cuda-cccl 0.3.0__cp312-cp312-manylinux_2_24_aarch64.whl → 0.3.2__cp312-cp312-manylinux_2_24_aarch64.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Potentially problematic release.


This version of cuda-cccl might be problematic. Click here for more details.

Files changed (294) hide show
  1. cuda/cccl/cooperative/__init__.py +7 -1
  2. cuda/cccl/cooperative/experimental/__init__.py +21 -5
  3. cuda/cccl/headers/include/cub/agent/agent_adjacent_difference.cuh +2 -5
  4. cuda/cccl/headers/include/cub/agent/agent_batch_memcpy.cuh +2 -5
  5. cuda/cccl/headers/include/cub/agent/agent_for.cuh +2 -5
  6. cuda/cccl/headers/include/cub/agent/agent_histogram.cuh +354 -572
  7. cuda/cccl/headers/include/cub/agent/agent_merge.cuh +23 -21
  8. cuda/cccl/headers/include/cub/agent/agent_merge_sort.cuh +21 -3
  9. cuda/cccl/headers/include/cub/agent/agent_radix_sort_downsweep.cuh +2 -5
  10. cuda/cccl/headers/include/cub/agent/agent_radix_sort_histogram.cuh +2 -5
  11. cuda/cccl/headers/include/cub/agent/agent_radix_sort_onesweep.cuh +2 -5
  12. cuda/cccl/headers/include/cub/agent/agent_radix_sort_upsweep.cuh +2 -5
  13. cuda/cccl/headers/include/cub/agent/agent_rle.cuh +2 -5
  14. cuda/cccl/headers/include/cub/agent/agent_scan.cuh +5 -1
  15. cuda/cccl/headers/include/cub/agent/agent_scan_by_key.cuh +2 -5
  16. cuda/cccl/headers/include/cub/agent/agent_segmented_radix_sort.cuh +2 -5
  17. cuda/cccl/headers/include/cub/agent/agent_select_if.cuh +2 -5
  18. cuda/cccl/headers/include/cub/agent/agent_sub_warp_merge_sort.cuh +2 -5
  19. cuda/cccl/headers/include/cub/agent/agent_three_way_partition.cuh +2 -5
  20. cuda/cccl/headers/include/cub/agent/agent_unique_by_key.cuh +22 -5
  21. cuda/cccl/headers/include/cub/block/block_adjacent_difference.cuh +6 -8
  22. cuda/cccl/headers/include/cub/block/block_discontinuity.cuh +24 -14
  23. cuda/cccl/headers/include/cub/block/block_exchange.cuh +5 -0
  24. cuda/cccl/headers/include/cub/block/block_histogram.cuh +4 -0
  25. cuda/cccl/headers/include/cub/block/block_load.cuh +4 -0
  26. cuda/cccl/headers/include/cub/block/block_radix_rank.cuh +4 -2
  27. cuda/cccl/headers/include/cub/block/block_radix_sort.cuh +4 -2
  28. cuda/cccl/headers/include/cub/block/block_reduce.cuh +1 -0
  29. cuda/cccl/headers/include/cub/block/block_scan.cuh +12 -2
  30. cuda/cccl/headers/include/cub/block/block_store.cuh +3 -2
  31. cuda/cccl/headers/include/cub/detail/device_memory_resource.cuh +1 -0
  32. cuda/cccl/headers/include/cub/detail/mdspan_utils.cuh +34 -30
  33. cuda/cccl/headers/include/cub/detail/ptx-json-parser.h +1 -1
  34. cuda/cccl/headers/include/cub/device/device_for.cuh +118 -40
  35. cuda/cccl/headers/include/cub/device/device_reduce.cuh +6 -7
  36. cuda/cccl/headers/include/cub/device/device_segmented_reduce.cuh +170 -260
  37. cuda/cccl/headers/include/cub/device/device_transform.cuh +122 -91
  38. cuda/cccl/headers/include/cub/device/dispatch/dispatch_merge.cuh +6 -7
  39. cuda/cccl/headers/include/cub/device/dispatch/dispatch_radix_sort.cuh +2 -11
  40. cuda/cccl/headers/include/cub/device/dispatch/dispatch_reduce.cuh +12 -29
  41. cuda/cccl/headers/include/cub/device/dispatch/dispatch_reduce_deterministic.cuh +2 -7
  42. cuda/cccl/headers/include/cub/device/dispatch/dispatch_reduce_nondeterministic.cuh +0 -1
  43. cuda/cccl/headers/include/cub/device/dispatch/dispatch_segmented_sort.cuh +2 -3
  44. cuda/cccl/headers/include/cub/device/dispatch/dispatch_streaming_reduce.cuh +4 -5
  45. cuda/cccl/headers/include/cub/device/dispatch/dispatch_streaming_reduce_by_key.cuh +0 -1
  46. cuda/cccl/headers/include/cub/device/dispatch/dispatch_topk.cuh +3 -5
  47. cuda/cccl/headers/include/cub/device/dispatch/dispatch_transform.cuh +13 -5
  48. cuda/cccl/headers/include/cub/device/dispatch/kernels/for_each.cuh +72 -37
  49. cuda/cccl/headers/include/cub/device/dispatch/kernels/reduce.cuh +2 -5
  50. cuda/cccl/headers/include/cub/device/dispatch/kernels/scan.cuh +2 -5
  51. cuda/cccl/headers/include/cub/device/dispatch/kernels/segmented_reduce.cuh +2 -5
  52. cuda/cccl/headers/include/cub/device/dispatch/kernels/transform.cuh +22 -27
  53. cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_adjacent_difference.cuh +2 -5
  54. cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_batch_memcpy.cuh +2 -5
  55. cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_for.cuh +2 -5
