cuda-cccl 0.3.2__cp313-cp313-manylinux_2_24_aarch64.whl → 0.3.4__cp313-cp313-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.
- cuda/cccl/headers/include/cub/agent/agent_adjacent_difference.cuh +12 -38
- cuda/cccl/headers/include/cub/agent/agent_batch_memcpy.cuh +16 -40
- cuda/cccl/headers/include/cub/agent/agent_for.cuh +2 -28
- cuda/cccl/headers/include/cub/agent/agent_histogram.cuh +24 -56
- cuda/cccl/headers/include/cub/agent/agent_merge_sort.cuh +12 -38
- cuda/cccl/headers/include/cub/agent/agent_radix_sort_downsweep.cuh +31 -56
- cuda/cccl/headers/include/cub/agent/agent_radix_sort_histogram.cuh +31 -35
- cuda/cccl/headers/include/cub/agent/agent_radix_sort_onesweep.cuh +47 -48
- cuda/cccl/headers/include/cub/agent/agent_radix_sort_upsweep.cuh +39 -42
- cuda/cccl/headers/include/cub/agent/agent_reduce.cuh +33 -60
- cuda/cccl/headers/include/cub/agent/agent_reduce_by_key.cuh +18 -44
- cuda/cccl/headers/include/cub/agent/agent_rle.cuh +26 -55
- cuda/cccl/headers/include/cub/agent/agent_scan.cuh +22 -49
- cuda/cccl/headers/include/cub/agent/agent_scan_by_key.cuh +15 -41
- cuda/cccl/headers/include/cub/agent/agent_segmented_radix_sort.cuh +9 -35
- cuda/cccl/headers/include/cub/agent/agent_select_if.cuh +20 -49
- cuda/cccl/headers/include/cub/agent/agent_sub_warp_merge_sort.cuh +14 -40
- cuda/cccl/headers/include/cub/agent/agent_three_way_partition.cuh +18 -40
- cuda/cccl/headers/include/cub/agent/agent_topk.cuh +0 -2
- cuda/cccl/headers/include/cub/agent/agent_unique_by_key.cuh +20 -46
- cuda/cccl/headers/include/cub/agent/single_pass_scan_operators.cuh +3 -28
- cuda/cccl/headers/include/cub/block/block_adjacent_difference.cuh +7 -31
- cuda/cccl/headers/include/cub/block/block_discontinuity.cuh +10 -34
- cuda/cccl/headers/include/cub/block/block_exchange.cuh +120 -154
- cuda/cccl/headers/include/cub/block/block_histogram.cuh +28 -52
- cuda/cccl/headers/include/cub/block/block_load.cuh +124 -146
- cuda/cccl/headers/include/cub/block/block_load_to_shared.cuh +0 -16
- cuda/cccl/headers/include/cub/block/block_merge_sort.cuh +58 -87
- cuda/cccl/headers/include/cub/block/block_radix_rank.cuh +81 -100
- cuda/cccl/headers/include/cub/block/block_radix_sort.cuh +92 -156
- cuda/cccl/headers/include/cub/block/block_raking_layout.cuh +8 -32
- cuda/cccl/headers/include/cub/block/block_reduce.cuh +21 -46
- cuda/cccl/headers/include/cub/block/block_run_length_decode.cuh +51 -79
- cuda/cccl/headers/include/cub/block/block_scan.cuh +94 -401
- cuda/cccl/headers/include/cub/block/block_shuffle.cuh +10 -34
- cuda/cccl/headers/include/cub/block/block_store.cuh +73 -97
- cuda/cccl/headers/include/cub/block/radix_rank_sort_operations.cuh +2 -29
- cuda/cccl/headers/include/cub/block/specializations/block_histogram_atomic.cuh +5 -29
- cuda/cccl/headers/include/cub/block/specializations/block_histogram_sort.cuh +25 -49
- cuda/cccl/headers/include/cub/block/specializations/block_reduce_raking.cuh +12 -34
- cuda/cccl/headers/include/cub/block/specializations/block_reduce_raking_commutative_only.cuh +10 -34
- cuda/cccl/headers/include/cub/block/specializations/block_reduce_warp_reductions.cuh +3 -27
- cuda/cccl/headers/include/cub/block/specializations/block_scan_raking.cuh +12 -36
- cuda/cccl/headers/include/cub/block/specializations/block_scan_warp_scans.cuh +9 -33
- cuda/cccl/headers/include/cub/config.cuh +2 -26
- cuda/cccl/headers/include/cub/cub.cuh +3 -27
- cuda/cccl/headers/include/cub/detail/array_utils.cuh +2 -26
- cuda/cccl/headers/include/cub/detail/choose_offset.cuh +2 -28
- cuda/cccl/headers/include/cub/detail/detect_cuda_runtime.cuh +3 -27
- cuda/cccl/headers/include/cub/detail/device_double_buffer.cuh +0 -2
- cuda/cccl/headers/include/cub/detail/device_memory_resource.cuh +1 -3
- cuda/cccl/headers/include/cub/detail/fast_modulo_division.cuh +2 -28
- cuda/cccl/headers/include/cub/detail/integer_utils.cuh +0 -2
- cuda/cccl/headers/include/cub/detail/launcher/cuda_driver.cuh +0 -2
- cuda/cccl/headers/include/cub/detail/launcher/cuda_runtime.cuh +0 -2
- cuda/cccl/headers/include/cub/detail/mdspan_utils.cuh +0 -2
- cuda/cccl/headers/include/cub/detail/ptx-json/README.md +7 -12
- cuda/cccl/headers/include/cub/detail/ptx-json/array.h +6 -33
- cuda/cccl/headers/include/cub/detail/ptx-json/json.h +13 -36
- cuda/cccl/headers/include/cub/detail/ptx-json/object.h +9 -38
- cuda/cccl/headers/include/cub/detail/ptx-json/string.h +58 -32
- cuda/cccl/headers/include/cub/detail/ptx-json/value.h +51 -51
- cuda/cccl/headers/include/cub/detail/ptx-json-parser.h +7 -31
- cuda/cccl/headers/include/cub/detail/rfa.cuh +2 -27
- cuda/cccl/headers/include/cub/detail/strong_load.cuh +3 -29
- cuda/cccl/headers/include/cub/detail/strong_store.cuh +3 -29
- cuda/cccl/headers/include/cub/detail/temporary_storage.cuh +2 -9
- cuda/cccl/headers/include/cub/detail/type_traits.cuh +0 -2
- cuda/cccl/headers/include/cub/detail/uninitialized_copy.cuh +6 -31
- cuda/cccl/headers/include/cub/detail/unsafe_bitcast.cuh +2 -25
- cuda/cccl/headers/include/cub/device/device_adjacent_difference.cuh +2 -26
- cuda/cccl/headers/include/cub/device/device_for.cuh +3 -5
- cuda/cccl/headers/include/cub/device/device_histogram.cuh +3 -27
- cuda/cccl/headers/include/cub/device/device_memcpy.cuh +2 -26
