% | -0.050 us | -0.05% | SAME |
| F64 | I64 | I32 | 2^28 | 2^1 | 1.463 ms | 0.19% | 1.464 ms | 0.20% | 0.164 us | 0.01% | SAME |
| F64 | I64 | I32 | 2^16 | 2^4 | 11.420 us | 4.10% | 11.393 us | 3.56% | -0.027 us | -0.23% | SAME |
| F64 | I64 | I32 | 2^20 | 2^4 | 15.991 us | 5.62% | 16.009 us | 5.90% | 0.017 us | 0.11% | SAME |
| F64 | I64 | I32 | 2^24 | 2^4 | 80.813 us | 1.20% | 80.796 us | 1.26% | -0.017 us | -0.02% | SAME |
| F64 | I64 | I32 | 2^28 | 2^4 | 1.091 ms | 0.18% | 1.092 ms | 0.18% | 0.354 us | 0.03% | SAME |
| F64 | I64 | I32 | 2^16 | 2^8 | 11.391 us | 3.95% | 11.386 us | 4.04% | -0.005 us | -0.04% | SAME |
| F64 | I64 | I32 | 2^20 | 2^8 | 15.518 us | 3.33% | 15.522 us | 3.48% | 0.004 us | 0.03% | SAME |
| F64 | I64 | I32 | 2^24 | 2^8 | 76.947 us | 0.81% | 76.997 us | 0.90% | 0.050 us | 0.06% | SAME |
| F64 | I64 | I32 | 2^28 | 2^8 | 1.027 ms | 0.16% | 1.027 ms | 0.16% | 0.009 us | 0.00% | SAME |
| F64 | I64 | I64 | 2^16 | 2^1 | 13.360 us | 5.74% | 13.345 us | 4.45% | -0.015 us | -0.11% | SAME |
| F64 | I64 | I64 | 2^20 | 2^1 | 19.534 us | 2.88% | 19.571 us | 3.09% | 0.037 us | 0.19% | SAME |
| F64 | I64 | I64 | 2^24 | 2^1 | 130.207 us | 0.98% | 130.396 us | 1.07% | 0.189 us | 0.15% | SAME |
| F64 | I64 | I64 | 2^28 | 2^1 | 1.890 ms | 0.18% | 1.890 ms | 0.19% | -0.105 us | -0.01% | SAME |
| F64 | I64 | I64 | 2^16 | 2^4 | 12.879 us | 8.74% | 12.885 us | 8.17% | 0.005 us | 0.04% | SAME |
| F64 | I64 | I64 | 2^20 | 2^4 | 18.055 us | 5.39% | 18.083 us | 5.58% | 0.028 us | 0.15% | SAME |
| F64 | I64 | I64 | 2^24 | 2^4 | 115.113 us | 1.27% | 115.170 us | 1.34% | 0.057 us | 0.05% | SAME |
| F64 | I64 | I64 | 2^28 | 2^4 | 1.632 ms | 0.26% | 1.632 ms | 0.26% | 0.018 us | 0.00% | SAME |
| F64 | I64 | I64 | 2^16 | 2^8 | 12.951 us | 9.37% | 12.941 us | 8.72% | -0.009 us | -0.07% | SAME |
| F64 | I64 | I64 | 2^20 | 2^8 | 17.862 us | 4.39% | 17.894 us | 4.64% | 0.032 us | 0.18% | SAME |
| F64 | I64 | I64 | 2^24 | 2^8 | 110.200 us | 1.44% | 110.296 us | 1.32% | 0.096 us | 0.09% | SAME |
| F64 | I64 | I64 | 2^28 | 2^8 | 1.546 ms | 0.29% | 1.545 ms | 0.29% | -0.151 us | -0.01% | SAME |
| C64 | I32 | I32 | 2^16 | 2^1 | 13.357 us | 2.96% | 13.346 us | 2.44% | -0.012 us | -0.09% | SAME |
| C64 | I32 | I32 | 2^20 | 2^1 | 19.269 us | 4.50% | 19.511 us | 2.61% | 0.242 us | 1.25% | SAME |
| C64 | I32 | I32 | 2^24 | 2^1 | 121.824 us | 0.92% | 128.793 us | 1.00% | 6.969 us | 5.72% | SLOW |
| C64 | I32 | I32 | 2^28 | 2^1 | 1.511 ms | 0.28% | 1.600 ms | 0.30% | 88.977 us | 5.89% | SLOW |
| C64 | I32 | I32 | 2^16 | 2^4 | 13.306 us | 4.98% | 13.202 us | 5.99% | -0.104 us | -0.79% | SAME |
| C64 | I32 | I32 | 2^20 | 2^4 | 17.916 us | 4.99% | 18.098 us | 5.29% | 0.181 us | 1.01% | SAME |
| C64 | I32 | I32 | 2^24 | 2^4 | 106.512 us | 1.37% | 112.074 us | 1.54% | 5.562 us | 5.22% | SLOW |
| C64 | I32 | I32 | 2^28 | 2^4 | 1.478 ms | 0.30% | 1.564 ms | 0.32% | 85.964 us | 5.82% | SLOW |
| C64 | I32 | I32 | 2^16 | 2^8 | 13.288 us | 5.37% | 13.263 us | 6.06% | -0.025 us | -0.19% | SAME |
| C64 | I32 | I32 | 2^20 | 2^8 | 17.768 us | 3.95% | 17.837 us | 4.46% | 0.069 us | 0.39% | SAME |
| C64 | I32 | I32 | 2^24 | 2^8 | 103.128 us | 1.48% | 108.381 us | 1.60% | 5.253 us | 5.09% | SLOW |
| C64 | I32 | I32 | 2^28 | 2^8 | 1.427 ms | 0.30% | 1.509 ms | 0.31% | 82.147 us | 5.76% | SLOW |
| C64 | I32 | I64 | 2^16 | 2^1 | 13.558 us | 5.11% | 13.548 us | 4.76% | -0.010 us | -0.08% | SAME |
| C64 | I32 | I64 | 2^20 | 2^1 | 19.723 us | 3.22% | 19.774 us | 3.13% | 0.051 us | 0.26% | SAME |
| C64 | I32 | I64 | 2^24 | 2^1 | 132.359 us | 0.75% | 132.452 us | 0.75% | 0.093 us | 0.07% | SAME |
| C64 | I32 | I64 | 2^28 | 2^1 | 1.609 ms | 0.25% | 1.612 ms | 0.25% | 3.078 us | 0.19% | SAME |
| C64 | I32 | I64 | 2^16 | 2^4 | 13.209 us | 5.60% | 13.276 us | 5.25% | 0.067 us | 0.50% | SAME |
| C64 | I32 | I64 | 2^20 | 2^4 | 18.616 us | 6.16% | 18.612 us | 5.68% | -0.004 us | -0.02% | SAME |
| C64 | I32 | I64 | 2^24 | 2^4 | 111.960 us | 1.25% | 112.032 us | 1.19% | 0.072 us | 0.06% | SAME |
| C64 | I32 | I64 | 2^28 | 2^4 | 1.568 ms | 0.28% | 1.571 ms | 0.26% | 2.461 us | 0.16% | SAME |
| C64 | I32 | I64 | 2^16 | 2^8 | 13.234 us | 5.86% | 13.265 us | 5.06% | 0.031 us | 0.24% | SAME |
| C64 | I32 | I64 | 2^20 | 2^8 | 18.406 us | 5.91% | 18.414 us | 6.18% | 0.007 us | 0.04% | SAME |
| C64 | I32 | I64 | 2^24 | 2^8 | 107.946 us | 1.26% | 108.025 us | 1.19% | 0.080 us | 0.07% | SAME |
| C64 | I32 | I64 | 2^28 | 2^8 | 1.504 ms | 0.25% | 1.506 ms | 0.26% | 2.234 us | 0.15% | SAME |
| C64 | I64 | I32 | 2^16 | 2^1 | 13.225 us | 6.08% | 13.309 us | 6.16% | 0.084 us | 0.64% | SAME |
| C64 | I64 | I32 | 2^20 | 2^1 | 19.232 us | 5.13% | 19.265 us | 5.37% | 0.033 us | 0.17% | SAME |
| C64 | I64 | I32 | 2^24 | 2^1 | 127.363 us | 1.04% | 127.029 us | 1.08% | -0.334 us | -0.26% | SAME |
| C64 | I64 | I32 | 2^28 | 2^1 | 1.581 ms | 0.31% | 1.579 ms | 0.30% | -1.902 us | -0.12% | SAME |
| C64 | I64 | I32 | 2^16 | 2^4 | 13.027 us | 7.75% | 12.838 us | 8.35% | -0.189 us | -1.45% | SAME |
| C64 | I64 | I32 | 2^20 | 2^4 | 17.824 us | 4.50% | 17.731 us | 3.94% | -0.093 us | -0.52% | SAME |
| C64 | I64 | I32 | 2^24 | 2^4 | 110.466 us | 1.52% | 110.422 us | 1.44% | -0.044 us | -0.04% | SAME |
| C64 | I64 | I32 | 2^28 | 2^4 | 1.545 ms | 0.31% | 1.543 ms | 0.32% | -1.545 us | -0.10% | SAME |
| C64 | I64 | I32 | 2^16 | 2^8 | 12.806 us | 7.68% | 12.748 us | 8.17% | -0.057 us | -0.45% | SAME |
| C64 | I64 | I32 | 2^20 | 2^8 | 17.560 us | 3.08% | 17.578 us | 3.09% | 0.018 us | 0.10% | SAME |
| C64 | I64 | I32 | 2^24 | 2^8 | 106.164 us | 1.56% | 106.175 us | 1.64% | 0.011 us | 0.01% | SAME |
| C64 | I64 | I32 | 2^28 | 2^8 | 1.486 ms | 0.32% | 1.486 ms | 0.34% | -0.359 us | -0.02% | SAME |
| C64 | I64 | I64 | 2^16 | 2^1 | 13.269 us | 5.29% | 13.278 us | 5.22% | 0.009 us | 0.07% | SAME |
| C64 | I64 | I64 | 2^20 | 2^1 | 19.601 us | 2.43% | 19.614 us | 2.48% | 0.013 us | 0.07% | SAME |
| C64 | I64 | I64 | 2^24 | 2^1 | 138.302 us | 0.69% | 138.329 us | 0.77% | 0.027 us | 0.02% | SAME |
| C64 | I64 | I64 | 2^28 | 2^1 | 1.695 ms | 0.22% | 1.692 ms | 0.22% | -2.925 us | -0.17% | SAME |
| C64 | I64 | I64 | 2^16 | 2^4 | 13.034 us | 6.85% | 13.037 us | 6.92% | 0.002 us | 0.02% | SAME |