  56. cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_histogram.cuh +2 -5
  57. cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_merge.cuh +2 -5
  58. cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_merge_sort.cuh +8 -0
  59. cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_radix_sort.cuh +2 -5
  60. cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_reduce_by_key.cuh +2 -5
  61. cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_run_length_encode.cuh +2 -5
  62. cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_scan.cuh +2 -5
  63. cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_scan_by_key.cuh +2 -5
  64. cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_segmented_sort.cuh +2 -5
  65. cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_three_way_partition.cuh +2 -5
  66. cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_transform.cuh +61 -70
  67. cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_unique_by_key.cuh +10 -0
  68. cuda/cccl/headers/include/cub/thread/thread_reduce.cuh +24 -17
  69. cuda/cccl/headers/include/cub/warp/specializations/warp_reduce_shfl.cuh +3 -2
  70. cuda/cccl/headers/include/cub/warp/specializations/warp_reduce_smem.cuh +3 -2
  71. cuda/cccl/headers/include/cub/warp/specializations/warp_scan_shfl.cuh +2 -2
  72. cuda/cccl/headers/include/cub/warp/warp_load.cuh +6 -6
  73. cuda/cccl/headers/include/cub/warp/warp_reduce.cuh +7 -2
  74. cuda/cccl/headers/include/cub/warp/warp_scan.cuh +7 -3
  75. cuda/cccl/headers/include/cub/warp/warp_store.cuh +1 -0
  76. cuda/cccl/headers/include/cuda/__algorithm/common.h +1 -1
  77. cuda/cccl/headers/include/cuda/__algorithm/copy.h +1 -1
  78. cuda/cccl/headers/include/cuda/__algorithm/fill.h +1 -1
  79. cuda/cccl/headers/include/cuda/__barrier/barrier_block_scope.h +19 -0
  80. cuda/cccl/headers/include/cuda/__cccl_config +1 -0
  81. cuda/cccl/headers/include/cuda/__cmath/fast_modulo_division.h +3 -74
  82. cuda/cccl/headers/include/cuda/__cmath/mul_hi.h +146 -0
  83. cuda/cccl/headers/include/cuda/__complex/get_real_imag.h +0 -4
  84. cuda/cccl/headers/include/cuda/__device/all_devices.h +46 -143
  85. cuda/cccl/headers/include/cuda/__device/arch_id.h +176 -0
  86. cuda/cccl/headers/include/cuda/__device/arch_traits.h +247 -323
  87. cuda/cccl/headers/include/cuda/__device/attributes.h +174 -123
  88. cuda/cccl/headers/include/cuda/__device/compute_capability.h +171 -0
  89. cuda/cccl/headers/include/cuda/__device/device_ref.h +27 -49
  90. cuda/cccl/headers/include/cuda/__device/physical_device.h +100 -96
  91. cuda/cccl/headers/include/cuda/__driver/driver_api.h +105 -3
  92. cuda/cccl/headers/include/cuda/__event/event.h +27 -26
  93. cuda/cccl/headers/include/cuda/__event/event_ref.h +5 -5
  94. cuda/cccl/headers/include/cuda/__event/timed_event.h +10 -7
  95. cuda/cccl/headers/include/cuda/__fwd/devices.h +44 -0
  96. cuda/cccl/headers/include/cuda/__fwd/zip_iterator.h +9 -0
  97. cuda/cccl/headers/include/cuda/__iterator/constant_iterator.h +46 -31
  98. cuda/cccl/headers/include/cuda/__iterator/strided_iterator.h +79 -47
  99. cuda/cccl/headers/include/cuda/__iterator/tabulate_output_iterator.h +59 -36
  100. cuda/cccl/headers/include/cuda/__iterator/transform_input_output_iterator.h +79 -49
  101. cuda/cccl/headers/include/cuda/__iterator/transform_iterator.h +74 -48
  102. cuda/cccl/headers/include/cuda/__iterator/transform_output_iterator.h +80 -55
  103. cuda/cccl/headers/include/cuda/__iterator/zip_common.h +148 -0
  104. cuda/cccl/headers/include/cuda/__iterator/zip_iterator.h +21 -137
  105. cuda/cccl/headers/include/cuda/__iterator/zip_transform_iterator.h +592 -0
  106. cuda/cccl/headers/include/cuda/__mdspan/host_device_accessor.h +127 -60
  107. cuda/cccl/headers/include/cuda/__mdspan/host_device_mdspan.h +178 -3
  108. cuda/cccl/headers/include/cuda/__mdspan/restrict_accessor.h +38 -8
  109. cuda/cccl/headers/include/cuda/__mdspan/restrict_mdspan.h +67 -1
  110. cuda/cccl/headers/include/cuda/__memory/ptr_in_range.h +93 -0
  111. cuda/cccl/headers/include/cuda/__memory_resource/get_memory_resource.h +4 -4
  112. cuda/cccl/headers/include/cuda/__memory_resource/properties.h +44 -0
  113. cuda/cccl/headers/include/cuda/__memory_resource/resource.h +1 -1
  114. cuda/cccl/headers/include/cuda/__memory_resource/resource_ref.h +4 -6
  115. cuda/cccl/headers/include/cuda/__nvtx/nvtx3.h +2 -1