- cuda/cccl/headers/include/cub/device/device_merge_sort.cuh +2 -26
- cuda/cccl/headers/include/cub/device/device_partition.cuh +3 -27
- cuda/cccl/headers/include/cub/device/device_radix_sort.cuh +3 -27
- cuda/cccl/headers/include/cub/device/device_reduce.cuh +10 -31
- cuda/cccl/headers/include/cub/device/device_run_length_encode.cuh +3 -27
- cuda/cccl/headers/include/cub/device/device_scan.cuh +16 -34
- cuda/cccl/headers/include/cub/device/device_segmented_radix_sort.cuh +3 -27
- cuda/cccl/headers/include/cub/device/device_segmented_reduce.cuh +3 -27
- cuda/cccl/headers/include/cub/device/device_segmented_sort.cuh +2 -26
- cuda/cccl/headers/include/cub/device/device_select.cuh +3 -27
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_adjacent_difference.cuh +2 -28
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_batch_memcpy.cuh +2 -27
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_copy_mdspan.cuh +0 -2
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_for.cuh +3 -29
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_histogram.cuh +14 -34
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_merge_sort.cuh +5 -30
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_radix_sort.cuh +4 -29
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_reduce.cuh +5 -32
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_reduce_by_key.cuh +3 -29
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_reduce_deterministic.cuh +2 -29
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_reduce_nondeterministic.cuh +1 -2
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_rle.cuh +47 -59
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_scan.cuh +21 -30
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_scan_by_key.cuh +2 -27
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_segmented_sort.cuh +3 -27
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_select_if.cuh +3 -27
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_streaming_reduce.cuh +0 -2
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_streaming_reduce_by_key.cuh +51 -36
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_three_way_partition.cuh +3 -28
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_topk.cuh +0 -1
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_transform.cuh +27 -55
- cuda/cccl/headers/include/cub/device/dispatch/dispatch_unique_by_key.cuh +4 -28
- cuda/cccl/headers/include/cub/device/dispatch/kernels/{for_each.cuh → kernel_for_each.cuh} +0 -2
- cuda/cccl/headers/include/cub/device/dispatch/kernels/{histogram.cuh → kernel_histogram.cuh} +149 -157
- cuda/cccl/headers/include/cub/device/dispatch/kernels/{merge_sort.cuh → kernel_merge_sort.cuh} +0 -2
- cuda/cccl/headers/include/cub/device/dispatch/kernels/{radix_sort.cuh → kernel_radix_sort.cuh} +0 -2
- cuda/cccl/headers/include/cub/device/dispatch/kernels/{reduce.cuh → kernel_reduce.cuh} +2 -28
- cuda/cccl/headers/include/cub/device/dispatch/kernels/{scan.cuh → kernel_scan.cuh} +2 -28
- cuda/cccl/headers/include/cub/device/dispatch/kernels/{segmented_reduce.cuh → kernel_segmented_reduce.cuh} +3 -29
- cuda/cccl/headers/include/cub/device/dispatch/kernels/{segmented_sort.cuh → kernel_segmented_sort.cuh} +0 -1
- cuda/cccl/headers/include/cub/device/dispatch/kernels/{three_way_partition.cuh → kernel_three_way_partition.cuh} +0 -1
- cuda/cccl/headers/include/cub/device/dispatch/kernels/{transform.cuh → kernel_transform.cuh} +11 -11
- cuda/cccl/headers/include/cub/device/dispatch/kernels/{unique_by_key.cuh → kernel_unique_by_key.cuh} +0 -1
- cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_adjacent_difference.cuh +2 -26
- cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_batch_memcpy.cuh +2 -26
- cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_for.cuh +2 -28
- cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_histogram.cuh +6 -26
- cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_merge.cuh +2 -26
- cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_merge_sort.cuh +5 -31
- cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_radix_sort.cuh +31 -33
- cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_reduce.cuh +15 -40
- cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_reduce_by_key.cuh +2 -26
- cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_run_length_encode.cuh +2 -28
- cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_scan.cuh +20 -44
- cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_scan_by_key.cuh +2 -26
- cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_segmented_sort.cuh +20 -45
- cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_select_if.cuh +2 -27
- cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_three_way_partition.cuh +11 -36
- cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_topk.cuh +0 -1
- cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_transform.cuh +2 -27
- cuda/cccl/headers/include/cub/device/dispatch/tuning/tuning_unique_by_key.cuh +14 -40
- cuda/cccl/headers/include/cub/grid/grid_even_share.cuh +3 -27
- cuda/cccl/headers/include/cub/grid/grid_mapping.cuh +3 -27
- cuda/cccl/headers/include/cub/grid/grid_queue.cuh +3 -27
- cuda/cccl/headers/include/cub/iterator/arg_index_input_iterator.cuh +3 -27