| C64 | I64 | I64 | 2^20 | 2^4 | 18.395 us | 6.08% | 18.328 us | 6.04% | -0.067 us | -0.36% | SAME |
| C64 | I64 | I64 | 2^24 | 2^4 | 117.101 us | 1.24% | 116.840 us | 1.24% | -0.261 us | -0.22% | SAME |
| C64 | I64 | I64 | 2^28 | 2^4 | 1.652 ms | 0.25% | 1.649 ms | 0.23% | -3.053 us | -0.18% | SAME |
| C64 | I64 | I64 | 2^16 | 2^8 | 12.921 us | 8.21% | 12.971 us | 7.54% | 0.050 us | 0.39% | SAME |
| C64 | I64 | I64 | 2^20 | 2^8 | 17.986 us | 5.03% | 18.009 us | 5.30% | 0.023 us | 0.13% | SAME |
| C64 | I64 | I64 | 2^24 | 2^8 | 112.151 us | 1.29% | 111.859 us | 1.23% | -0.292 us | -0.26% | SAME |
| C64 | I64 | I64 | 2^28 | 2^8 | 1.582 ms | 0.25% | 1.576 ms | 0.24% | -5.108 us | -0.32% | FAST |
That's somewhat surprising and I think we should still investigate a bit more what's going on. Especially for those runs:
| C64 | I32 | I32 | 2^24 | 2^1 | 121.824 us | 0.92% | 128.793 us | 1.00% | 6.969 us | 5.72% | SLOW |
| C64 | I32 | I32 | 2^28 | 2^1 | 1.511 ms | 0.28% | 1.600 ms | 0.30% | 88.977 us | 5.89% | SLOW |
| C64 | I32 | I32 | 2^24 | 2^4 | 106.512 us | 1.37% | 112.074 us | 1.54% | 5.562 us | 5.22% | SLOW |
| C64 | I32 | I32 | 2^28 | 2^4 | 1.478 ms | 0.30% | 1.564 ms | 0.32% | 85.964 us | 5.82% | SLOW |
| C64 | I32 | I32 | 2^24 | 2^8 | 103.128 us | 1.48% | 108.381 us | 1.60% | 5.253 us | 5.09% | SLOW |
| C64 | I32 | I32 | 2^28 | 2^8 | 1.427 ms | 0.30% | 1.509 ms | 0.31% | 82.147 us | 5.76% | SLOW |
Since changes to CUB headers seem to not affect SASS, let's try reverting the changes to the benchmark definition and nvbench:
diff --git a/cub/benchmarks/bench/run_length_encode/encode.cu b/cub/benchmarks/bench/run_length_encode/encode.cu
index 80ffe0599d..a79f918ccf 100644
--- a/cub/benchmarks/bench/run_length_encode/encode.cu
+++ b/cub/benchmarks/bench/run_length_encode/encode.cu
@@ -28,6 +28,7 @@
#include <cub/device/device_run_length_encode.cuh>
#include <cuda/iterator>
+#include <thrust/iterator/constant_iterator.h>
#include <look_back_helper.cuh>
#include <nvbench_helper.cuh>
@@ -82,7 +83,7 @@ static void rle(nvbench::state& state, nvbench::type_list<T, OffsetT, RunLengthT
using keys_input_it_t = const T*;
using unique_output_it_t = T*;
- using run_length_input_it_t = cuda::constant_iterator<run_length_t, offset_t>;
+ using run_length_input_it_t = thrust::constant_iterator<run_length_t, offset_t>;
using run_length_output_it_t = run_length_t*;
using num_runs_output_iterator_t = num_runs_t*;
using equality_op_t = ::cuda::std::equal_to<>;
diff --git a/nvbench_helper/nvbench_helper/nvbench_helper.cu b/nvbench_helper/nvbench_helper/nvbench_helper.cu
index b187d33964..8f9809f1c3 100644
--- a/nvbench_helper/nvbench_helper/nvbench_helper.cu
+++ b/nvbench_helper/nvbench_helper/nvbench_helper.cu
@@ -8,6 +8,7 @@
#include <thrust/fill.h>
#include <thrust/for_each.h>
#include <thrust/host_vector.h>
+#include <thrust/iterator/constant_iterator.h>
#include <thrust/iterator/counting_iterator.h>
#include <thrust/iterator/transform_output_iterator.h>
#include <thrust/iterator/zip_iterator.h>
@@ -582,9 +583,9 @@ struct offset_to_iterator_t
template <class T>
struct repeat_index_t
{
- __host__ __device__ __forceinline__ cuda::constant_iterator<T> operator()(std::size_t i)
+ __host__ __device__ __forceinline__ thrust::constant_iterator<T> operator()(std::size_t i)
{
- return cuda::constant_iterator<T>(static_cast<T>(i));
+ return thrust::constant_iterator<T>(static_cast<T>(i));
}
};
This now results in the same, or better, performance as before this PR:
## [0] NVIDIA B200
| T{ct} | OffsetT{ct} | RunLengthT{ct} | Elements{io} | MaxSegSize | Ref Time | Ref Noise | Cmp Time | Cmp Noise | Diff | %Diff | Status |
|---------|---------------|------------------|----------------|--------------|------------|-------------|------------|-------------|-----------|---------|----------|
| I8 | I32 | I32 | 2^16 | 2^1 | 11.563 us | 6.18% | 11.600 us | 6.39% | 0.036 us | 0.31% | SAME |
| I8 | I32 | I32 | 2^20 | 2^1 | 13.735 us | 6.15% | 13.788 us | 6.53% | 0.053 us | 0.39% | SAME |
| I8 | I32 | I32 | 2^24 | 2^1 | 70.537 us | 1.21% | 70.678 us | 1.12% | 0.141 us | 0.20% | SAME |
| I8 | I32 | I32 | 2^28 | 2^1 | 960.569 us | 0.20% | 960.670 us | 0.20% | 0.101 us | 0.01% | SAME |
| I8 | I32 | I32 | 2^16 | 2^4 | 11.363 us | 3.68% | 11.400 us | 3.70% | 0.038 us | 0.33% | SAME |
| I8 | I32 | I32 | 2^20 | 2^4 | 13.421 us | 3.41% | 13.456 us | 3.56% | 0.035 us | 0.26% | SAME |
| I8 | I32 | I32 | 2^24 | 2^4 | 55.936 us | 1.83% | 55.962 us | 1.88% | 0.026 us | 0.05% | SAME |
| I8 | I32 | I32 | 2^28 | 2^4 | 734.661 us | 0.26% | 734.916 us | 0.28% | 0.255 us | 0.03% | SAME |
| I8 | I32 | I32 | 2^16 | 2^8 | 11.431 us | 5.00% | 11.434 us | 4.33% | 0.003 us | 0.03% | SAME |
| I8 | I32 | I32 | 2^20 | 2^8 | 13.412 us | 3.15% | 13.525 us | 3.77% | 0.113 us | 0.84% | SAME |
| I8 | I32 | I32 | 2^24 | 2^8 | 54.474 us | 1.31% | 54.455 us | 1.12% | -0.020 us | -0.04% | SAME |
| I8 | I32 | I32 | 2^28 | 2^8 | 710.287 us | 0.27% | 710.586 us | 0.27% | 0.299 us | 0.04% | SAME |
| I8 | I32 | I64 | 2^16 | 2^1 | 14.090 us | 7.49% | 13.503 us | 4.31% | -0.586 us | -4.16% | SAME |
| I8 | I32 | I64 | 2^20 | 2^1 | 18.300 us | 6.59% | 17.949 us | 4.76% | -0.351 us | -1.92% | SAME |
| I8 | I32 | I64 | 2^24 | 2^1 | 110.951 us | 0.92% | 109.900 us | 0.63% | -1.051 us | -0.95% | FAST |
| I8 | I32 | I64 | 2^28 | 2^1 | 1.580 ms | 0.08% | 1.580 ms | 0.08% | -0.016 us | -0.00% | SAME |
| I8 | I32 | I64 | 2^16 | 2^4 | 13.942 us | 7.32% | 13.312 us | 4.21% | -0.630 us | -4.52% | FAST |
| I8 | I32 | I64 | 2^20 | 2^4 | 17.897 us | 5.88% | 17.377 us | 3.53% | -0.521 us | -2.91% | SAME |
| I8 | I32 | I64 | 2^24 | 2^4 | 93.895 us | 1.52% | 92.833 us | 1.32% | -1.062 us | -1.13% | SAME |
| I8 | I32 | I64 | 2^28 | 2^4 | 1.304 ms | 0.16% | 1.304 ms | 0.16% | -0.074 us | -0.01% | SAME |
| I8 | I32 | I64 | 2^16 | 2^8 | 13.917 us | 6.83% | 13.364 us | 4.21% | -0.553 us | -3.97% | SAME |
| I8 | I32 | I64 | 2^20 | 2^8 | 17.810 us | 7.34% | 17.402 us | 5.11% | -0.408 us | -2.29% | SAME |
| I8 | I32 | I64 | 2^24 | 2^8 | 91.669 us | 1.60% | 90.729 us | 1.31% | -0.940 us | -1.03% | SAME |