  116. cuda/cccl/headers/include/cuda/__runtime/ensure_current_context.h +9 -7
  117. cuda/cccl/headers/include/cuda/__stream/stream.h +8 -8
  118. cuda/cccl/headers/include/cuda/__stream/stream_ref.h +18 -16
  119. cuda/cccl/headers/include/cuda/__utility/basic_any.h +1 -1
  120. cuda/cccl/headers/include/cuda/__utility/in_range.h +65 -0
  121. cuda/cccl/headers/include/cuda/algorithm +1 -1
  122. cuda/cccl/headers/include/cuda/cmath +1 -0
  123. cuda/cccl/headers/include/cuda/devices +13 -0
  124. cuda/cccl/headers/include/cuda/iterator +1 -0
  125. cuda/cccl/headers/include/cuda/memory +1 -0
  126. cuda/cccl/headers/include/cuda/std/__algorithm/equal_range.h +2 -2
  127. cuda/cccl/headers/include/cuda/std/__algorithm/find.h +1 -1
  128. cuda/cccl/headers/include/cuda/std/__algorithm/includes.h +2 -4
  129. cuda/cccl/headers/include/cuda/std/__algorithm/lower_bound.h +1 -1
  130. cuda/cccl/headers/include/cuda/std/__algorithm/make_projected.h +7 -15
  131. cuda/cccl/headers/include/cuda/std/__algorithm/min_element.h +1 -1
  132. cuda/cccl/headers/include/cuda/std/__algorithm/minmax_element.h +1 -2
  133. cuda/cccl/headers/include/cuda/std/__algorithm/partial_sort_copy.h +2 -2
  134. cuda/cccl/headers/include/cuda/std/__algorithm/upper_bound.h +1 -1
  135. cuda/cccl/headers/include/cuda/std/__bit/countl.h +8 -1
  136. cuda/cccl/headers/include/cuda/std/__bit/countr.h +2 -2
  137. cuda/cccl/headers/include/cuda/std/__bit/reference.h +11 -11
  138. cuda/cccl/headers/include/cuda/std/__cccl/algorithm_wrapper.h +36 -0
  139. cuda/cccl/headers/include/cuda/std/__cccl/builtin.h +46 -49
  140. cuda/cccl/headers/include/cuda/std/__cccl/execution_space.h +6 -0
  141. cuda/cccl/headers/include/cuda/std/__cccl/host_std_lib.h +52 -0
  142. cuda/cccl/headers/include/cuda/std/__cccl/memory_wrapper.h +36 -0
  143. cuda/cccl/headers/include/cuda/std/__cccl/numeric_wrapper.h +36 -0
  144. cuda/cccl/headers/include/cuda/std/__chrono/duration.h +16 -16
  145. cuda/cccl/headers/include/cuda/std/__chrono/steady_clock.h +5 -5
  146. cuda/cccl/headers/include/cuda/std/__chrono/system_clock.h +5 -5
  147. cuda/cccl/headers/include/cuda/std/__cmath/isnan.h +3 -2
  148. cuda/cccl/headers/include/cuda/std/__complex/complex.h +3 -2
  149. cuda/cccl/headers/include/cuda/std/__complex/literals.h +14 -34
  150. cuda/cccl/headers/include/cuda/std/__complex/nvbf16.h +2 -1
  151. cuda/cccl/headers/include/cuda/std/__complex/nvfp16.h +4 -3
  152. cuda/cccl/headers/include/cuda/std/__concepts/invocable.h +2 -2
  153. cuda/cccl/headers/include/cuda/std/__cstdlib/malloc.h +3 -2
  154. cuda/cccl/headers/include/cuda/std/__floating_point/fp.h +1 -1
  155. cuda/cccl/headers/include/cuda/std/__functional/bind.h +10 -13
  156. cuda/cccl/headers/include/cuda/std/__functional/function.h +5 -8
  157. cuda/cccl/headers/include/cuda/std/__functional/invoke.h +71 -335
  158. cuda/cccl/headers/include/cuda/std/__functional/mem_fn.h +1 -2
  159. cuda/cccl/headers/include/cuda/std/__functional/reference_wrapper.h +3 -3
  160. cuda/cccl/headers/include/cuda/std/__functional/weak_result_type.h +0 -6
  161. cuda/cccl/headers/include/cuda/std/__fwd/allocator.h +13 -0
  162. cuda/cccl/headers/include/cuda/std/__fwd/char_traits.h +13 -0
  163. cuda/cccl/headers/include/cuda/std/__fwd/complex.h +13 -4
  164. cuda/cccl/headers/include/cuda/std/__fwd/mdspan.h +23 -0
  165. cuda/cccl/headers/include/cuda/std/__fwd/pair.h +13 -0
  166. cuda/cccl/headers/include/cuda/std/__fwd/string.h +22 -0
  167. cuda/cccl/headers/include/cuda/std/__fwd/string_view.h +14 -0
  168. cuda/cccl/headers/include/cuda/std/__internal/features.h +0 -5
  169. cuda/cccl/headers/include/cuda/std/__internal/namespaces.h +21 -0
  170. cuda/cccl/headers/include/cuda/std/__iterator/iterator_traits.h +5 -5
  171. cuda/cccl/headers/include/cuda/std/__mdspan/extents.h +7 -1
  172. cuda/cccl/headers/include/cuda/std/__mdspan/mdspan.h +53 -39
  173. cuda/cccl/headers/include/cuda/std/__memory/allocator.h +3 -3
  174. cuda/cccl/headers/include/cuda/std/__memory/construct_at.h +1 -3
  175. cuda/cccl/headers/include/cuda/std/__optional/optional_base.h +1 -0
  176. cuda/cccl/headers/include/cuda/std/__ranges/compressed_movable_box.h +892 -0
  177. cuda/cccl/headers/include/cuda/std/__ranges/movable_box.h +2 -2
  178. cuda/cccl/headers/include/cuda/std/__tuple_dir/make_tuple_types.h +23 -1
  179. cuda/cccl/headers/include/cuda/std/__tuple_dir/tuple_like.h +4 -0