- cuda/cccl/headers/include/cub/iterator/cache_modified_input_iterator.cuh +3 -27
- cuda/cccl/headers/include/cub/iterator/cache_modified_output_iterator.cuh +3 -27
- cuda/cccl/headers/include/cub/iterator/tex_obj_input_iterator.cuh +3 -27
- cuda/cccl/headers/include/cub/thread/thread_load.cuh +3 -28
- cuda/cccl/headers/include/cub/thread/thread_operators.cuh +3 -27
- cuda/cccl/headers/include/cub/thread/thread_reduce.cuh +3 -26
- cuda/cccl/headers/include/cub/thread/thread_scan.cuh +3 -29
- cuda/cccl/headers/include/cub/thread/thread_search.cuh +3 -27
- cuda/cccl/headers/include/cub/thread/thread_simd.cuh +0 -2
- cuda/cccl/headers/include/cub/thread/thread_sort.cuh +2 -26
- cuda/cccl/headers/include/cub/thread/thread_store.cuh +3 -27
- cuda/cccl/headers/include/cub/util_allocator.cuh +3 -27
- cuda/cccl/headers/include/cub/util_arch.cuh +3 -29
- cuda/cccl/headers/include/cub/util_cpp_dialect.cuh +2 -26
- cuda/cccl/headers/include/cub/util_debug.cuh +3 -27
- cuda/cccl/headers/include/cub/util_device.cuh +18 -59
- cuda/cccl/headers/include/cub/util_macro.cuh +4 -28
- cuda/cccl/headers/include/cub/util_math.cuh +2 -28
- cuda/cccl/headers/include/cub/util_namespace.cuh +3 -28
- cuda/cccl/headers/include/cub/util_policy_wrapper_t.cuh +3 -27
- cuda/cccl/headers/include/cub/util_ptx.cuh +6 -30
- cuda/cccl/headers/include/cub/util_temporary_storage.cuh +3 -29
- cuda/cccl/headers/include/cub/util_type.cuh +5 -32
- cuda/cccl/headers/include/cub/util_vsmem.cuh +2 -28
- cuda/cccl/headers/include/cub/version.cuh +2 -26
- cuda/cccl/headers/include/cub/warp/specializations/warp_exchange_shfl.cuh +10 -35
- cuda/cccl/headers/include/cub/warp/specializations/warp_exchange_smem.cuh +5 -30
- cuda/cccl/headers/include/cub/warp/specializations/warp_reduce_shfl.cuh +15 -39
- cuda/cccl/headers/include/cub/warp/specializations/warp_reduce_smem.cuh +5 -35
- cuda/cccl/headers/include/cub/warp/specializations/warp_scan_shfl.cuh +22 -46
- cuda/cccl/headers/include/cub/warp/specializations/warp_scan_smem.cuh +3 -27
- cuda/cccl/headers/include/cub/warp/warp_exchange.cuh +2 -26
- cuda/cccl/headers/include/cub/warp/warp_load.cuh +4 -27
- cuda/cccl/headers/include/cub/warp/warp_merge_sort.cuh +2 -26
- cuda/cccl/headers/include/cub/warp/warp_reduce.cuh +3 -22
- cuda/cccl/headers/include/cub/warp/warp_scan.cuh +3 -27
- cuda/cccl/headers/include/cub/warp/warp_store.cuh +4 -27
- cuda/cccl/headers/include/cub/warp/warp_utils.cuh +0 -2
- cuda/cccl/headers/include/cuda/__barrier/barrier.h +1 -1
- cuda/cccl/headers/include/cuda/__barrier/barrier_arrive_tx.h +0 -1
- cuda/cccl/headers/include/cuda/__barrier/barrier_block_scope.h +277 -235
- cuda/cccl/headers/include/cuda/__barrier/barrier_expect_tx.h +0 -1
- cuda/cccl/headers/include/cuda/__driver/driver_api.h +13 -0
- cuda/cccl/headers/include/cuda/__execution/determinism.h +0 -2
- cuda/cccl/headers/include/cuda/__execution/output_ordering.h +0 -2
- cuda/cccl/headers/include/cuda/__functional/maximum.h +25 -7
- cuda/cccl/headers/include/cuda/__functional/minimum.h +25 -7
- cuda/cccl/headers/include/cuda/__functional/minimum_maximum_common.h +52 -0
- cuda/cccl/headers/include/cuda/__functional/proclaim_return_type.h +0 -2
- cuda/cccl/headers/include/cuda/__iterator/counting_iterator.h +13 -4
- cuda/cccl/headers/include/cuda/__iterator/zip_function.h +4 -2
- cuda/cccl/headers/include/cuda/__mdspan/host_device_accessor.h +0 -1
- cuda/cccl/headers/include/cuda/__memcpy_async/cp_async_bulk_shared_global.h +28 -7
- cuda/cccl/headers/include/cuda/__memcpy_async/dispatch_memcpy_async.h +1 -1
- cuda/cccl/headers/include/cuda/__memcpy_async/elect_one.h +52 -0
- cuda/cccl/headers/include/cuda/__memcpy_async/memcpy_async_tx.h +2 -3
- cuda/cccl/headers/include/cuda/__memcpy_async/memcpy_completion.h +1 -7
- cuda/cccl/headers/include/cuda/__memcpy_async/try_get_barrier_handle.h +0 -1
- cuda/cccl/headers/include/cuda/__memory/get_device_address.h +1 -1
- cuda/cccl/headers/include/cuda/__memory/ranges_overlap.h +126 -0
- cuda/cccl/headers/include/cuda/__memory_resource/any_resource.h +898 -0
- cuda/cccl/headers/include/cuda/__memory_resource/device_memory_pool.h +149 -0
- cuda/cccl/headers/include/cuda/__memory_resource/get_property.h +3 -3
- cuda/cccl/headers/include/cuda/__memory_resource/legacy_managed_memory_resource.h +148 -0
- cuda/cccl/headers/include/cuda/__memory_resource/legacy_pinned_memory_resource.h +139 -0
- cuda/cccl/headers/include/cuda/__memory_resource/managed_memory_pool.h +146 -0
- cuda/cccl/headers/include/cuda/__memory_resource/memory_resource_base.h +578 -0
- cuda/cccl/headers/include/cuda/__memory_resource/pinned_memory_pool.h +188 -0
- cuda/cccl/headers/include/cuda/__memory_resource/properties.h +3 -3
- cuda/cccl/headers/include/cuda/__memory_resource/resource.h +37 -3
- cuda/cccl/headers/include/cuda/__numeric/add_overflow.h +13 -3
- cuda/cccl/headers/include/cuda/__numeric/div_overflow.h +150 -0
- cuda/cccl/headers/include/cuda/__numeric/overflow_cast.h +2 -2
- cuda/cccl/headers/include/cuda/__numeric/sub_overflow.h +344 -0
- cuda/cccl/headers/include/cuda/__nvtx/nvtx3.h +0 -6
- cuda/cccl/headers/include/cuda/__ptx/instructions/shfl_sync.h +1 -1
- cuda/cccl/headers/include/cuda/__ptx/pragmas/enable_smem_spilling.h +47 -0