| I8 | I32 | I64 | 2^28 | 2^8 | 1.267 ms | 0.18% | 1.267 ms | 0.18% | -0.110 us | -0.01% | SAME |
| I8 | I64 | I32 | 2^16 | 2^1 | 11.729 us | 7.27% | 11.333 us | 2.96% | -0.396 us | -3.38% | FAST |
| I8 | I64 | I32 | 2^20 | 2^1 | 14.312 us | 10.88% | 13.669 us | 6.30% | -0.643 us | -4.49% | SAME |
| I8 | I64 | I32 | 2^24 | 2^1 | 69.735 us | 1.51% | 68.716 us | 0.87% | -1.020 us | -1.46% | FAST |
| I8 | I64 | I32 | 2^28 | 2^1 | 936.946 us | 0.18% | 937.268 us | 0.17% | 0.322 us | 0.03% | SAME |
| I8 | I64 | I32 | 2^16 | 2^4 | 11.683 us | 6.99% | 11.428 us | 4.08% | -0.254 us | -2.18% | SAME |
| I8 | I64 | I32 | 2^20 | 2^4 | 14.036 us | 6.96% | 13.533 us | 3.97% | -0.502 us | -3.58% | SAME |
| I8 | I64 | I32 | 2^24 | 2^4 | 55.488 us | 2.04% | 54.456 us | 1.18% | -1.032 us | -1.86% | FAST |
| I8 | I64 | I32 | 2^28 | 2^4 | 712.085 us | 0.21% | 711.074 us | 0.27% | -1.011 us | -0.14% | SAME |
| I8 | I64 | I32 | 2^16 | 2^8 | 11.661 us | 6.92% | 11.301 us | 2.33% | -0.359 us | -3.08% | FAST |
| I8 | I64 | I32 | 2^20 | 2^8 | 13.946 us | 6.79% | 13.447 us | 3.52% | -0.499 us | -3.58% | FAST |
| I8 | I64 | I32 | 2^24 | 2^8 | 53.260 us | 2.28% | 52.247 us | 1.47% | -1.013 us | -1.90% | FAST |
| I8 | I64 | I32 | 2^28 | 2^8 | 673.695 us | 0.29% | 673.773 us | 0.28% | 0.078 us | 0.01% | SAME |
| I8 | I64 | I64 | 2^16 | 2^1 | 13.843 us | 8.91% | 13.255 us | 6.43% | -0.588 us | -4.25% | SAME |
| I8 | I64 | I64 | 2^20 | 2^1 | 18.136 us | 5.89% | 17.702 us | 4.07% | -0.434 us | -2.39% | SAME |
| I8 | I64 | I64 | 2^24 | 2^1 | 114.873 us | 0.99% | 113.792 us | 0.70% | -1.081 us | -0.94% | FAST |
| I8 | I64 | I64 | 2^28 | 2^1 | 1.647 ms | 0.09% | 1.647 ms | 0.09% | -0.100 us | -0.01% | SAME |
| I8 | I64 | I64 | 2^16 | 2^4 | 13.610 us | 9.17% | 13.051 us | 6.54% | -0.559 us | -4.11% | SAME |
| I8 | I64 | I64 | 2^20 | 2^4 | 17.886 us | 6.44% | 17.395 us | 4.89% | -0.491 us | -2.75% | SAME |
| I8 | I64 | I64 | 2^24 | 2^4 | 97.437 us | 1.57% | 96.279 us | 1.24% | -1.158 us | -1.19% | SAME |
| I8 | I64 | I64 | 2^28 | 2^4 | 1.367 ms | 0.17% | 1.367 ms | 0.17% | 0.140 us | 0.01% | SAME |
| I8 | I64 | I64 | 2^16 | 2^8 | 13.768 us | 9.40% | 12.999 us | 8.23% | -0.769 us | -5.59% | SAME |
| I8 | I64 | I64 | 2^20 | 2^8 | 17.503 us | 8.36% | 17.020 us | 6.62% | -0.483 us | -2.76% | SAME |
| I8 | I64 | I64 | 2^24 | 2^8 | 94.823 us | 1.35% | 93.839 us | 1.22% | -0.984 us | -1.04% | SAME |
| I8 | I64 | I64 | 2^28 | 2^8 | 1.324 ms | 0.19% | 1.324 ms | 0.18% | 0.066 us | 0.00% | SAME |
| I16 | I32 | I32 | 2^16 | 2^1 | 11.499 us | 5.74% | 11.315 us | 2.72% | -0.183 us | -1.60% | SAME |
| I16 | I32 | I32 | 2^20 | 2^1 | 16.022 us | 5.96% | 15.360 us | 0.00% | -0.662 us | -4.13% | ???? |
| I16 | I32 | I32 | 2^24 | 2^1 | 69.291 us | 2.06% | 68.184 us | 1.70% | -1.107 us | -1.60% | SAME |
| I16 | I32 | I32 | 2^28 | 2^1 | 915.073 us | 0.20% | 914.281 us | 0.23% | -0.793 us | -0.09% | SAME |
| I16 | I32 | I32 | 2^16 | 2^4 | 11.650 us | 6.86% | 11.393 us | 3.90% | -0.257 us | -2.21% | SAME |
| I16 | I32 | I32 | 2^20 | 2^4 | 13.992 us | 8.02% | 13.700 us | 5.58% | -0.292 us | -2.09% | SAME |
| I16 | I32 | I32 | 2^24 | 2^4 | 56.103 us | 2.56% | 55.061 us | 1.97% | -1.042 us | -1.86% | SAME |
| I16 | I32 | I32 | 2^28 | 2^4 | 712.815 us | 0.23% | 712.432 us | 0.27% | -0.383 us | -0.05% | SAME |
| I16 | I32 | I32 | 2^16 | 2^8 | 11.528 us | 5.86% | 11.360 us | 3.17% | -0.168 us | -1.46% | SAME |
| I16 | I32 | I32 | 2^20 | 2^8 | 14.066 us | 7.37% | 13.654 us | 5.69% | -0.412 us | -2.93% | SAME |
| I16 | I32 | I32 | 2^24 | 2^8 | 54.878 us | 2.47% | 53.769 us | 2.06% | -1.109 us | -2.02% | SAME |
| I16 | I32 | I32 | 2^28 | 2^8 | 686.441 us | 0.26% | 685.992 us | 0.27% | -0.448 us | -0.07% | SAME |
| I16 | I32 | I64 | 2^16 | 2^1 | 13.684 us | 6.18% | 13.399 us | 4.06% | -0.285 us | -2.08% | SAME |
| I16 | I32 | I64 | 2^20 | 2^1 | 18.372 us | 6.94% | 17.947 us | 5.39% | -0.425 us | -2.31% | SAME |
| I16 | I32 | I64 | 2^24 | 2^1 | 113.095 us | 1.01% | 112.057 us | 0.76% | -1.038 us | -0.92% | FAST |
| I16 | I32 | I64 | 2^28 | 2^1 | 1.613 ms | 0.09% | 1.613 ms | 0.09% | -0.055 us | -0.00% | SAME |
| I16 | I32 | I64 | 2^16 | 2^4 | 13.838 us | 8.04% | 13.299 us | 4.72% | -0.539 us | -3.90% | SAME |
| I16 | I32 | I64 | 2^20 | 2^4 | 17.416 us | 4.07% | 17.414 us | 3.47% | -0.003 us | -0.02% | SAME |
| I16 | I32 | I64 | 2^24 | 2^4 | 96.225 us | 1.26% | 96.087 us | 1.22% | -0.138 us | -0.14% | SAME |
| I16 | I32 | I64 | 2^28 | 2^4 | 1.354 ms | 0.19% | 1.354 ms | 0.20% | -0.053 us | -0.00% | SAME |
| I16 | I32 | I64 | 2^16 | 2^8 | 13.159 us | 6.31% | 13.160 us | 5.92% | 0.001 us | 0.01% | SAME |
| I16 | I32 | I64 | 2^20 | 2^8 | 17.423 us | 3.95% | 17.470 us | 3.77% | 0.047 us | 0.27% | SAME |
| I16 | I32 | I64 | 2^24 | 2^8 | 93.660 us | 1.19% | 93.636 us | 1.10% | -0.024 us | -0.03% | SAME |
| I16 | I32 | I64 | 2^28 | 2^8 | 1.309 ms | 0.19% | 1.309 ms | 0.19% | -0.126 us | -0.01% | SAME |
| I16 | I64 | I32 | 2^16 | 2^1 | 11.405 us | 3.89% | 11.421 us | 4.32% | 0.016 us | 0.14% | SAME |
| I16 | I64 | I32 | 2^20 | 2^1 | 15.401 us | 3.65% | 15.448 us | 3.40% | 0.047 us | 0.31% | SAME |
| I16 | I64 | I32 | 2^24 | 2^1 | 74.955 us | 1.09% | 75.070 us | 1.24% | 0.115 us | 0.15% | SAME |
| I16 | I64 | I32 | 2^28 | 2^1 | 1.040 ms | 0.23% | 1.040 ms | 0.22% | 0.060 us | 0.01% | SAME |
| I16 | I64 | I32 | 2^16 | 2^4 | 11.519 us | 5.96% | 11.407 us | 3.90% | -0.112 us | -0.97% | SAME |
| I16 | I64 | I32 | 2^20 | 2^4 | 13.651 us | 6.07% | 13.643 us | 5.44% | -0.008 us | -0.06% | SAME |
| I16 | I64 | I32 | 2^24 | 2^4 | 61.315 us | 1.90% | 61.180 us | 1.70% | -0.134 us | -0.22% | SAME |
| I16 | I64 | I32 | 2^28 | 2^4 | 821.175 us | 0.23% | 820.734 us | 0.24% | -0.441 us | -0.05% | SAME |
| I16 | I64 | I32 | 2^16 | 2^8 | 11.832 us | 7.74% | 11.357 us | 3.58% | -0.475 us | -4.01% | FAST |
| I16 | I64 | I32 | 2^20 | 2^8 | 13.853 us | 7.48% | 13.448 us | 3.72% | -0.405 us | -2.93% | SAME |
| I16 | I64 | I32 | 2^24 | 2^8 | 59.830 us | 2.30% | 58.892 us | 1.71% | -0.938 us | -1.57% | SAME |
| I16 | I64 | I32 | 2^28 | 2^8 | 780.664 us | 0.24% | 780.071 us | 0.28% | -0.593 us | -0.08% | SAME |