  180. cuda/cccl/headers/include/cuda/std/__tuple_dir/tuple_like_ext.h +4 -0
  181. cuda/cccl/headers/include/cuda/std/__type_traits/is_primary_template.h +7 -5
  182. cuda/cccl/headers/include/cuda/std/__type_traits/result_of.h +1 -1
  183. cuda/cccl/headers/include/cuda/std/__utility/pair.h +0 -5
  184. cuda/cccl/headers/include/cuda/std/bitset +1 -1
  185. cuda/cccl/headers/include/cuda/std/detail/libcxx/include/__config +15 -12
  186. cuda/cccl/headers/include/cuda/std/detail/libcxx/include/variant +11 -9
  187. cuda/cccl/headers/include/cuda/std/inplace_vector +4 -4
  188. cuda/cccl/headers/include/cuda/std/numbers +5 -0
  189. cuda/cccl/headers/include/cuda/std/string_view +155 -13
  190. cuda/cccl/headers/include/cuda/std/version +1 -4
  191. cuda/cccl/headers/include/cuda/stream_ref +5 -0
  192. cuda/cccl/headers/include/cuda/utility +1 -0
  193. cuda/cccl/headers/include/nv/target +7 -2
  194. cuda/cccl/headers/include/thrust/allocate_unique.h +1 -1
  195. cuda/cccl/headers/include/thrust/detail/allocator/allocator_traits.h +309 -33
  196. cuda/cccl/headers/include/thrust/detail/allocator/copy_construct_range.h +151 -4
  197. cuda/cccl/headers/include/thrust/detail/allocator/destroy_range.h +60 -3
  198. cuda/cccl/headers/include/thrust/detail/allocator/fill_construct_range.h +45 -3
  199. cuda/cccl/headers/include/thrust/detail/allocator/malloc_allocator.h +31 -6
  200. cuda/cccl/headers/include/thrust/detail/allocator/tagged_allocator.h +29 -16
  201. cuda/cccl/headers/include/thrust/detail/allocator/temporary_allocator.h +41 -4
  202. cuda/cccl/headers/include/thrust/detail/allocator/value_initialize_range.h +42 -4
  203. cuda/cccl/headers/include/thrust/detail/complex/ccosh.h +3 -3
  204. cuda/cccl/headers/include/thrust/detail/integer_math.h +3 -20
  205. cuda/cccl/headers/include/thrust/detail/internal_functional.h +1 -1
  206. cuda/cccl/headers/include/thrust/detail/memory_algorithms.h +1 -1
  207. cuda/cccl/headers/include/thrust/detail/temporary_array.h +1 -1
  208. cuda/cccl/headers/include/thrust/detail/type_traits.h +1 -1
  209. cuda/cccl/headers/include/thrust/device_delete.h +18 -3
  210. cuda/cccl/headers/include/thrust/device_free.h +16 -3
  211. cuda/cccl/headers/include/thrust/device_new.h +29 -8
  212. cuda/cccl/headers/include/thrust/host_vector.h +1 -1
  213. cuda/cccl/headers/include/thrust/iterator/iterator_traits.h +11 -0
  214. cuda/cccl/headers/include/thrust/iterator/tabulate_output_iterator.h +5 -2
  215. cuda/cccl/headers/include/thrust/mr/disjoint_pool.h +1 -1
  216. cuda/cccl/headers/include/thrust/mr/pool.h +1 -1
  217. cuda/cccl/headers/include/thrust/system/cuda/detail/copy.h +33 -0
  218. cuda/cccl/headers/include/thrust/system/cuda/detail/find.h +13 -115
  219. cuda/cccl/headers/include/thrust/system/cuda/detail/mismatch.h +8 -2
  220. cuda/cccl/headers/include/thrust/type_traits/is_contiguous_iterator.h +7 -7
  221. cuda/cccl/parallel/experimental/__init__.py +21 -74
  222. cuda/compute/__init__.py +79 -0
  223. cuda/{cccl/parallel/experimental → compute}/_bindings.pyi +43 -1
  224. cuda/{cccl/parallel/experimental → compute}/_bindings_impl.pyx +157 -8
  225. cuda/{cccl/parallel/experimental → compute}/algorithms/_histogram.py +2 -2
  226. cuda/{cccl/parallel/experimental → compute}/algorithms/_merge_sort.py +2 -2
  227. cuda/{cccl/parallel/experimental → compute}/algorithms/_radix_sort.py +3 -3
  228. cuda/{cccl/parallel/experimental → compute}/algorithms/_reduce.py +2 -2
  229. cuda/{cccl/parallel/experimental → compute}/algorithms/_scan.py +112 -40
  230. cuda/{cccl/parallel/experimental → compute}/algorithms/_segmented_reduce.py +2 -2
  231. cuda/{cccl/parallel/experimental → compute}/algorithms/_three_way_partition.py +2 -2
  232. cuda/{cccl/parallel/experimental → compute}/algorithms/_transform.py +36 -15
  233. cuda/{cccl/parallel/experimental → compute}/algorithms/_unique_by_key.py +2 -2
  234. cuda/compute/cu12/_bindings_impl.cpython-312-aarch64-linux-gnu.so +0 -0
  235. cuda/{cccl/parallel/experimental → compute}/cu12/cccl/libcccl.c.parallel.so +0 -0
  236. cuda/compute/cu13/_bindings_impl.cpython-312-aarch64-linux-gnu.so +0 -0
  237. cuda/{cccl/parallel/experimental → compute}/cu13/cccl/libcccl.c.parallel.so +0 -0
  238. cuda/{cccl/parallel/experimental → compute}/iterators/__init__.py +2 -0
  239. cuda/{cccl/parallel/experimental → compute}/iterators/_factories.py +36 -8