- cuda/cccl/headers/include/cuda/{std/__cuda → __runtime}/api_wrapper.h +3 -3
- cuda/cccl/headers/include/cuda/__stream/get_stream.h +0 -1
- cuda/cccl/headers/include/cuda/{__fwd/barrier_native_handle.h → __stream/internal_streams.h} +17 -15
- cuda/cccl/headers/include/cuda/__utility/__basic_any/basic_any_ptr.h +2 -2
- cuda/cccl/headers/include/cuda/__utility/__basic_any/basic_any_value.h +1 -0
- cuda/cccl/headers/include/cuda/__utility/__basic_any/semiregular.h +1 -0
- cuda/cccl/headers/include/cuda/__utility/__basic_any/virtcall.h +2 -1
- cuda/cccl/headers/include/cuda/barrier +42 -16
- cuda/cccl/headers/include/cuda/memory +1 -0
- cuda/cccl/headers/include/cuda/memory_resource +6 -1
- cuda/cccl/headers/include/cuda/numeric +2 -0
- cuda/cccl/headers/include/cuda/pipeline +3 -2
- cuda/cccl/headers/include/cuda/ptx +1 -0
- cuda/cccl/headers/include/cuda/std/__algorithm/unique_copy.h +0 -2
- cuda/cccl/headers/include/cuda/std/__atomic/api/reference.h +1 -1
- cuda/cccl/headers/include/cuda/std/__atomic/functions/cuda_ptx_derived.h +115 -58
- cuda/cccl/headers/include/cuda/std/__atomic/functions/cuda_ptx_generated.h +844 -378
- cuda/cccl/headers/include/cuda/std/__atomic/functions/cuda_ptx_generated_helper.h +12 -5
- cuda/cccl/headers/include/cuda/std/__atomic/functions/host.h +31 -0
- cuda/cccl/headers/include/cuda/std/__atomic/types/small.h +10 -0
- cuda/cccl/headers/include/cuda/std/__atomic/types.h +2 -3
- cuda/cccl/headers/include/cuda/std/__bit/byteswap.h +37 -13
- cuda/cccl/headers/include/cuda/std/__cccl/builtin.h +0 -28
- cuda/cccl/headers/include/cuda/std/__cccl/dialect.h +7 -0
- cuda/cccl/headers/include/cuda/std/__cccl/epilogue.h +10 -0
- cuda/cccl/headers/include/cuda/std/__cccl/exceptions.h +2 -45
- cuda/cccl/headers/include/cuda/std/__cccl/is_non_narrowing_convertible.h +0 -2
- cuda/cccl/headers/include/cuda/std/__cccl/prologue.h +8 -0
- cuda/cccl/headers/include/cuda/std/__chrono/calendar.h +0 -2
- cuda/cccl/headers/include/cuda/std/__chrono/day.h +0 -2
- cuda/cccl/headers/include/cuda/std/__chrono/duration.h +13 -17
- cuda/cccl/headers/include/cuda/std/__chrono/file_clock.h +0 -2
- cuda/cccl/headers/include/cuda/std/__chrono/high_resolution_clock.h +0 -2
- cuda/cccl/headers/include/cuda/std/__chrono/month.h +0 -2
- cuda/cccl/headers/include/cuda/std/__chrono/steady_clock.h +0 -2
- cuda/cccl/headers/include/cuda/std/__chrono/system_clock.h +0 -2
- cuda/cccl/headers/include/cuda/std/__chrono/time_point.h +5 -8
- cuda/cccl/headers/include/cuda/std/__chrono/year.h +0 -2
- cuda/cccl/headers/include/cuda/std/__cmath/error_functions.h +4 -0
- cuda/cccl/headers/include/cuda/std/__cmath/exponential_functions.h +2 -3
- cuda/cccl/headers/include/cuda/std/__cmath/fdim.h +4 -0
- cuda/cccl/headers/include/cuda/std/__cmath/fma.h +4 -0
- cuda/cccl/headers/include/cuda/std/__cmath/fpclassify.h +2 -3
- cuda/cccl/headers/include/cuda/std/__cmath/gamma.h +2 -3
- cuda/cccl/headers/include/cuda/std/__cmath/hyperbolic_functions.h +2 -3
- cuda/cccl/headers/include/cuda/std/__cmath/hypot.h +2 -3
- cuda/cccl/headers/include/cuda/std/__cmath/inverse_hyperbolic_functions.h +2 -3
- cuda/cccl/headers/include/cuda/std/__cmath/inverse_trigonometric_functions.h +2 -3
- cuda/cccl/headers/include/cuda/std/__cmath/isfinite.h +2 -3
- cuda/cccl/headers/include/cuda/std/__cmath/isinf.h +2 -3
- cuda/cccl/headers/include/cuda/std/__cmath/isnan.h +2 -3
- cuda/cccl/headers/include/cuda/std/__cmath/logarithms.h +2 -3
- cuda/cccl/headers/include/cuda/std/__cmath/min_max.h +2 -2
- cuda/cccl/headers/include/cuda/std/__cmath/remainder.h +4 -0
- cuda/cccl/headers/include/cuda/std/__cmath/roots.h +2 -3
- cuda/cccl/headers/include/cuda/std/__cmath/rounding_functions.h +2 -3
- cuda/cccl/headers/include/cuda/std/__cmath/traits.h +4 -0
- cuda/cccl/headers/include/cuda/std/__cmath/trigonometric_functions.h +2 -3
- cuda/cccl/headers/include/cuda/std/__complex/complex.h +0 -6
- cuda/cccl/headers/include/cuda/std/__complex/exponential_functions.h +2 -2
- cuda/cccl/headers/include/cuda/std/__concepts/concept_macros.h +27 -1
- cuda/cccl/headers/include/cuda/std/__concepts/equality_comparable.h +2 -4
- cuda/cccl/headers/include/cuda/std/__exception/cuda_error.h +15 -36
- cuda/cccl/headers/include/cuda/std/__exception/exception_macros.h +93 -0
- cuda/cccl/headers/include/cuda/std/{detail/libcxx/include/stdexcept → __exception/throw_error.h} +3 -3
- cuda/cccl/headers/include/cuda/std/__expected/expected.h +28 -43
- cuda/cccl/headers/include/cuda/std/__expected/unexpected.h +2 -10
- cuda/cccl/headers/include/cuda/std/__format/format_arg_store.h +2 -2
- cuda/cccl/headers/include/cuda/std/__functional/bind.h +6 -6
- cuda/cccl/headers/include/cuda/std/__functional/function.h +2 -6
- cuda/cccl/headers/include/cuda/std/__functional/invoke.h +5 -5
- cuda/cccl/headers/include/cuda/std/__fwd/allocator.h +5 -0
- cuda/cccl/headers/include/cuda/std/__fwd/array.h +2 -2
- cuda/cccl/headers/include/cuda/std/__fwd/char_traits.h +12 -0
- cuda/cccl/headers/include/cuda/std/__fwd/expected.h +46 -0
- cuda/cccl/headers/include/cuda/std/__fwd/get.h +21 -22
- cuda/cccl/headers/include/cuda/std/{detail/libcxx/include/iosfwd → __fwd/ios.h} +5 -10
- cuda/cccl/headers/include/cuda/std/__fwd/mdspan.h +19 -10
- cuda/cccl/headers/include/cuda/std/__fwd/optional.h +2 -2