| I16 | I64 | I64 | 2^16 | 2^1 | 13.535 us | 9.56% | 12.910 us | 7.76% | -0.625 us | -4.62% | SAME |
| I16 | I64 | I64 | 2^20 | 2^1 | 18.061 us | 5.34% | 17.622 us | 3.12% | -0.440 us | -2.44% | SAME |
| I16 | I64 | I64 | 2^24 | 2^1 | 115.032 us | 1.02% | 113.935 us | 0.66% | -1.097 us | -0.95% | FAST |
| I16 | I64 | I64 | 2^28 | 2^1 | 1.649 ms | 0.11% | 1.649 ms | 0.11% | -0.258 us | -0.02% | SAME |
| I16 | I64 | I64 | 2^16 | 2^4 | 12.590 us | 11.02% | 12.089 us | 9.02% | -0.500 us | -3.97% | SAME |
| I16 | I64 | I64 | 2^20 | 2^4 | 17.483 us | 7.98% | 17.036 us | 5.99% | -0.448 us | -2.56% | SAME |
| I16 | I64 | I64 | 2^24 | 2^4 | 98.317 us | 1.48% | 97.193 us | 1.33% | -1.124 us | -1.14% | SAME |
| I16 | I64 | I64 | 2^28 | 2^4 | 1.380 ms | 0.22% | 1.380 ms | 0.22% | 0.225 us | 0.02% | SAME |
| I16 | I64 | I64 | 2^16 | 2^8 | 12.799 us | 11.01% | 12.034 us | 8.75% | -0.765 us | -5.98% | SAME |
| I16 | I64 | I64 | 2^20 | 2^8 | 17.210 us | 8.35% | 16.581 us | 6.79% | -0.629 us | -3.65% | SAME |
| I16 | I64 | I64 | 2^24 | 2^8 | 95.180 us | 1.50% | 94.213 us | 1.36% | -0.967 us | -1.02% | SAME |
| I16 | I64 | I64 | 2^28 | 2^8 | 1.325 ms | 0.23% | 1.325 ms | 0.23% | 0.279 us | 0.02% | SAME |
| I32 | I32 | I32 | 2^16 | 2^1 | 11.518 us | 5.80% | 11.313 us | 2.54% | -0.205 us | -1.78% | SAME |
| I32 | I32 | I32 | 2^20 | 2^1 | 16.012 us | 5.93% | 15.465 us | 2.86% | -0.547 us | -3.42% | FAST |
| I32 | I32 | I32 | 2^24 | 2^1 | 67.852 us | 1.80% | 66.837 us | 1.16% | -1.015 us | -1.50% | FAST |
| I32 | I32 | I32 | 2^28 | 2^1 | 893.816 us | 0.23% | 892.906 us | 0.22% | -0.910 us | -0.10% | SAME |
| I32 | I32 | I32 | 2^16 | 2^4 | 11.645 us | 6.85% | 11.369 us | 3.43% | -0.276 us | -2.37% | SAME |
| I32 | I32 | I32 | 2^20 | 2^4 | 15.132 us | 8.73% | 14.890 us | 7.25% | -0.242 us | -1.60% | SAME |
| I32 | I32 | I32 | 2^24 | 2^4 | 55.815 us | 2.39% | 54.832 us | 1.71% | -0.983 us | -1.76% | FAST |
| I32 | I32 | I32 | 2^28 | 2^4 | 696.998 us | 0.21% | 696.550 us | 0.27% | -0.449 us | -0.06% | SAME |
| I32 | I32 | I32 | 2^16 | 2^8 | 11.854 us | 7.82% | 11.384 us | 3.84% | -0.470 us | -3.97% | FAST |
| I32 | I32 | I32 | 2^20 | 2^8 | 14.994 us | 10.17% | 14.746 us | 7.52% | -0.248 us | -1.66% | SAME |
| I32 | I32 | I32 | 2^24 | 2^8 | 54.843 us | 2.77% | 53.678 us | 2.15% | -1.165 us | -2.12% | SAME |
| I32 | I32 | I32 | 2^28 | 2^8 | 667.165 us | 0.28% | 667.214 us | 0.27% | 0.049 us | 0.01% | SAME |
| I32 | I32 | I64 | 2^16 | 2^1 | 14.055 us | 7.58% | 13.559 us | 4.53% | -0.496 us | -3.53% | SAME |
| I32 | I32 | I64 | 2^20 | 2^1 | 19.140 us | 8.02% | 18.756 us | 6.03% | -0.383 us | -2.00% | SAME |
| I32 | I32 | I64 | 2^24 | 2^1 | 114.935 us | 0.94% | 113.940 us | 0.71% | -0.994 us | -0.87% | FAST |
| I32 | I32 | I64 | 2^28 | 2^1 | 1.631 ms | 0.11% | 1.631 ms | 0.12% | 0.115 us | 0.01% | SAME |
| I32 | I32 | I64 | 2^16 | 2^4 | 13.668 us | 5.95% | 13.387 us | 4.23% | -0.281 us | -2.06% | SAME |
| I32 | I32 | I64 | 2^20 | 2^4 | 18.115 us | 5.69% | 17.589 us | 3.33% | -0.526 us | -2.91% | SAME |
| I32 | I32 | I64 | 2^24 | 2^4 | 98.015 us | 1.66% | 96.840 us | 1.31% | -1.176 us | -1.20% | SAME |
| I32 | I32 | I64 | 2^28 | 2^4 | 1.350 ms | 0.25% | 1.350 ms | 0.26% | 0.061 us | 0.00% | SAME |
| I32 | I32 | I64 | 2^16 | 2^8 | 13.932 us | 7.01% | 13.431 us | 3.75% | -0.501 us | -3.59% | SAME |
| I32 | I32 | I64 | 2^20 | 2^8 | 18.097 us | 5.55% | 17.668 us | 3.55% | -0.429 us | -2.37% | SAME |
| I32 | I32 | I64 | 2^24 | 2^8 | 95.006 us | 1.59% | 93.950 us | 1.32% | -1.055 us | -1.11% | SAME |
| I32 | I32 | I64 | 2^28 | 2^8 | 1.297 ms | 0.25% | 1.297 ms | 0.24% | 0.106 us | 0.01% | SAME |
| I32 | I64 | I32 | 2^16 | 2^1 | 11.555 us | 6.07% | 11.337 us | 2.97% | -0.218 us | -1.89% | SAME |
| I32 | I64 | I32 | 2^20 | 2^1 | 15.873 us | 6.66% | 15.481 us | 4.33% | -0.392 us | -2.47% | SAME |
| I32 | I64 | I32 | 2^24 | 2^1 | 73.329 us | 1.86% | 72.398 us | 1.44% | -0.931 us | -1.27% | SAME |
| I32 | I64 | I32 | 2^28 | 2^1 | 990.122 us | 0.23% | 989.519 us | 0.25% | -0.603 us | -0.06% | SAME |
| I32 | I64 | I32 | 2^16 | 2^4 | 11.597 us | 6.50% | 11.398 us | 3.84% | -0.199 us | -1.71% | SAME |
| I32 | I64 | I32 | 2^20 | 2^4 | 14.935 us | 8.96% | 14.825 us | 7.76% | -0.110 us | -0.74% | SAME |
| I32 | I64 | I32 | 2^24 | 2^4 | 60.266 us | 2.37% | 59.188 us | 1.85% | -1.078 us | -1.79% | SAME |
| I32 | I64 | I32 | 2^28 | 2^4 | 781.092 us | 0.22% | 779.896 us | 0.26% | -1.196 us | -0.15% | SAME |
| I32 | I64 | I32 | 2^16 | 2^8 | 11.875 us | 7.80% | 11.332 us | 3.30% | -0.543 us | -4.57% | FAST |
| I32 | I64 | I32 | 2^20 | 2^8 | 15.094 us | 9.00% | 14.567 us | 7.42% | -0.526 us | -3.49% | SAME |
| I32 | I64 | I32 | 2^24 | 2^8 | 58.159 us | 2.57% | 57.257 us | 1.94% | -0.903 us | -1.55% | SAME |
| I32 | I64 | I32 | 2^28 | 2^8 | 743.407 us | 0.26% | 743.368 us | 0.28% | -0.040 us | -0.01% | SAME |
| I32 | I64 | I64 | 2^16 | 2^1 | 13.669 us | 6.78% | 13.246 us | 5.14% | -0.423 us | -3.09% | SAME |
| I32 | I64 | I64 | 2^20 | 2^1 | 18.456 us | 7.77% | 18.083 us | 5.58% | -0.374 us | -2.02% | SAME |
| I32 | I64 | I64 | 2^24 | 2^1 | 118.827 us | 1.10% | 117.635 us | 0.89% | -1.191 us | -1.00% | FAST |
| I32 | I64 | I64 | 2^28 | 2^1 | 1.701 ms | 0.13% | 1.701 ms | 0.13% | -0.003 us | -0.00% | SAME |
| I32 | I64 | I64 | 2^16 | 2^4 | 13.503 us | 7.59% | 13.144 us | 5.83% | -0.359 us | -2.66% | SAME |
| I32 | I64 | I64 | 2^20 | 2^4 | 18.016 us | 5.59% | 17.563 us | 3.61% | -0.453 us | -2.52% | SAME |
| I32 | I64 | I64 | 2^24 | 2^4 | 101.608 us | 1.44% | 100.563 us | 1.26% | -1.045 us | -1.03% | SAME |
| I32 | I64 | I64 | 2^28 | 2^4 | 1.419 ms | 0.25% | 1.419 ms | 0.24% | 0.197 us | 0.01% | SAME |
| I32 | I64 | I64 | 2^16 | 2^8 | 13.864 us | 8.23% | 13.259 us | 5.88% | -0.605 us | -4.36% | SAME |
| I32 | I64 | I64 | 2^20 | 2^8 | 17.933 us | 6.93% | 17.494 us | 4.57% | -0.439 us | -2.45% | SAME |
| I32 | I64 | I64 | 2^24 | 2^8 | 98.340 us | 1.62% | 97.451 us | 1.36% | -0.888 us | -0.90% | SAME |
| I32 | I64 | I64 | 2^28 | 2^8 | 1.358 ms | 0.25% | 1.358 ms | 0.25% | -0.005 us | -0.00% | SAME |
| I64 | I32 | I32 | 2^16 | 2^1 | 11.887 us | 7.86% | 11.367 us | 3.83% | -0.520 us | -4.38% | FAST |
| I64 | I32 | I32 | 2^20 | 2^1 | 17.941 us | 5.12% | 17.503 us | 2.41% | -0.438 us | -2.44% | FAST |