  240. cuda/{cccl/parallel/experimental → compute}/iterators/_iterators.py +206 -1
  241. cuda/{cccl/parallel/experimental → compute}/numba_utils.py +2 -2
  242. cuda/{cccl/parallel/experimental → compute}/struct.py +2 -2
  243. cuda/{cccl/parallel/experimental → compute}/typing.py +2 -0
  244. cuda/coop/__init__.py +8 -0
  245. cuda/{cccl/cooperative/experimental → coop}/_nvrtc.py +3 -2
  246. cuda/{cccl/cooperative/experimental → coop}/_scan_op.py +3 -3
  247. cuda/{cccl/cooperative/experimental → coop}/_types.py +2 -2
  248. cuda/{cccl/cooperative/experimental → coop}/_typing.py +1 -1
  249. cuda/{cccl/cooperative/experimental → coop}/block/__init__.py +6 -6
  250. cuda/{cccl/cooperative/experimental → coop}/block/_block_exchange.py +4 -4
  251. cuda/{cccl/cooperative/experimental → coop}/block/_block_load_store.py +6 -6
  252. cuda/{cccl/cooperative/experimental → coop}/block/_block_merge_sort.py +4 -4
  253. cuda/{cccl/cooperative/experimental → coop}/block/_block_radix_sort.py +6 -6
  254. cuda/{cccl/cooperative/experimental → coop}/block/_block_reduce.py +6 -6
  255. cuda/{cccl/cooperative/experimental → coop}/block/_block_scan.py +7 -7
  256. cuda/coop/warp/__init__.py +9 -0
  257. cuda/{cccl/cooperative/experimental → coop}/warp/_warp_merge_sort.py +3 -3
  258. cuda/{cccl/cooperative/experimental → coop}/warp/_warp_reduce.py +6 -6
  259. cuda/{cccl/cooperative/experimental → coop}/warp/_warp_scan.py +4 -4
  260. {cuda_cccl-0.3.0.dist-info → cuda_cccl-0.3.2.dist-info}/METADATA +1 -1
  261. {cuda_cccl-0.3.0.dist-info → cuda_cccl-0.3.2.dist-info}/RECORD +275 -276
  262. cuda/cccl/cooperative/experimental/warp/__init__.py +0 -9
  263. cuda/cccl/headers/include/cub/device/dispatch/dispatch_advance_iterators.cuh +0 -111
  264. cuda/cccl/headers/include/thrust/detail/algorithm_wrapper.h +0 -37
  265. cuda/cccl/headers/include/thrust/detail/allocator/allocator_traits.inl +0 -371
  266. cuda/cccl/headers/include/thrust/detail/allocator/copy_construct_range.inl +0 -242
  267. cuda/cccl/headers/include/thrust/detail/allocator/destroy_range.inl +0 -137
  268. cuda/cccl/headers/include/thrust/detail/allocator/fill_construct_range.inl +0 -99
  269. cuda/cccl/headers/include/thrust/detail/allocator/malloc_allocator.inl +0 -68
  270. cuda/cccl/headers/include/thrust/detail/allocator/tagged_allocator.inl +0 -86
  271. cuda/cccl/headers/include/thrust/detail/allocator/temporary_allocator.inl +0 -79
  272. cuda/cccl/headers/include/thrust/detail/allocator/value_initialize_range.inl +0 -98
  273. cuda/cccl/headers/include/thrust/detail/device_delete.inl +0 -52
  274. cuda/cccl/headers/include/thrust/detail/device_free.inl +0 -47
  275. cuda/cccl/headers/include/thrust/detail/device_new.inl +0 -61
  276. cuda/cccl/headers/include/thrust/detail/memory_wrapper.h +0 -40
  277. cuda/cccl/headers/include/thrust/detail/numeric_wrapper.h +0 -37
  278. cuda/cccl/parallel/experimental/.gitignore +0 -4
  279. cuda/cccl/parallel/experimental/cu12/_bindings_impl.cpython-312-aarch64-linux-gnu.so +0 -0
  280. cuda/cccl/parallel/experimental/cu13/_bindings_impl.cpython-312-aarch64-linux-gnu.so +0 -0
  281. /cuda/{cccl/parallel/experimental → compute}/_bindings.py +0 -0
  282. /cuda/{cccl/parallel/experimental → compute}/_caching.py +0 -0
  283. /cuda/{cccl/parallel/experimental → compute}/_cccl_interop.py +0 -0
  284. /cuda/{cccl/parallel/experimental → compute}/_utils/__init__.py +0 -0
  285. /cuda/{cccl/parallel/experimental → compute}/_utils/protocols.py +0 -0
  286. /cuda/{cccl/parallel/experimental → compute}/_utils/temp_storage_buffer.py +0 -0
  287. /cuda/{cccl/parallel/experimental → compute}/algorithms/__init__.py +0 -0
  288. /cuda/{cccl/parallel/experimental → compute}/cccl/.gitkeep +0 -0
  289. /cuda/{cccl/parallel/experimental → compute}/iterators/_zip_iterator.py +0 -0
  290. /cuda/{cccl/parallel/experimental → compute}/op.py +0 -0
  291. /cuda/{cccl/cooperative/experimental → coop}/_caching.py +0 -0
  292. /cuda/{cccl/cooperative/experimental → coop}/_common.py +0 -0
  293. {cuda_cccl-0.3.0.dist-info → cuda_cccl-0.3.2.dist-info}/WHEEL +0 -0
  294. {cuda_cccl-0.3.0.dist-info → cuda_cccl-0.3.2.dist-info}/licenses/LICENSE +0 -0
@@ -1,79 +0,0 @@
1
- /*
2
- * Copyright 2008-2013 NVIDIA Corporation
3
- *
4
- * Licensed under the Apache License, Version 2.0 (the "License");
5
- * you may not use this file except in compliance with the License.