- cuda/cccl/headers/include/cuda/std/__fwd/reference_wrapper.h +5 -0
- cuda/cccl/headers/include/cuda/std/__fwd/span.h +2 -2
- cuda/cccl/headers/include/cuda/std/__fwd/string.h +7 -0
- cuda/cccl/headers/include/cuda/std/__fwd/string_view.h +18 -0
- cuda/cccl/headers/include/cuda/std/__fwd/tuple.h +3 -0
- cuda/cccl/headers/include/cuda/std/__fwd/unexpected.h +40 -0
- cuda/cccl/headers/include/cuda/std/{__type_traits/is_reference_wrapper.h → __fwd/variant.h} +16 -15
- cuda/cccl/headers/include/cuda/std/__internal/features.h +14 -0
- cuda/cccl/headers/include/cuda/std/__iterator/istream_iterator.h +1 -1
- cuda/cccl/headers/include/cuda/std/__iterator/istreambuf_iterator.h +1 -1
- cuda/cccl/headers/include/cuda/std/__iterator/iter_swap.h +58 -40
- cuda/cccl/headers/include/cuda/std/__iterator/ostream_iterator.h +1 -1
- cuda/cccl/headers/include/cuda/std/__iterator/ostreambuf_iterator.h +1 -1
- cuda/cccl/headers/include/cuda/std/__iterator/reverse_iterator.h +0 -5
- cuda/cccl/headers/include/cuda/std/__limits/numeric_limits.h +4 -18
- cuda/cccl/headers/include/cuda/std/__linalg/conj_if_needed.h +1 -2
- cuda/cccl/headers/include/cuda/std/__linalg/conjugate_transposed.h +0 -2
- cuda/cccl/headers/include/cuda/std/__linalg/conjugated.h +0 -2
- cuda/cccl/headers/include/cuda/std/__linalg/scaled.h +0 -4
- cuda/cccl/headers/include/cuda/std/__linalg/transposed.h +0 -5
- cuda/cccl/headers/include/cuda/std/__mdspan/concepts.h +3 -10
- cuda/cccl/headers/include/cuda/std/__mdspan/extents.h +4 -15
- cuda/cccl/headers/include/cuda/std/__mdspan/layout_left.h +4 -4
- cuda/cccl/headers/include/cuda/std/__mdspan/layout_right.h +4 -4
- cuda/cccl/headers/include/cuda/std/__mdspan/layout_stride.h +2 -4
- cuda/cccl/headers/include/cuda/std/__mdspan/mdspan.h +3 -3
- cuda/cccl/headers/include/cuda/std/__mdspan/submdspan_helper.h +1 -1
- cuda/cccl/headers/include/cuda/std/__memory/allocator_arg_t.h +1 -0
- cuda/cccl/headers/include/cuda/std/__memory/allocator_traits.h +6 -12
- cuda/cccl/headers/include/cuda/std/__memory/construct_at.h +1 -5
- cuda/cccl/headers/include/cuda/std/__memory/is_sufficiently_aligned.h +7 -2
- cuda/cccl/headers/include/cuda/std/__memory/uninitialized_algorithms.h +1 -0
- cuda/cccl/headers/include/cuda/std/__memory/uses_allocator.h +5 -0
- cuda/cccl/headers/include/cuda/std/__new/allocate.h +5 -0
- cuda/cccl/headers/include/cuda/{__barrier/barrier_native_handle.h → std/__new/device_new.h} +9 -24
- cuda/cccl/headers/include/cuda/std/__new_ +1 -0
- cuda/cccl/headers/include/cuda/std/__optional/optional.h +5 -4
- cuda/cccl/headers/include/cuda/std/__optional/optional_ref.h +4 -4
- cuda/cccl/headers/include/cuda/std/__random/linear_congruential_engine.h +1 -1
- cuda/cccl/headers/include/cuda/std/__random/philox_engine.h +562 -0
- cuda/cccl/headers/include/cuda/std/__random/seed_seq.h +204 -0
- cuda/cccl/headers/include/cuda/std/__random_ +2 -0
- cuda/cccl/headers/include/cuda/std/__ranges/concepts.h +7 -19
- cuda/cccl/headers/include/cuda/std/__ranges/movable_box.h +2 -4
- cuda/cccl/headers/include/cuda/std/__ranges/owning_view.h +5 -4
- cuda/cccl/headers/include/cuda/std/__ranges/repeat_view.h +1 -1
- cuda/cccl/headers/include/cuda/std/__string/string_view.h +5 -5
- cuda/cccl/headers/include/cuda/std/__tuple_dir/apply.h +82 -0
- cuda/cccl/headers/include/cuda/std/__tuple_dir/get.h +122 -0
- cuda/cccl/headers/include/cuda/std/__tuple_dir/sfinae_helpers.h +0 -160
- cuda/cccl/headers/include/cuda/std/__tuple_dir/structured_bindings.h +123 -129
- cuda/cccl/headers/include/cuda/std/__tuple_dir/tie.h +55 -0
- cuda/cccl/headers/include/cuda/std/__tuple_dir/tuple.h +457 -0
- cuda/cccl/headers/include/cuda/std/__tuple_dir/tuple_cat.h +158 -0
- cuda/cccl/headers/include/cuda/std/__tuple_dir/tuple_constraints.h +286 -0
- cuda/cccl/headers/include/cuda/std/__tuple_dir/tuple_element.h +7 -0
- cuda/cccl/headers/include/cuda/std/__tuple_dir/tuple_leaf.h +452 -0
- cuda/cccl/headers/include/cuda/std/__tuple_dir/tuple_like.h +1 -2
- cuda/cccl/headers/include/cuda/std/__type_traits/is_comparable.h +78 -0
- cuda/cccl/headers/include/cuda/std/__type_traits/is_convertible.h +1 -1
- cuda/cccl/headers/include/cuda/std/__type_traits/is_fully_bounded_array.h +47 -0
- cuda/cccl/headers/include/cuda/std/__type_traits/is_swappable.h +0 -2
- cuda/cccl/headers/include/cuda/std/__utility/in_place.h +4 -24
- cuda/cccl/headers/include/cuda/std/__utility/integer_sequence.h +0 -2
- cuda/cccl/headers/include/cuda/std/__utility/pair.h +20 -20
- cuda/cccl/headers/include/cuda/std/__utility/rel_ops.h +0 -2
- cuda/cccl/headers/include/cuda/std/__variant/bad_variant_access.h +74 -0
- cuda/cccl/headers/include/cuda/std/__variant/comparison.h +207 -0
- cuda/cccl/headers/include/cuda/std/__variant/get.h +192 -0
- cuda/cccl/headers/include/cuda/std/__variant/hash.h +82 -0
- cuda/cccl/headers/include/cuda/std/__variant/sfinae_helpers.h +89 -0
- cuda/cccl/headers/include/cuda/std/__variant/variant.h +250 -0
- cuda/cccl/headers/include/cuda/std/__variant/variant_access.h +70 -0
- cuda/cccl/headers/include/cuda/std/__variant/variant_base.h +683 -0
- cuda/cccl/headers/include/cuda/std/__variant/variant_constraints.h +135 -0
- cuda/cccl/headers/include/cuda/std/__variant/variant_match.h +126 -0