| I64 | I32 | I32 | 2^24 | 2^1 | 104.278 us | 1.48% | 103.065 us | 1.08% | -1.213 us | -1.16% | FAST |
| I64 | I32 | I32 | 2^28 | 2^1 | 1.459 ms | 0.19% | 1.458 ms | 0.20% | -0.151 us | -0.01% | SAME |
| I64 | I32 | I32 | 2^16 | 2^4 | 11.448 us | 5.12% | 11.426 us | 4.21% | -0.022 us | -0.19% | SAME |
| I64 | I32 | I32 | 2^20 | 2^4 | 16.583 us | 7.47% | 16.455 us | 6.69% | -0.128 us | -0.77% | SAME |
| I64 | I32 | I32 | 2^24 | 2^4 | 81.196 us | 1.46% | 80.769 us | 1.12% | -0.427 us | -0.53% | SAME |
| I64 | I32 | I32 | 2^28 | 2^4 | 1.091 ms | 0.17% | 1.091 ms | 0.19% | -0.301 us | -0.03% | SAME |
| I64 | I32 | I32 | 2^16 | 2^8 | 11.877 us | 7.84% | 11.417 us | 4.18% | -0.460 us | -3.87% | SAME |
| I64 | I32 | I32 | 2^20 | 2^8 | 16.170 us | 7.47% | 15.815 us | 5.36% | -0.356 us | -2.20% | SAME |
| I64 | I32 | I32 | 2^24 | 2^8 | 79.204 us | 2.02% | 77.978 us | 1.54% | -1.225 us | -1.55% | FAST |
| I64 | I32 | I32 | 2^28 | 2^8 | 1.042 ms | 0.15% | 1.041 ms | 0.18% | -0.296 us | -0.03% | SAME |
| I64 | I32 | I64 | 2^16 | 2^1 | 13.952 us | 8.09% | 13.444 us | 5.26% | -0.508 us | -3.64% | SAME |
| I64 | I32 | I64 | 2^20 | 2^1 | 20.127 us | 5.49% | 19.689 us | 3.36% | -0.438 us | -2.18% | SAME |
| I64 | I32 | I64 | 2^24 | 2^1 | 121.090 us | 1.14% | 120.124 us | 0.95% | -0.966 us | -0.80% | SAME |
| I64 | I32 | I64 | 2^28 | 2^1 | 1.717 ms | 0.17% | 1.717 ms | 0.16% | 0.485 us | 0.03% | SAME |
| I64 | I32 | I64 | 2^16 | 2^4 | 13.350 us | 8.26% | 13.005 us | 6.67% | -0.345 us | -2.58% | SAME |
| I64 | I32 | I64 | 2^20 | 2^4 | 19.494 us | 7.24% | 18.948 us | 5.55% | -0.546 us | -2.80% | SAME |
| I64 | I32 | I64 | 2^24 | 2^4 | 107.303 us | 1.63% | 106.130 us | 1.36% | -1.173 us | -1.09% | SAME |
| I64 | I32 | I64 | 2^28 | 2^4 | 1.479 ms | 0.28% | 1.479 ms | 0.28% | 0.135 us | 0.01% | SAME |
| I64 | I32 | I64 | 2^16 | 2^8 | 13.584 us | 10.14% | 13.095 us | 6.95% | -0.489 us | -3.60% | SAME |
| I64 | I32 | I64 | 2^20 | 2^8 | 19.152 us | 7.49% | 18.621 us | 6.27% | -0.531 us | -2.77% | SAME |
| I64 | I32 | I64 | 2^24 | 2^8 | 103.452 us | 1.48% | 102.538 us | 1.40% | -0.914 us | -0.88% | SAME |
| I64 | I32 | I64 | 2^28 | 2^8 | 1.410 ms | 0.24% | 1.410 ms | 0.25% | -0.207 us | -0.01% | SAME |
| I64 | I64 | I32 | 2^16 | 2^1 | 11.565 us | 6.23% | 11.379 us | 3.82% | -0.186 us | -1.60% | SAME |
| I64 | I64 | I32 | 2^20 | 2^1 | 18.047 us | 5.28% | 17.534 us | 2.72% | -0.513 us | -2.84% | FAST |
| I64 | I64 | I32 | 2^24 | 2^1 | 104.948 us | 1.34% | 104.036 us | 1.01% | -0.912 us | -0.87% | SAME |
| I64 | I64 | I32 | 2^28 | 2^1 | 1.472 ms | 0.17% | 1.472 ms | 0.18% | 0.013 us | 0.00% | SAME |
| I64 | I64 | I32 | 2^16 | 2^4 | 11.884 us | 7.79% | 11.400 us | 3.81% | -0.484 us | -4.07% | FAST |
| I64 | I64 | I32 | 2^20 | 2^4 | 16.651 us | 8.83% | 16.179 us | 6.46% | -0.472 us | -2.84% | SAME |
| I64 | I64 | I32 | 2^24 | 2^4 | 82.197 us | 1.44% | 81.267 us | 0.97% | -0.930 us | -1.13% | FAST |
| I64 | I64 | I32 | 2^28 | 2^4 | 1.103 ms | 0.16% | 1.104 ms | 0.15% | 0.094 us | 0.01% | SAME |
| I64 | I64 | I32 | 2^16 | 2^8 | 11.869 us | 7.83% | 11.336 us | 3.12% | -0.533 us | -4.49% | FAST |
| I64 | I64 | I32 | 2^20 | 2^8 | 16.178 us | 7.24% | 15.626 us | 4.47% | -0.552 us | -3.41% | SAME |
| I64 | I64 | I32 | 2^24 | 2^8 | 78.772 us | 1.94% | 77.610 us | 1.42% | -1.162 us | -1.48% | FAST |
| I64 | I64 | I32 | 2^28 | 2^8 | 1.039 ms | 0.16% | 1.039 ms | 0.18% | -0.599 us | -0.06% | SAME |
| I64 | I64 | I64 | 2^16 | 2^1 | 13.744 us | 9.68% | 13.186 us | 6.55% | -0.559 us | -4.07% | SAME |
| I64 | I64 | I64 | 2^20 | 2^1 | 20.105 us | 4.89% | 19.616 us | 3.09% | -0.489 us | -2.43% | SAME |
| I64 | I64 | I64 | 2^24 | 2^1 | 131.443 us | 1.14% | 130.315 us | 1.00% | -1.129 us | -0.86% | SAME |
| I64 | I64 | I64 | 2^28 | 2^1 | 1.894 ms | 0.20% | 1.894 ms | 0.19% | 0.161 us | 0.01% | SAME |
| I64 | I64 | I64 | 2^16 | 2^4 | 13.392 us | 10.84% | 12.806 us | 8.45% | -0.586 us | -4.38% | SAME |
| I64 | I64 | I64 | 2^20 | 2^4 | 18.751 us | 7.63% | 18.371 us | 6.00% | -0.380 us | -2.03% | SAME |
| I64 | I64 | I64 | 2^24 | 2^4 | 116.530 us | 1.49% | 115.455 us | 1.28% | -1.075 us | -0.92% | SAME |
| I64 | I64 | I64 | 2^28 | 2^4 | 1.641 ms | 0.24% | 1.641 ms | 0.26% | 0.226 us | 0.01% | SAME |
| I64 | I64 | I64 | 2^16 | 2^8 | 13.526 us | 10.11% | 12.901 us | 8.12% | -0.625 us | -4.62% | SAME |
| I64 | I64 | I64 | 2^20 | 2^8 | 18.348 us | 6.73% | 18.062 us | 5.40% | -0.286 us | -1.56% | SAME |
| I64 | I64 | I64 | 2^24 | 2^8 | 111.825 us | 1.53% | 110.849 us | 1.32% | -0.976 us | -0.87% | SAME |
| I64 | I64 | I64 | 2^28 | 2^8 | 1.556 ms | 0.27% | 1.556 ms | 0.26% | -0.129 us | -0.01% | SAME |
| I128 | I32 | I32 | 2^16 | 2^1 | 14.408 us | 9.37% | 13.906 us | 6.94% | -0.502 us | -3.48% | SAME |
| I128 | I32 | I32 | 2^20 | 2^1 | 23.757 us | 5.65% | 23.283 us | 3.71% | -0.475 us | -2.00% | SAME |
| I128 | I32 | I32 | 2^24 | 2^1 | 164.236 us | 0.78% | 163.102 us | 0.52% | -1.133 us | -0.69% | FAST |
| I128 | I32 | I32 | 2^28 | 2^1 | 2.402 ms | 0.07% | 2.402 ms | 0.07% | 0.081 us | 0.00% | SAME |
| I128 | I32 | I32 | 2^16 | 2^4 | 13.958 us | 7.13% | 13.555 us | 4.64% | -0.403 us | -2.89% | SAME |
| I128 | I32 | I32 | 2^20 | 2^4 | 20.545 us | 6.64% | 19.945 us | 4.48% | -0.600 us | -2.92% | SAME |
| I128 | I32 | I32 | 2^24 | 2^4 | 121.223 us | 1.05% | 120.365 us | 0.83% | -0.858 us | -0.71% | SAME |
| I128 | I32 | I32 | 2^28 | 2^4 | 1.707 ms | 0.10% | 1.707 ms | 0.10% | 0.164 us | 0.01% | SAME |
| I128 | I32 | I32 | 2^16 | 2^8 | 13.918 us | 7.16% | 13.429 us | 4.44% | -0.489 us | -3.51% | SAME |
| I128 | I32 | I32 | 2^20 | 2^8 | 20.172 us | 4.92% | 19.640 us | 2.72% | -0.532 us | -2.64% | SAME |
| I128 | I32 | I32 | 2^24 | 2^8 | 115.667 us | 1.43% | 114.486 us | 0.94% | -1.182 us | -1.02% | FAST |
| I128 | I32 | I32 | 2^28 | 2^8 | 1.611 ms | 0.10% | 1.611 ms | 0.11% | -0.075 us | -0.00% | SAME |
| I128 | I32 | I64 | 2^16 | 2^1 | 15.031 us | 9.79% | 14.357 us | 8.08% | -0.674 us | -4.48% | SAME |
| I128 | I32 | I64 | 2^20 | 2^1 | 30.460 us | 3.53% | 30.007 us | 2.22% | -0.453 us | -1.49% | SAME |
| I128 | I32 | I64 | 2^24 | 2^1 | 286.711 us | 0.44% | 285.639 us | 0.57% | -1.072 us | -0.37% | SAME |