6
- * You may obtain a copy of the License at
7
- *
8
- * http://www.apache.org/licenses/LICENSE-2.0
9
- *
10
- * Unless required by applicable law or agreed to in writing, software
11
- * distributed under the License is distributed on an "AS IS" BASIS,
12
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13
- * See the License for the specific language governing permissions and
14
- * limitations under the License.
15
- */
16
-
17
- #pragma once
18
-
19
- #include <thrust/detail/config.h>
20
-
21
- #if defined(_CCCL_IMPLICIT_SYSTEM_HEADER_GCC)
22
- # pragma GCC system_header
23
- #elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_CLANG)
24
- # pragma clang system_header
25
- #elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_MSVC)
26
- # pragma system_header
27
- #endif // no system header
28
- #include <thrust/detail/allocator/temporary_allocator.h>
29
- #include <thrust/detail/temporary_buffer.h>
30
- #include <thrust/system/detail/bad_alloc.h>
31
-
32
- #include <cuda/std/cassert>
33
-
34
- #include <nv/target>
35
-
36
- #if _CCCL_CUDA_COMPILATION() && _CCCL_DEVICE_COMPILATION()
37
- # include <thrust/system/cuda/detail/terminate.h>
38
- #endif // _CCCL_CUDA_COMPILATION() && _CCCL_DEVICE_COMPILATION()
39
-
40
- THRUST_NAMESPACE_BEGIN
41
- namespace detail
42
- {
43
-
44
- template <typename T, typename System>
45
- _CCCL_HOST_DEVICE typename temporary_allocator<T, System>::pointer
46
- temporary_allocator<T, System>::allocate(typename temporary_allocator<T, System>::size_type cnt)
47
- {
48
- pointer_and_size result = thrust::get_temporary_buffer<T>(system(), cnt);
49
-
50
- // handle failure
51
- if (result.second < cnt)
52
- {
53
- // deallocate and throw
54
- // note that we pass cnt to deallocate, not a value derived from result.second
55
- deallocate(result.first, cnt);
56
-
57
- #if _CCCL_CUDA_COMPILATION()
58
- NV_IF_TARGET(NV_IS_HOST,
59
- (throw thrust::system::detail::bad_alloc("temporary_buffer::allocate: get_temporary_buffer failed");),
60
- ( // NV_IS_DEVICE
61
- thrust::system::cuda::detail::terminate_with_message("temporary_buffer::allocate: "
62
- "get_temporary_buffer failed");));
63
- #else
64
- throw thrust::system::detail::bad_alloc("temporary_buffer::allocate: get_temporary_buffer failed");
65
- #endif
66
- } // end if
67
-
68
- return result.first;
69
- } // end temporary_allocator::allocate()
70
-
71
- template <typename T, typename System>
72
- _CCCL_HOST_DEVICE void temporary_allocator<T, System>::deallocate(
73
- typename temporary_allocator<T, System>::pointer p, typename temporary_allocator<T, System>::size_type n) noexcept
74
- {
75
- return thrust::return_temporary_buffer(system(), p, n);
76
- } // end temporary_allocator
77
-
78
- } // namespace detail
79
- THRUST_NAMESPACE_END
@@ -1,98 +0,0 @@
1
- /*
2
- * Copyright 2008-2013 NVIDIA Corporation
3
- *
4
- * Licensed under the Apache License, Version 2.0 (the "License");
5
- * you may not use this file except in compliance with the License.
6
- * You may obtain a copy of the License at
7
- *
8
- * http://www.apache.org/licenses/LICENSE-2.0
9
- *
10
- * Unless required by applicable law or agreed to in writing, software
11
- * distributed under the License is distributed on an "AS IS" BASIS,
12
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13
- * See the License for the specific language governing permissions and
14
- * limitations under the License.
15
- */
16
-
17
- #pragma once
18
-
19
- #include <thrust/detail/config.h>
20
-
21
- #if defined(_CCCL_IMPLICIT_SYSTEM_HEADER_GCC)
22
- # pragma GCC system_header
23
- #elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_CLANG)
24
- # pragma clang system_header
25
- #elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_MSVC)
26
- # pragma system_header
27
- #endif // no system header
28
- #include <thrust/detail/allocator/allocator_traits.h>
29
- #include <thrust/detail/type_traits.h>
30
- #include <thrust/detail/type_traits/pointer_traits.h>
31
- #include <thrust/for_each.h>
32
- #include <thrust/uninitialized_fill.h>
33
-
34
- THRUST_NAMESPACE_BEGIN
35
- namespace detail
36
- {
37
- namespace allocator_traits_detail
38
- {
39
-
40
- template <typename Allocator>
41
- struct construct1_via_allocator
42
- {
43
- Allocator& a;
44
-
45
- _CCCL_HOST_DEVICE construct1_via_allocator(Allocator& a)
46
- : a(a)
47
- {}
48
-
49
- template <typename T>
50
- inline _CCCL_HOST_DEVICE void operator()(T& x)
51
- {
52
- allocator_traits<Allocator>::construct(a, &x);
53
- }
54
- };
55
-
56
- // we need to construct T via the allocator if...