- cuda/cccl/headers/include/cuda/std/__variant/variant_traits.h +184 -0
- cuda/cccl/headers/include/cuda/std/__variant/variant_visit.h +225 -0
- cuda/cccl/headers/include/cuda/std/__variant/visit.h +148 -0
- cuda/cccl/headers/include/cuda/std/array +1 -1
- cuda/cccl/headers/include/cuda/std/atomic +1 -1
- cuda/cccl/headers/include/cuda/std/bitset +2 -10
- cuda/cccl/headers/include/cuda/std/detail/libcxx/include/__config +6 -6
- cuda/cccl/headers/include/cuda/std/detail/libcxx/include/algorithm +1 -4
- cuda/cccl/headers/include/cuda/std/detail/libcxx/include/chrono +3 -6
- cuda/cccl/headers/include/cuda/std/functional +1 -1
- cuda/cccl/headers/include/cuda/std/initializer_list +8 -0
- cuda/cccl/headers/include/cuda/std/inplace_vector +6 -5
- cuda/cccl/headers/include/cuda/std/iterator +1 -1
- cuda/cccl/headers/include/cuda/std/numbers +0 -2
- cuda/cccl/headers/include/cuda/std/ratio +2 -2
- cuda/cccl/headers/include/cuda/std/span +2 -2
- cuda/cccl/headers/include/cuda/std/string_view +24 -42
- cuda/cccl/headers/include/cuda/std/tuple +18 -1
- cuda/cccl/headers/include/cuda/std/type_traits +0 -1
- cuda/cccl/headers/include/cuda/std/variant +8 -1
- cuda/cccl/headers/include/nv/target +2 -6
- cuda/cccl/headers/include/thrust/detail/adjacent_difference.inl +15 -2
- cuda/cccl/headers/include/thrust/detail/allocator/allocator_traits.h +0 -2
- cuda/cccl/headers/include/thrust/detail/allocator/copy_construct_range.h +0 -1
- cuda/cccl/headers/include/thrust/detail/allocator/destroy_range.h +0 -1
- cuda/cccl/headers/include/thrust/detail/allocator/fill_construct_range.h +0 -2
- cuda/cccl/headers/include/thrust/detail/allocator/malloc_allocator.h +0 -2
- cuda/cccl/headers/include/thrust/detail/allocator/no_throw_allocator.h +0 -2
- cuda/cccl/headers/include/thrust/detail/allocator/tagged_allocator.h +0 -2
- cuda/cccl/headers/include/thrust/detail/allocator/temporary_allocator.h +0 -2
- cuda/cccl/headers/include/thrust/detail/allocator_aware_execution_policy.h +0 -4
- cuda/cccl/headers/include/thrust/detail/binary_search.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/complex/arithmetic.h +2 -7
- cuda/cccl/headers/include/thrust/detail/complex/c99math.h +2 -8
- cuda/cccl/headers/include/thrust/detail/complex/catrig.h +2 -8
- cuda/cccl/headers/include/thrust/detail/complex/catrigf.h +2 -8
- cuda/cccl/headers/include/thrust/detail/complex/ccosh.h +2 -8
- cuda/cccl/headers/include/thrust/detail/complex/ccoshf.h +2 -8
- cuda/cccl/headers/include/thrust/detail/complex/cexp.h +2 -7
- cuda/cccl/headers/include/thrust/detail/complex/cexpf.h +2 -8
- cuda/cccl/headers/include/thrust/detail/complex/clog.h +2 -8
- cuda/cccl/headers/include/thrust/detail/complex/clogf.h +2 -8
- cuda/cccl/headers/include/thrust/detail/complex/cproj.h +2 -7
- cuda/cccl/headers/include/thrust/detail/complex/csinh.h +2 -8
- cuda/cccl/headers/include/thrust/detail/complex/csinhf.h +2 -8
- cuda/cccl/headers/include/thrust/detail/complex/csqrt.h +2 -8
- cuda/cccl/headers/include/thrust/detail/complex/csqrtf.h +2 -8
- cuda/cccl/headers/include/thrust/detail/complex/ctanh.h +2 -8
- cuda/cccl/headers/include/thrust/detail/complex/ctanhf.h +2 -8
- cuda/cccl/headers/include/thrust/detail/complex/math_private.h +2 -8
- cuda/cccl/headers/include/thrust/detail/config/device_system.h +2 -0
- cuda/cccl/headers/include/thrust/detail/config/host_system.h +2 -0
- cuda/cccl/headers/include/thrust/detail/config/namespace.h +0 -1
- cuda/cccl/headers/include/thrust/detail/contiguous_storage.h +0 -2
- cuda/cccl/headers/include/thrust/detail/contiguous_storage.inl +0 -2
- cuda/cccl/headers/include/thrust/detail/copy.h +0 -2
- cuda/cccl/headers/include/thrust/detail/copy.inl +14 -4
- cuda/cccl/headers/include/thrust/detail/copy_if.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/count.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/equal.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/execute_with_allocator.h +4 -5
- cuda/cccl/headers/include/thrust/detail/extrema.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/fill.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/find.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/for_each.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/functional/actor.h +2 -5
- cuda/cccl/headers/include/thrust/detail/functional/operators.h +2 -5
- cuda/cccl/headers/include/thrust/detail/gather.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/generate.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/get_iterator_value.h +0 -2
- cuda/cccl/headers/include/thrust/detail/inner_product.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/internal_functional.h +1 -0
- cuda/cccl/headers/include/thrust/detail/logical.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/malloc_and_free.h +13 -1
- cuda/cccl/headers/include/thrust/detail/merge.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/mismatch.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/overlapped_copy.h +0 -4
- cuda/cccl/headers/include/thrust/detail/partition.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/random_bijection.h +0 -2
- cuda/cccl/headers/include/thrust/detail/range/head_flags.h +0 -2
- cuda/cccl/headers/include/thrust/detail/range/tail_flags.h +0 -2
- cuda/cccl/headers/include/thrust/detail/raw_reference_cast.h +0 -6