| I128 | I32 | I64 | 2^28 | 2^1 | 4.363 ms | 0.10% | 4.362 ms | 0.10% | -1.590 us | -0.04% | SAME |
| I128 | I32 | I64 | 2^16 | 2^4 | 14.753 us | 9.84% | 14.091 us | 7.71% | -0.662 us | -4.49% | SAME |
| I128 | I32 | I64 | 2^20 | 2^4 | 28.489 us | 4.17% | 27.877 us | 2.60% | -0.612 us | -2.15% | SAME |
| I128 | I32 | I64 | 2^24 | 2^4 | 260.310 us | 0.69% | 259.403 us | 0.82% | -0.907 us | -0.35% | SAME |
| I128 | I32 | I64 | 2^28 | 2^4 | 3.929 ms | 0.16% | 3.929 ms | 0.16% | -0.161 us | -0.00% | SAME |
| I128 | I32 | I64 | 2^16 | 2^8 | 14.042 us | 7.48% | 14.069 us | 7.63% | 0.027 us | 0.19% | SAME |
| I128 | I32 | I64 | 2^20 | 2^8 | 27.879 us | 2.78% | 27.842 us | 2.52% | -0.036 us | -0.13% | SAME |
| I128 | I32 | I64 | 2^24 | 2^8 | 254.133 us | 0.71% | 254.278 us | 0.75% | 0.145 us | 0.06% | SAME |
| I128 | I32 | I64 | 2^28 | 2^8 | 3.845 ms | 0.15% | 3.845 ms | 0.15% | -0.031 us | -0.00% | SAME |
| I128 | I64 | I32 | 2^16 | 2^1 | 13.639 us | 5.65% | 13.570 us | 5.03% | -0.069 us | -0.50% | SAME |
| I128 | I64 | I32 | 2^20 | 2^1 | 23.414 us | 3.66% | 23.370 us | 3.61% | -0.044 us | -0.19% | SAME |
| I128 | I64 | I32 | 2^24 | 2^1 | 163.992 us | 0.68% | 164.152 us | 0.73% | 0.160 us | 0.10% | SAME |
| I128 | I64 | I32 | 2^28 | 2^1 | 2.420 ms | 0.08% | 2.420 ms | 0.08% | -0.028 us | -0.00% | SAME |
| I128 | I64 | I32 | 2^16 | 2^4 | 13.579 us | 4.60% | 13.498 us | 4.27% | -0.080 us | -0.59% | SAME |
| I128 | I64 | I32 | 2^20 | 2^4 | 19.794 us | 3.28% | 19.803 us | 3.29% | 0.009 us | 0.05% | SAME |
| I128 | I64 | I32 | 2^24 | 2^4 | 121.403 us | 0.93% | 121.479 us | 0.99% | 0.076 us | 0.06% | SAME |
| I128 | I64 | I32 | 2^28 | 2^4 | 1.725 ms | 0.11% | 1.725 ms | 0.12% | 0.022 us | 0.00% | SAME |
| I128 | I64 | I32 | 2^16 | 2^8 | 13.400 us | 4.00% | 13.309 us | 5.60% | -0.090 us | -0.68% | SAME |
| I128 | I64 | I32 | 2^20 | 2^8 | 19.529 us | 1.63% | 19.547 us | 1.87% | 0.017 us | 0.09% | SAME |
| I128 | I64 | I32 | 2^24 | 2^8 | 113.787 us | 0.72% | 113.813 us | 0.65% | 0.026 us | 0.02% | SAME |
| I128 | I64 | I32 | 2^28 | 2^8 | 1.602 ms | 0.10% | 1.602 ms | 0.11% | -0.060 us | -0.00% | SAME |
| I128 | I64 | I64 | 2^16 | 2^1 | 14.214 us | 8.00% | 13.994 us | 7.12% | -0.220 us | -1.55% | SAME |
| I128 | I64 | I64 | 2^20 | 2^1 | 29.963 us | 2.23% | 30.010 us | 2.34% | 0.047 us | 0.16% | SAME |
| I128 | I64 | I64 | 2^24 | 2^1 | 297.619 us | 0.62% | 297.631 us | 0.63% | 0.011 us | 0.00% | SAME |
| I128 | I64 | I64 | 2^28 | 2^1 | 4.571 ms | 0.12% | 4.571 ms | 0.12% | -0.006 us | -0.00% | SAME |
| I128 | I64 | I64 | 2^16 | 2^4 | 13.989 us | 7.24% | 13.881 us | 6.99% | -0.109 us | -0.78% | SAME |
| I128 | I64 | I64 | 2^20 | 2^4 | 28.373 us | 3.56% | 28.413 us | 3.78% | 0.039 us | 0.14% | SAME |
| I128 | I64 | I64 | 2^24 | 2^4 | 269.877 us | 0.70% | 270.065 us | 0.73% | 0.188 us | 0.07% | SAME |
| I128 | I64 | I64 | 2^28 | 2^4 | 4.126 ms | 0.15% | 4.127 ms | 0.16% | 0.947 us | 0.02% | SAME |
| I128 | I64 | I64 | 2^16 | 2^8 | 14.057 us | 7.67% | 13.956 us | 7.62% | -0.102 us | -0.72% | SAME |
| I128 | I64 | I64 | 2^20 | 2^8 | 28.059 us | 3.29% | 28.138 us | 3.20% | 0.078 us | 0.28% | SAME |
| I128 | I64 | I64 | 2^24 | 2^8 | 264.801 us | 0.75% | 264.775 us | 0.78% | -0.026 us | -0.01% | SAME |
| I128 | I64 | I64 | 2^28 | 2^8 | 4.030 ms | 0.16% | 4.030 ms | 0.15% | -0.442 us | -0.01% | SAME |
| F32 | I32 | I32 | 2^16 | 2^1 | 11.321 us | 2.77% | 11.362 us | 3.68% | 0.040 us | 0.36% | SAME |
| F32 | I32 | I32 | 2^20 | 2^1 | 15.437 us | 2.25% | 15.406 us | 1.92% | -0.031 us | -0.20% | SAME |
| F32 | I32 | I32 | 2^24 | 2^1 | 66.540 us | 1.27% | 66.593 us | 1.14% | 0.053 us | 0.08% | SAME |
| F32 | I32 | I32 | 2^28 | 2^1 | 713.251 us | 0.26% | 713.394 us | 0.27% | 0.142 us | 0.02% | SAME |
| F32 | I32 | I32 | 2^16 | 2^4 | 11.371 us | 3.52% | 11.406 us | 3.91% | 0.035 us | 0.30% | SAME |
| F32 | I32 | I32 | 2^20 | 2^4 | 14.557 us | 7.73% | 14.651 us | 7.66% | 0.094 us | 0.65% | SAME |
| F32 | I32 | I32 | 2^24 | 2^4 | 54.541 us | 1.36% | 54.531 us | 1.30% | -0.010 us | -0.02% | SAME |
| F32 | I32 | I32 | 2^28 | 2^4 | 690.972 us | 0.30% | 690.980 us | 0.29% | 0.008 us | 0.00% | SAME |
| F32 | I32 | I32 | 2^16 | 2^8 | 11.373 us | 3.31% | 11.358 us | 3.20% | -0.015 us | -0.13% | SAME |
| F32 | I32 | I32 | 2^20 | 2^8 | 14.708 us | 7.29% | 14.807 us | 7.53% | 0.100 us | 0.68% | SAME |
| F32 | I32 | I32 | 2^24 | 2^8 | 52.878 us | 1.87% | 53.001 us | 1.99% | 0.123 us | 0.23% | SAME |
| F32 | I32 | I32 | 2^28 | 2^8 | 662.018 us | 0.27% | 662.101 us | 0.26% | 0.083 us | 0.01% | SAME |
| F32 | I32 | I64 | 2^16 | 2^1 | 13.593 us | 5.01% | 13.525 us | 4.31% | -0.068 us | -0.50% | SAME |
| F32 | I32 | I64 | 2^20 | 2^1 | 19.187 us | 5.68% | 19.152 us | 5.92% | -0.035 us | -0.18% | SAME |
| F32 | I32 | I64 | 2^24 | 2^1 | 115.529 us | 0.92% | 115.576 us | 1.00% | 0.048 us | 0.04% | SAME |
| F32 | I32 | I64 | 2^28 | 2^1 | 1.399 ms | 0.19% | 1.399 ms | 0.21% | -0.019 us | -0.00% | SAME |
| F32 | I32 | I64 | 2^16 | 2^4 | 13.388 us | 3.04% | 13.443 us | 3.75% | 0.055 us | 0.41% | SAME |
| F32 | I32 | I64 | 2^20 | 2^4 | 17.581 us | 3.15% | 17.647 us | 3.91% | 0.066 us | 0.37% | SAME |
| F32 | I32 | I64 | 2^24 | 2^4 | 97.948 us | 1.22% | 97.996 us | 1.25% | 0.048 us | 0.05% | SAME |
| F32 | I32 | I64 | 2^28 | 2^4 | 1.365 ms | 0.23% | 1.365 ms | 0.22% | 0.195 us | 0.01% | SAME |
| F32 | I32 | I64 | 2^16 | 2^8 | 13.390 us | 4.86% | 13.396 us | 4.53% | 0.006 us | 0.04% | SAME |
| F32 | I32 | I64 | 2^20 | 2^8 | 17.663 us | 3.69% | 17.667 us | 3.54% | 0.004 us | 0.02% | SAME |
| F32 | I32 | I64 | 2^24 | 2^8 | 95.266 us | 1.28% | 95.364 us | 1.43% | 0.099 us | 0.10% | SAME |
| F32 | I32 | I64 | 2^28 | 2^8 | 1.323 ms | 0.23% | 1.323 ms | 0.22% | -0.009 us | -0.00% | SAME |
| F32 | I64 | I32 | 2^16 | 2^1 | 11.385 us | 3.58% | 11.388 us | 3.88% | 0.003 us | 0.02% | SAME |
| F32 | I64 | I32 | 2^20 | 2^1 | 15.201 us | 5.26% | 15.285 us | 5.59% | 0.084 us | 0.55% | SAME |
| F32 | I64 | I32 | 2^24 | 2^1 | 72.412 us | 1.46% | 72.583 us | 1.38% | 0.171 us | 0.24% | SAME |
| F32 | I64 | I32 | 2^28 | 2^1 | 809.182 us | 0.25% | 809.067 us | 0.26% | -0.115 us | -0.01% | SAME |