57
- template <typename Allocator, typename T>
58
- struct needs_default_construct_via_allocator
59
- : ::cuda::std::disjunction<has_member_construct1<Allocator, T>, // if the Allocator does something interesting
60
- // or if T's default constructor does something
61
- // interesting
62
- thrust::detail::not_<::cuda::std::is_trivially_default_constructible<T>>>
63
- {};
64
-
65
- // we know that std::allocator::construct's only effect is to call T's
66
- // default constructor, so we needn't use it for default construction
67
- // unless T's constructor does something interesting
68
- template <typename U, typename T>
69
- struct needs_default_construct_via_allocator<std::allocator<U>, T>
70
- : thrust::detail::not_<::cuda::std::is_trivially_default_constructible<T>>
71
- {};
72
-
73
- template <typename Allocator, typename Pointer, typename Size>
74
- _CCCL_HOST_DEVICE ::cuda::std::enable_if_t<
75
- needs_default_construct_via_allocator<Allocator, typename pointer_element<Pointer>::type>::value>
76
- value_initialize_range(Allocator& a, Pointer p, Size n)
77
- {
78
- thrust::for_each_n(allocator_system<Allocator>::get(a), p, n, construct1_via_allocator<Allocator>(a));
79
- }
80
-
81
- template <typename Allocator, typename Pointer, typename Size>
82
- _CCCL_HOST_DEVICE
83
- typename disable_if<needs_default_construct_via_allocator<Allocator, typename pointer_element<Pointer>::type>::value>::type
84
- value_initialize_range(Allocator& a, Pointer p, Size n)
85
- {
86
- thrust::uninitialized_fill_n(allocator_system<Allocator>::get(a), p, n, typename pointer_element<Pointer>::type());
87
- }
88
-
89
- } // namespace allocator_traits_detail
90
-
91
- template <typename Allocator, typename Pointer, typename Size>
92
- _CCCL_HOST_DEVICE void value_initialize_range(Allocator& a, Pointer p, Size n)
93
- {
94
- return allocator_traits_detail::value_initialize_range(a, p, n);
95
- }
96
-
97
- } // namespace detail
98
- THRUST_NAMESPACE_END
@@ -1,52 +0,0 @@
1
- /*
2
- * Copyright 2008-2013 NVIDIA Corporation
3
- *
4
- * Licensed under the Apache License, Version 2.0 (the "License");
5
- * you may not use this file except in compliance with the License.
6
- * You may obtain a copy of the License at
7
- *
8
- * http://www.apache.org/licenses/LICENSE-2.0
9
- *
10
- * Unless required by applicable law or agreed to in writing, software
11
- * distributed under the License is distributed on an "AS IS" BASIS,
12
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13
- * See the License for the specific language governing permissions and
14
- * limitations under the License.
15
- */
16
-
17
- #pragma once
18
-
19
- #include <thrust/detail/config.h>
20
-
21
- #if defined(_CCCL_IMPLICIT_SYSTEM_HEADER_GCC)
22
- # pragma GCC system_header
23
- #elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_CLANG)
24
- # pragma clang system_header
25
- #elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_MSVC)
26
- # pragma system_header
27
- #endif // no system header
28
- #include <thrust/detail/allocator/destroy_range.h>
29
- #include <thrust/device_delete.h>
30
- #include <thrust/device_free.h>
31
-
32
- THRUST_NAMESPACE_BEGIN
33
-
34
- namespace detail
35
- {
36
-
37
- // define an empty allocator class to use below
38
- struct device_delete_allocator
39
- {};
40
-
41
- } // namespace detail
42
-
43
- template <typename T>
44
- void device_delete(device_ptr<T> ptr, const size_t n)
45
- {
46
- // we can use device_allocator to destroy the range
47
- thrust::detail::device_delete_allocator a;
48
- thrust::detail::destroy_range(a, ptr, n);
49
- thrust::device_free(ptr);
50
- } // end device_delete()
51
-
52
- THRUST_NAMESPACE_END
@@ -1,47 +0,0 @@
1
- /*
2
- * Copyright 2008-2013 NVIDIA Corporation
3
- *
4
- * Licensed under the Apache License, Version 2.0 (the "License");
5
- * you may not use this file except in compliance with the License.
6
- * You may obtain a copy of the License at
7
- *
8
- * http://www.apache.org/licenses/LICENSE-2.0
9
- *
10
- * Unless required by applicable law or agreed to in writing, software
11
- * distributed under the License is distributed on an "AS IS" BASIS,
12
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13
- * See the License for the specific language governing permissions and
14
- * limitations under the License.
15
- */
16
-
17
- #pragma once
18
-
19
- #include <thrust/detail/config.h>
20
-
21
- #if defined(_CCCL_IMPLICIT_SYSTEM_HEADER_GCC)
22
- # pragma GCC system_header
23
- #elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_CLANG)
24
- # pragma clang system_header
25
- #elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_MSVC)
26
- # pragma system_header
27
- #endif // no system header
28
- #include <thrust/detail/malloc_and_free.h>
29
- #include <thrust/device_free.h>
30
- #include <thrust/iterator/iterator_traits.h>
31
- #include <thrust/system/detail/generic/select_system.h>
32
-
33
- THRUST_NAMESPACE_BEGIN
34
-
35
- void device_free(thrust::device_ptr<void> ptr)
36
- {
37
- using thrust::system::detail::generic::select_system;
38
-
39
- using system = thrust::iterator_system<thrust::device_ptr<void>>::type;
40
-
41
- // XXX lower to select_system(system) here
42
- system s;
43
-
44
- thrust::free(s, ptr);
45
- } // end device_free()
46
-
47
- THRUST_NAMESPACE_END
@@ -1,61 +0,0 @@
1
- /*
2
- * Copyright 2008-2013 NVIDIA Corporation
3
- *
4
- * Licensed under the Apache License, Version 2.0 (the "License");
5
- * you may not use this file except in compliance with the License.