- cuda/cccl/headers/include/thrust/detail/reduce.inl +21 -3
- cuda/cccl/headers/include/thrust/detail/reference.h +27 -3
- cuda/cccl/headers/include/thrust/detail/remove.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/replace.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/reverse.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/scan.inl +21 -3
- cuda/cccl/headers/include/thrust/detail/scatter.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/sequence.inl +13 -1
- cuda/cccl/headers/include/thrust/detail/set_operations.inl +13 -1
- cuda/cccl/headers/include/thrust/detail/sort.inl +13 -1
- cuda/cccl/headers/include/thrust/detail/static_assert.h +0 -2
- cuda/cccl/headers/include/thrust/detail/static_map.h +0 -3
- cuda/cccl/headers/include/thrust/detail/swap_ranges.inl +13 -1
- cuda/cccl/headers/include/thrust/detail/tabulate.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/temporary_array.h +0 -4
- cuda/cccl/headers/include/thrust/detail/temporary_array.inl +0 -1
- cuda/cccl/headers/include/thrust/detail/temporary_buffer.h +14 -3
- cuda/cccl/headers/include/thrust/detail/transform_reduce.inl +13 -1
- cuda/cccl/headers/include/thrust/detail/transform_scan.inl +13 -1
- cuda/cccl/headers/include/thrust/detail/trivial_sequence.h +0 -2
- cuda/cccl/headers/include/thrust/detail/tuple_meta_transform.h +0 -2
- cuda/cccl/headers/include/thrust/detail/type_traits/is_call_possible.h +2 -7
- cuda/cccl/headers/include/thrust/detail/type_traits/is_commutative.h +0 -2
- cuda/cccl/headers/include/thrust/detail/type_traits/is_thrust_pointer.h +0 -4
- cuda/cccl/headers/include/thrust/detail/type_traits/pointer_traits.h +0 -4
- cuda/cccl/headers/include/thrust/detail/uninitialized_copy.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/uninitialized_fill.inl +14 -2
- cuda/cccl/headers/include/thrust/detail/unique.inl +21 -3
- cuda/cccl/headers/include/thrust/detail/vector_base.h +0 -2
- cuda/cccl/headers/include/thrust/detail/vector_base.inl +0 -2
- cuda/cccl/headers/include/thrust/execution_policy.h +10 -9
- cuda/cccl/headers/include/thrust/functional.h +0 -2
- cuda/cccl/headers/include/thrust/iterator/detail/device_system_tag.h +9 -4
- cuda/cccl/headers/include/thrust/iterator/detail/host_system_tag.h +8 -4
- cuda/cccl/headers/include/thrust/iterator/detail/iterator_adaptor_base.h +0 -1
- cuda/cccl/headers/include/thrust/iterator/detail/iterator_category_with_system_and_traversal.h +0 -1
- cuda/cccl/headers/include/thrust/iterator/detail/iterator_facade_category.h +0 -1
- cuda/cccl/headers/include/thrust/iterator/detail/minimum_system.h +0 -1
- cuda/cccl/headers/include/thrust/iterator/detail/tagged_iterator.h +0 -1
- cuda/cccl/headers/include/thrust/iterator/detail/tuple_of_iterator_references.h +2 -6
- cuda/cccl/headers/include/thrust/iterator/transform_input_output_iterator.h +0 -1
- cuda/cccl/headers/include/thrust/iterator/transform_iterator.h +0 -2
- cuda/cccl/headers/include/thrust/mr/allocator.h +0 -2
- cuda/cccl/headers/include/thrust/mr/device_memory_resource.h +9 -4
- cuda/cccl/headers/include/thrust/mr/disjoint_pool.h +10 -10
- cuda/cccl/headers/include/thrust/mr/disjoint_sync_pool.h +0 -2
- cuda/cccl/headers/include/thrust/mr/disjoint_tls_pool.h +0 -2
- cuda/cccl/headers/include/thrust/mr/fancy_pointer_resource.h +0 -2
- cuda/cccl/headers/include/thrust/mr/host_memory_resource.h +8 -4
- cuda/cccl/headers/include/thrust/mr/memory_resource.h +0 -2
- cuda/cccl/headers/include/thrust/mr/new.h +0 -2
- cuda/cccl/headers/include/thrust/mr/polymorphic_adaptor.h +0 -2
- cuda/cccl/headers/include/thrust/mr/pool.h +10 -10
- cuda/cccl/headers/include/thrust/mr/pool_options.h +4 -6
- cuda/cccl/headers/include/thrust/mr/sync_pool.h +0 -2
- cuda/cccl/headers/include/thrust/mr/tls_pool.h +0 -2
- cuda/cccl/headers/include/thrust/mr/validator.h +0 -2
- cuda/cccl/headers/include/thrust/per_device_resource.h +13 -1
- cuda/cccl/headers/include/thrust/random/detail/discard_block_engine.inl +0 -2
- cuda/cccl/headers/include/thrust/random/detail/linear_congruential_engine.inl +0 -2
- cuda/cccl/headers/include/thrust/random/detail/linear_congruential_engine_discard.h +2 -9
- cuda/cccl/headers/include/thrust/random/detail/linear_feedback_shift_engine.inl +0 -2
- cuda/cccl/headers/include/thrust/random/detail/linear_feedback_shift_engine_wordmask.h +2 -9
- cuda/cccl/headers/include/thrust/random/detail/mod.h +2 -9
- cuda/cccl/headers/include/thrust/random/detail/normal_distribution.inl +0 -2
- cuda/cccl/headers/include/thrust/random/detail/normal_distribution_base.h +2 -7
- cuda/cccl/headers/include/thrust/random/detail/random_core_access.h +2 -9
- cuda/cccl/headers/include/thrust/random/detail/subtract_with_carry_engine.inl +0 -2
- cuda/cccl/headers/include/thrust/random/detail/uniform_int_distribution.inl +0 -2
- cuda/cccl/headers/include/thrust/random/detail/uniform_real_distribution.inl +0 -2
- cuda/cccl/headers/include/thrust/random/detail/xor_combine_engine.inl +0 -2
- cuda/cccl/headers/include/thrust/random/discard_block_engine.h +0 -2
- cuda/cccl/headers/include/thrust/random/linear_congruential_engine.h +0 -2
- cuda/cccl/headers/include/thrust/random/linear_feedback_shift_engine.h +0 -2
- cuda/cccl/headers/include/thrust/random/normal_distribution.h +0 -2