| F32 | I64 | I32 | 2^16 | 2^4 | 11.389 us | 3.50% | 11.407 us | 4.18% | 0.017 us | 0.15% | SAME |
| F32 | I64 | I32 | 2^20 | 2^4 | 14.957 us | 7.25% | 14.951 us | 7.37% | -0.007 us | -0.04% | SAME |
| F32 | I64 | I32 | 2^24 | 2^4 | 59.630 us | 2.03% | 59.638 us | 2.10% | 0.008 us | 0.01% | SAME |
| F32 | I64 | I32 | 2^28 | 2^4 | 784.961 us | 0.26% | 784.989 us | 0.26% | 0.028 us | 0.00% | SAME |
| F32 | I64 | I32 | 2^16 | 2^8 | 11.353 us | 3.31% | 11.306 us | 2.64% | -0.048 us | -0.42% | SAME |
| F32 | I64 | I32 | 2^20 | 2^8 | 14.522 us | 7.71% | 14.559 us | 7.58% | 0.037 us | 0.26% | SAME |
| F32 | I64 | I32 | 2^24 | 2^8 | 57.358 us | 1.91% | 57.459 us | 1.99% | 0.101 us | 0.18% | SAME |
| F32 | I64 | I32 | 2^28 | 2^8 | 744.046 us | 0.27% | 744.183 us | 0.26% | 0.137 us | 0.02% | SAME |
| F32 | I64 | I64 | 2^16 | 2^1 | 13.209 us | 6.24% | 13.326 us | 4.71% | 0.117 us | 0.89% | SAME |
| F32 | I64 | I64 | 2^20 | 2^1 | 17.991 us | 5.39% | 18.090 us | 5.42% | 0.099 us | 0.55% | SAME |
| F32 | I64 | I64 | 2^24 | 2^1 | 117.384 us | 0.90% | 117.479 us | 1.04% | 0.095 us | 0.08% | SAME |
| F32 | I64 | I64 | 2^28 | 2^1 | 1.439 ms | 0.22% | 1.439 ms | 0.22% | -0.118 us | -0.01% | SAME |
| F32 | I64 | I64 | 2^16 | 2^4 | 13.193 us | 6.28% | 13.124 us | 5.67% | -0.069 us | -0.52% | SAME |
| F32 | I64 | I64 | 2^20 | 2^4 | 17.497 us | 3.64% | 17.534 us | 3.93% | 0.037 us | 0.21% | SAME |
| F32 | I64 | I64 | 2^24 | 2^4 | 100.236 us | 1.24% | 100.263 us | 1.33% | 0.027 us | 0.03% | SAME |
| F32 | I64 | I64 | 2^28 | 2^4 | 1.406 ms | 0.25% | 1.406 ms | 0.24% | -0.364 us | -0.03% | SAME |
| F32 | I64 | I64 | 2^16 | 2^8 | 13.033 us | 7.29% | 13.078 us | 7.56% | 0.045 us | 0.34% | SAME |
| F32 | I64 | I64 | 2^20 | 2^8 | 17.490 us | 4.41% | 17.539 us | 4.37% | 0.048 us | 0.28% | SAME |
| F32 | I64 | I64 | 2^24 | 2^8 | 96.805 us | 1.47% | 96.815 us | 1.46% | 0.011 us | 0.01% | SAME |
| F32 | I64 | I64 | 2^28 | 2^8 | 1.352 ms | 0.24% | 1.352 ms | 0.25% | -0.043 us | -0.00% | SAME |
| F64 | I32 | I32 | 2^16 | 2^1 | 11.316 us | 2.76% | 11.321 us | 2.72% | 0.005 us | 0.04% | SAME |
| F64 | I32 | I32 | 2^20 | 2^1 | 17.563 us | 3.29% | 17.578 us | 3.38% | 0.015 us | 0.09% | SAME |
| F64 | I32 | I32 | 2^24 | 2^1 | 102.777 us | 1.20% | 102.869 us | 1.26% | 0.092 us | 0.09% | SAME |
| F64 | I32 | I32 | 2^28 | 2^1 | 1.452 ms | 0.21% | 1.452 ms | 0.20% | 0.204 us | 0.01% | SAME |
| F64 | I32 | I32 | 2^16 | 2^4 | 11.383 us | 3.77% | 11.414 us | 4.29% | 0.031 us | 0.27% | SAME |
| F64 | I32 | I32 | 2^20 | 2^4 | 16.110 us | 6.57% | 16.173 us | 6.52% | 0.062 us | 0.39% | SAME |
| F64 | I32 | I32 | 2^24 | 2^4 | 80.305 us | 1.53% | 80.387 us | 1.44% | 0.082 us | 0.10% | SAME |
| F64 | I32 | I32 | 2^28 | 2^4 | 1.084 ms | 0.17% | 1.084 ms | 0.16% | 0.016 us | 0.00% | SAME |
| F64 | I32 | I32 | 2^16 | 2^8 | 11.359 us | 3.15% | 11.368 us | 3.50% | 0.009 us | 0.08% | SAME |
| F64 | I32 | I32 | 2^20 | 2^8 | 15.799 us | 5.46% | 15.794 us | 5.21% | -0.005 us | -0.03% | SAME |
| F64 | I32 | I32 | 2^24 | 2^8 | 77.566 us | 1.31% | 77.693 us | 1.35% | 0.127 us | 0.16% | SAME |
| F64 | I32 | I32 | 2^28 | 2^8 | 1.037 ms | 0.17% | 1.037 ms | 0.17% | -0.125 us | -0.01% | SAME |
| F64 | I32 | I64 | 2^16 | 2^1 | 13.392 us | 4.18% | 13.491 us | 4.99% | 0.099 us | 0.74% | SAME |
| F64 | I32 | I64 | 2^20 | 2^1 | 19.648 us | 3.53% | 19.563 us | 3.92% | -0.085 us | -0.43% | SAME |
| F64 | I32 | I64 | 2^24 | 2^1 | 123.576 us | 1.06% | 123.567 us | 1.10% | -0.009 us | -0.01% | SAME |
| F64 | I32 | I64 | 2^28 | 2^1 | 1.784 ms | 0.22% | 1.784 ms | 0.22% | 0.392 us | 0.02% | SAME |
| F64 | I32 | I64 | 2^16 | 2^4 | 13.111 us | 6.78% | 13.169 us | 6.71% | 0.058 us | 0.44% | SAME |
| F64 | I32 | I64 | 2^20 | 2^4 | 18.237 us | 5.81% | 18.157 us | 5.57% | -0.080 us | -0.44% | SAME |
| F64 | I32 | I64 | 2^24 | 2^4 | 109.000 us | 1.40% | 109.146 us | 1.29% | 0.146 us | 0.13% | SAME |
| F64 | I32 | I64 | 2^28 | 2^4 | 1.537 ms | 0.31% | 1.537 ms | 0.32% | -0.123 us | -0.01% | SAME |
| F64 | I32 | I64 | 2^16 | 2^8 | 13.152 us | 6.99% | 13.242 us | 6.14% | 0.090 us | 0.69% | SAME |
| F64 | I32 | I64 | 2^20 | 2^8 | 18.064 us | 5.53% | 18.015 us | 5.22% | -0.049 us | -0.27% | SAME |
| F64 | I32 | I64 | 2^24 | 2^8 | 105.020 us | 1.44% | 104.972 us | 1.44% | -0.048 us | -0.05% | SAME |
| F64 | I32 | I64 | 2^28 | 2^8 | 1.454 ms | 0.30% | 1.454 ms | 0.29% | -0.204 us | -0.01% | SAME |
| F64 | I64 | I32 | 2^16 | 2^1 | 11.415 us | 3.95% | 11.354 us | 3.22% | -0.061 us | -0.53% | SAME |
| F64 | I64 | I32 | 2^20 | 2^1 | 17.500 us | 2.28% | 17.569 us | 2.87% | 0.069 us | 0.39% | SAME |
| F64 | I64 | I32 | 2^24 | 2^1 | 103.457 us | 1.08% | 103.366 us | 1.12% | -0.091 us | -0.09% | SAME |
| F64 | I64 | I32 | 2^28 | 2^1 | 1.463 ms | 0.19% | 1.463 ms | 0.19% | -0.047 us | -0.00% | SAME |
| F64 | I64 | I32 | 2^16 | 2^4 | 11.420 us | 4.10% | 11.410 us | 3.79% | -0.010 us | -0.09% | SAME |
| F64 | I64 | I32 | 2^20 | 2^4 | 15.991 us | 5.62% | 15.995 us | 5.71% | 0.004 us | 0.02% | SAME |
| F64 | I64 | I32 | 2^24 | 2^4 | 80.813 us | 1.20% | 80.788 us | 1.10% | -0.025 us | -0.03% | SAME |
| F64 | I64 | I32 | 2^28 | 2^4 | 1.091 ms | 0.18% | 1.091 ms | 0.18% | 0.027 us | 0.00% | SAME |
| F64 | I64 | I32 | 2^16 | 2^8 | 11.391 us | 3.95% | 11.397 us | 4.41% | 0.006 us | 0.05% | SAME |
| F64 | I64 | I32 | 2^20 | 2^8 | 15.518 us | 3.33% | 15.524 us | 3.25% | 0.007 us | 0.04% | SAME |
| F64 | I64 | I32 | 2^24 | 2^8 | 76.947 us | 0.81% | 77.055 us | 0.90% | 0.108 us | 0.14% | SAME |
| F64 | I64 | I32 | 2^28 | 2^8 | 1.027 ms | 0.16% | 1.027 ms | 0.17% | -0.080 us | -0.01% | SAME |
| F64 | I64 | I64 | 2^16 | 2^1 | 13.360 us | 5.74% | 13.308 us | 4.35% | -0.052 us | -0.39% | SAME |
| F64 | I64 | I64 | 2^20 | 2^1 | 19.534 us | 2.88% | 19.605 us | 3.12% | 0.070 us | 0.36% | SAME |
| F64 | I64 | I64 | 2^24 | 2^1 | 130.207 us | 0.98% | 130.267 us | 0.97% | 0.061 us | 0.05% | SAME |
| F64 | I64 | I64 | 2^28 | 2^1 | 1.890 ms | 0.18% | 1.890 ms | 0.20% | 0.427 us | 0.02% | SAME |
| F64 | I64 | I64 | 2^16 | 2^4 | 12.879 us | 8.74% | 12.820 us | 8.23% | -0.059 us | -0.46% | SAME |
| F64 | I64 | I64 | 2^20 | 2^4 | 18.055 us | 5.39% | 18.067 us | 5.31% | 0.012 us | 0.07% | SAME |