6
- * You may obtain a copy of the License at
7
- *
8
- * http://www.apache.org/licenses/LICENSE-2.0
9
- *
10
- * Unless required by applicable law or agreed to in writing, software
11
- * distributed under the License is distributed on an "AS IS" BASIS,
12
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13
- * See the License for the specific language governing permissions and
14
- * limitations under the License.
15
- */
16
-
17
- #pragma once
18
-
19
- #include <thrust/detail/config.h>
20
-
21
- #if defined(_CCCL_IMPLICIT_SYSTEM_HEADER_GCC)
22
- # pragma GCC system_header
23
- #elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_CLANG)
24
- # pragma clang system_header
25
- #elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_MSVC)
26
- # pragma system_header
27
- #endif // no system header
28
- #include <thrust/device_malloc.h>
29
- #include <thrust/device_new.h>
30
- #include <thrust/uninitialized_fill.h>
31
-
32
- THRUST_NAMESPACE_BEGIN
33
-
34
- template <typename T>
35
- device_ptr<T> device_new(device_ptr<void> p, const size_t n)
36
- {
37
- // XXX TODO dispatch n null device constructors at p here
38
- // in the meantime, dispatch 1 null host constructor here
39
- // and dispatch n copy constructors
40
- return device_new<T>(p, T(), n);
41
- } // end device_new()
42
-
43
- template <typename T>
44
- device_ptr<T> device_new(device_ptr<void> p, const T& exemplar, const size_t n)
45
- {
46
- device_ptr<T> result(reinterpret_cast<T*>(p.get()));
47
-
48
- // run copy constructors at p here
49
- thrust::uninitialized_fill(result, result + n, exemplar);
50
-
51
- return result;
52
- } // end device_new()
53
-
54
- template <typename T>
55
- device_ptr<T> device_new(const size_t n)
56
- {
57
- // call placement new
58
- return device_new<T>(thrust::device_malloc<T>(n));
59
- } // end device_new()
60
-
61
- THRUST_NAMESPACE_END
@@ -1,40 +0,0 @@
1
- /*
2
- * Copyright 2020 NVIDIA Corporation
3
- *
4
- * Licensed under the Apache License, Version 2.0 (the "License");
5
- * you may not use this file except in compliance with the License.
6
- * You may obtain a copy of the License at
7
- *
8
- * http://www.apache.org/licenses/LICENSE-2.0
9
- *
10
- * Unless required by applicable law or agreed to in writing, software
11
- * distributed under the License is distributed on an "AS IS" BASIS,
12
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13
- * See the License for the specific language governing permissions and
14
- * limitations under the License.
15
- */
16
-
17
- #pragma once
18
-
19
- #include <thrust/detail/config.h>
20
-
21
- #if defined(_CCCL_IMPLICIT_SYSTEM_HEADER_GCC)
22
- # pragma GCC system_header
23
- #elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_CLANG)
24
- # pragma clang system_header
25
- #elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_MSVC)
26
- # pragma system_header
27
- #endif // no system header
28
-
29
- // When a compiler uses Thrust as part of its implementation of Standard C++
30
- // algorithms, a cycle of included files may result when Thrust code tries to
31
- // use a standard algorithm. Having a macro that is defined only when Thrust
32
- // is including an algorithms-related header gives the compiler a chance to
33
- // detect and break the cycle of includes. (<memory> declares several standard
34
- // algorithms, including all of the uninitialized_* algorithms. "_ALGORITHMS_"
35
- // in the macro name is meant generically, not as a specific reference to
36
- // the header <algorithms>.)
37
-
38
- #define THRUST_INCLUDING_ALGORITHMS_HEADER
39
- #include <memory>
40
- #undef THRUST_INCLUDING_ALGORITHMS_HEADER
@@ -1,37 +0,0 @@
1
- /*
2
- * Copyright 2021 NVIDIA Corporation
3
- *
4
- * Licensed under the Apache License, Version 2.0 (the "License");
5
- * you may not use this file except in compliance with the License.
6
- * You may obtain a copy of the License at
7
- *
8
- * http://www.apache.org/licenses/LICENSE-2.0
9
- *
10
- * Unless required by applicable law or agreed to in writing, software
11
- * distributed under the License is distributed on an "AS IS" BASIS,
12
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13
- * See the License for the specific language governing permissions and
14
- * limitations under the License.
15
- */
16
-
17
- #pragma once
18
-
19
- #include <thrust/detail/config.h>
20
-
21
- #if defined(_CCCL_IMPLICIT_SYSTEM_HEADER_GCC)
22
- # pragma GCC system_header
23
- #elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_CLANG)
24
- # pragma clang system_header
25
- #elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_MSVC)
26
- # pragma system_header
27
- #endif // no system header
28
-
29
- // When a compiler uses Thrust as part of its implementation of Standard C++
30
- // algorithms, a cycle of included files may result when Thrust code tries to
31
- // use a standard algorithm. Having a macro that is defined only when Thrust
32
- // is including an algorithms-related header gives the compiler a chance to
33
- // detect and break the cycle of includes.
34
-
35
- #define THRUST_INCLUDING_ALGORITHMS_HEADER
36
- #include <numeric>
37
- #undef THRUST_INCLUDING_ALGORITHMS_HEADER
@@ -1,4 +0,0 @@
1
- # these are generated at build time depending on the CUDA
2
- # version we're building for:
3
- cu12/
4
- cu13/
File without changes
File without changes