| F64 | I64 | I64 | 2^24 | 2^4 | 115.113 us | 1.27% | 114.978 us | 1.22% | -0.135 us | -0.12% | SAME |
| F64 | I64 | I64 | 2^28 | 2^4 | 1.632 ms | 0.26% | 1.632 ms | 0.27% | -0.278 us | -0.02% | SAME |
| F64 | I64 | I64 | 2^16 | 2^8 | 12.951 us | 9.37% | 12.875 us | 8.86% | -0.076 us | -0.58% | SAME |
| F64 | I64 | I64 | 2^20 | 2^8 | 17.862 us | 4.39% | 17.814 us | 4.12% | -0.048 us | -0.27% | SAME |
| F64 | I64 | I64 | 2^24 | 2^8 | 110.200 us | 1.44% | 110.213 us | 1.41% | 0.013 us | 0.01% | SAME |
| F64 | I64 | I64 | 2^28 | 2^8 | 1.546 ms | 0.29% | 1.545 ms | 0.26% | -0.154 us | -0.01% | SAME |
| C64 | I32 | I32 | 2^16 | 2^1 | 13.357 us | 2.96% | 13.364 us | 2.91% | 0.006 us | 0.05% | SAME |
| C64 | I32 | I32 | 2^20 | 2^1 | 19.269 us | 4.50% | 19.328 us | 4.13% | 0.059 us | 0.31% | SAME |
| C64 | I32 | I32 | 2^24 | 2^1 | 121.824 us | 0.92% | 121.828 us | 1.04% | 0.004 us | 0.00% | SAME |
| C64 | I32 | I32 | 2^28 | 2^1 | 1.511 ms | 0.28% | 1.511 ms | 0.27% | 0.198 us | 0.01% | SAME |
| C64 | I32 | I32 | 2^16 | 2^4 | 13.306 us | 4.98% | 13.341 us | 5.16% | 0.034 us | 0.26% | SAME |
| C64 | I32 | I32 | 2^20 | 2^4 | 17.916 us | 4.99% | 18.018 us | 5.30% | 0.102 us | 0.57% | SAME |
| C64 | I32 | I32 | 2^24 | 2^4 | 106.512 us | 1.37% | 106.466 us | 1.34% | -0.046 us | -0.04% | SAME |
| C64 | I32 | I32 | 2^28 | 2^4 | 1.478 ms | 0.30% | 1.478 ms | 0.32% | 0.003 us | 0.00% | SAME |
| C64 | I32 | I32 | 2^16 | 2^8 | 13.288 us | 5.37% | 13.344 us | 5.42% | 0.056 us | 0.42% | SAME |
| C64 | I32 | I32 | 2^20 | 2^8 | 17.768 us | 3.95% | 17.909 us | 4.99% | 0.141 us | 0.79% | SAME |
| C64 | I32 | I32 | 2^24 | 2^8 | 103.128 us | 1.48% | 103.150 us | 1.42% | 0.022 us | 0.02% | SAME |
| C64 | I32 | I32 | 2^28 | 2^8 | 1.427 ms | 0.30% | 1.427 ms | 0.30% | 0.240 us | 0.02% | SAME |
| C64 | I32 | I64 | 2^16 | 2^1 | 13.558 us | 5.11% | 13.515 us | 4.96% | -0.043 us | -0.32% | SAME |
| C64 | I32 | I64 | 2^20 | 2^1 | 19.723 us | 3.22% | 19.742 us | 3.12% | 0.019 us | 0.09% | SAME |
| C64 | I32 | I64 | 2^24 | 2^1 | 132.359 us | 0.75% | 132.395 us | 0.70% | 0.035 us | 0.03% | SAME |
| C64 | I32 | I64 | 2^28 | 2^1 | 1.609 ms | 0.25% | 1.609 ms | 0.24% | 0.190 us | 0.01% | SAME |
| C64 | I32 | I64 | 2^16 | 2^4 | 13.209 us | 5.60% | 13.287 us | 5.59% | 0.078 us | 0.59% | SAME |
| C64 | I32 | I64 | 2^20 | 2^4 | 18.616 us | 6.16% | 18.554 us | 6.02% | -0.062 us | -0.33% | SAME |
| C64 | I32 | I64 | 2^24 | 2^4 | 111.960 us | 1.25% | 111.814 us | 1.16% | -0.146 us | -0.13% | SAME |
| C64 | I32 | I64 | 2^28 | 2^4 | 1.568 ms | 0.28% | 1.568 ms | 0.26% | -0.356 us | -0.02% | SAME |
| C64 | I32 | I64 | 2^16 | 2^8 | 13.234 us | 5.86% | 13.316 us | 5.65% | 0.082 us | 0.62% | SAME |
| C64 | I32 | I64 | 2^20 | 2^8 | 18.406 us | 5.91% | 18.390 us | 6.00% | -0.017 us | -0.09% | SAME |
| C64 | I32 | I64 | 2^24 | 2^8 | 107.946 us | 1.26% | 107.965 us | 1.28% | 0.019 us | 0.02% | SAME |
| C64 | I32 | I64 | 2^28 | 2^8 | 1.504 ms | 0.25% | 1.504 ms | 0.25% | -0.231 us | -0.02% | SAME |
| C64 | I64 | I32 | 2^16 | 2^1 | 13.225 us | 6.08% | 13.270 us | 5.96% | 0.045 us | 0.34% | SAME |
| C64 | I64 | I32 | 2^20 | 2^1 | 19.232 us | 5.13% | 19.239 us | 4.96% | 0.007 us | 0.03% | SAME |
| C64 | I64 | I32 | 2^24 | 2^1 | 127.363 us | 1.04% | 127.221 us | 1.02% | -0.141 us | -0.11% | SAME |
| C64 | I64 | I32 | 2^28 | 2^1 | 1.581 ms | 0.31% | 1.581 ms | 0.31% | -0.164 us | -0.01% | SAME |
| C64 | I64 | I32 | 2^16 | 2^4 | 13.027 us | 7.75% | 12.980 us | 7.54% | -0.048 us | -0.37% | SAME |
| C64 | I64 | I32 | 2^20 | 2^4 | 17.824 us | 4.50% | 17.784 us | 4.11% | -0.040 us | -0.23% | SAME |
| C64 | I64 | I32 | 2^24 | 2^4 | 110.466 us | 1.52% | 110.380 us | 1.51% | -0.086 us | -0.08% | SAME |
| C64 | I64 | I32 | 2^28 | 2^4 | 1.545 ms | 0.31% | 1.545 ms | 0.33% | 0.078 us | 0.01% | SAME |
| C64 | I64 | I32 | 2^16 | 2^8 | 12.806 us | 7.68% | 12.853 us | 7.33% | 0.048 us | 0.37% | SAME |
| C64 | I64 | I32 | 2^20 | 2^8 | 17.560 us | 3.08% | 17.558 us | 2.89% | -0.002 us | -0.01% | SAME |
| C64 | I64 | I32 | 2^24 | 2^8 | 106.164 us | 1.56% | 106.201 us | 1.54% | 0.037 us | 0.03% | SAME |
| C64 | I64 | I32 | 2^28 | 2^8 | 1.486 ms | 0.32% | 1.486 ms | 0.32% | 0.016 us | 0.00% | SAME |
| C64 | I64 | I64 | 2^16 | 2^1 | 13.269 us | 5.29% | 13.249 us | 5.71% | -0.021 us | -0.15% | SAME |
| C64 | I64 | I64 | 2^20 | 2^1 | 19.601 us | 2.43% | 19.612 us | 2.61% | 0.011 us | 0.06% | SAME |
| C64 | I64 | I64 | 2^24 | 2^1 | 138.302 us | 0.69% | 138.319 us | 0.77% | 0.017 us | 0.01% | SAME |
| C64 | I64 | I64 | 2^28 | 2^1 | 1.695 ms | 0.22% | 1.695 ms | 0.23% | 0.313 us | 0.02% | SAME |
| C64 | I64 | I64 | 2^16 | 2^4 | 13.034 us | 6.85% | 12.894 us | 7.45% | -0.140 us | -1.08% | SAME |
| C64 | I64 | I64 | 2^20 | 2^4 | 18.395 us | 6.08% | 18.411 us | 6.08% | 0.015 us | 0.08% | SAME |
| C64 | I64 | I64 | 2^24 | 2^4 | 117.101 us | 1.24% | 116.979 us | 1.17% | -0.121 us | -0.10% | SAME |
| C64 | I64 | I64 | 2^28 | 2^4 | 1.652 ms | 0.25% | 1.652 ms | 0.25% | 0.162 us | 0.01% | SAME |
| C64 | I64 | I64 | 2^16 | 2^8 | 12.921 us | 8.21% | 12.881 us | 7.59% | -0.040 us | -0.31% | SAME |
| C64 | I64 | I64 | 2^20 | 2^8 | 17.986 us | 5.03% | 18.113 us | 5.36% | 0.127 us | 0.70% | SAME |
| C64 | I64 | I64 | 2^24 | 2^8 | 112.151 us | 1.29% | 112.341 us | 1.22% | 0.190 us | 0.17% | SAME |
| C64 | I64 | I64 | 2^28 | 2^8 | 1.582 ms | 0.25% | 1.581 ms | 0.25% | -0.302 us | -0.02% | SAME |
😬 CI Workflow Results
🟥 Finished in 4h 59m: Pass: 70%/177 | Total: 6d 03h | Max: 4h 58m | Hits: 84%/269746
See results here.
😬 CI Workflow Results
🟥 Finished in 4h 37m: Pass: 70%/177 | Total: 6d 15h | Max: 4h 36m | Hits: 80%/265427
See results here.
😬 CI Workflow Results
🟥 Finished in 5h 29m: Pass: 89%/177 | Total: 9d 11h | Max: 5h 27m | Hits: 68%/327697
See results here.
😬 CI Workflow Results
🟥 Finished in 5h 04m: Pass: 91%/177 | Total: 10d 03h | Max: 5h 03m | Hits: 40%/337427
See results here.
😬 CI Workflow Results
🟥 Finished in 5h 02m: Pass: 90%/177 | Total: 8d 17h | Max: 5h 01m | Hits: 58%/333392
See results here.
😬 CI Workflow Results
🟥 Finished in 5h 14m: Pass: 89%/175 | Total: 9d 14h | Max: 5h 12m | Hits: 52%/328324
See results here.