- cuda/cccl/headers/include/thrust/random/subtract_with_carry_engine.h +0 -2
- cuda/cccl/headers/include/thrust/random/uniform_int_distribution.h +0 -2
- cuda/cccl/headers/include/thrust/random/uniform_real_distribution.h +0 -2
- cuda/cccl/headers/include/thrust/random/xor_combine_engine.h +0 -2
- cuda/cccl/headers/include/thrust/random.h +0 -2
- cuda/cccl/headers/include/thrust/system/cpp/detail/execution_policy.h +15 -11
- cuda/cccl/headers/include/thrust/system/cpp/detail/memory.inl +2 -7
- cuda/cccl/headers/include/thrust/system/cpp/memory.h +0 -1
- cuda/cccl/headers/include/thrust/system/cpp/memory_resource.h +0 -2
- cuda/cccl/headers/include/thrust/system/cpp/pointer.h +0 -2
- cuda/cccl/headers/include/thrust/system/cpp/vector.h +0 -1
- cuda/cccl/headers/include/thrust/system/cuda/detail/adjacent_difference.h +0 -4
- cuda/cccl/headers/include/thrust/system/cuda/detail/copy.h +0 -1
- cuda/cccl/headers/include/thrust/system/cuda/detail/copy_if.h +0 -4
- cuda/cccl/headers/include/thrust/system/cuda/detail/core/agent_launcher.h +2 -9
- cuda/cccl/headers/include/thrust/system/cuda/detail/core/triple_chevron_launch.h +4 -32
- cuda/cccl/headers/include/thrust/system/cuda/detail/core/util.h +2 -9
- cuda/cccl/headers/include/thrust/system/cuda/detail/count.h +0 -2
- cuda/cccl/headers/include/thrust/system/cuda/detail/cross_system.h +0 -2
- cuda/cccl/headers/include/thrust/system/cuda/detail/dispatch.h +23 -2
- cuda/cccl/headers/include/thrust/system/cuda/detail/equal.h +0 -2
- cuda/cccl/headers/include/thrust/system/cuda/detail/error.inl +2 -11
- cuda/cccl/headers/include/thrust/system/cuda/detail/execution_policy.h +2 -0
- cuda/cccl/headers/include/thrust/system/cuda/detail/extrema.h +0 -4
- cuda/cccl/headers/include/thrust/system/cuda/detail/fill.h +0 -1
- cuda/cccl/headers/include/thrust/system/cuda/detail/find.h +0 -5
- cuda/cccl/headers/include/thrust/system/cuda/detail/for_each.h +0 -1
- cuda/cccl/headers/include/thrust/system/cuda/detail/gather.h +0 -2
- cuda/cccl/headers/include/thrust/system/cuda/detail/generate.h +0 -2
- cuda/cccl/headers/include/thrust/system/cuda/detail/iter_swap.h +0 -1
- cuda/cccl/headers/include/thrust/system/cuda/detail/make_unsigned_special.h +2 -8
- cuda/cccl/headers/include/thrust/system/cuda/detail/malloc_and_free.h +0 -2
- cuda/cccl/headers/include/thrust/system/cuda/detail/memory.inl +0 -2
- cuda/cccl/headers/include/thrust/system/cuda/detail/merge.h +2 -26
- cuda/cccl/headers/include/thrust/system/cuda/detail/mismatch.h +7 -142
- cuda/cccl/headers/include/thrust/system/cuda/detail/parallel_for.h +0 -2
- cuda/cccl/headers/include/thrust/system/cuda/detail/partition.h +0 -4
- cuda/cccl/headers/include/thrust/system/cuda/detail/per_device_resource.h +0 -2
- cuda/cccl/headers/include/thrust/system/cuda/detail/reduce.h +0 -5
- cuda/cccl/headers/include/thrust/system/cuda/detail/reduce_by_key.h +0 -4
- cuda/cccl/headers/include/thrust/system/cuda/detail/remove.h +0 -2
- cuda/cccl/headers/include/thrust/system/cuda/detail/replace.h +0 -1
- cuda/cccl/headers/include/thrust/system/cuda/detail/reverse.h +0 -4
- cuda/cccl/headers/include/thrust/system/cuda/detail/scan.h +0 -4
- cuda/cccl/headers/include/thrust/system/cuda/detail/scan_by_key.h +0 -3
- cuda/cccl/headers/include/thrust/system/cuda/detail/scatter.h +0 -2
- cuda/cccl/headers/include/thrust/system/cuda/detail/set_operations.h +3 -5
- cuda/cccl/headers/include/thrust/system/cuda/detail/sort.h +8 -10
- cuda/cccl/headers/include/thrust/system/cuda/detail/temporary_buffer.h +0 -2
- cuda/cccl/headers/include/thrust/system/cuda/detail/transform.h +0 -1
- cuda/cccl/headers/include/thrust/system/cuda/detail/transform_reduce.h +0 -4
- cuda/cccl/headers/include/thrust/system/cuda/detail/transform_scan.h +0 -2
- cuda/cccl/headers/include/thrust/system/cuda/detail/uninitialized_copy.h +1 -7
- cuda/cccl/headers/include/thrust/system/cuda/detail/uninitialized_fill.h +2 -7
- cuda/cccl/headers/include/thrust/system/cuda/detail/unique.h +0 -3
- cuda/cccl/headers/include/thrust/system/cuda/detail/unique_by_key.h +0 -4
- cuda/cccl/headers/include/thrust/system/cuda/error.h +2 -11
- cuda/cccl/headers/include/thrust/system/cuda/memory.h +2 -6
- cuda/cccl/headers/include/thrust/system/cuda/memory_resource.h +2 -9
- cuda/cccl/headers/include/thrust/system/cuda/pointer.h +2 -7
- cuda/cccl/headers/include/thrust/system/cuda/vector.h +2 -6
- cuda/cccl/headers/include/thrust/system/detail/bad_alloc.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/errno.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/error_category.inl +0 -4
- cuda/cccl/headers/include/thrust/system/detail/error_code.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/error_condition.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/adjacent_difference.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/adjacent_difference.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/binary_search.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/binary_search.inl +0 -3
- cuda/cccl/headers/include/thrust/system/detail/generic/copy.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/copy.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/copy_if.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/copy_if.inl +0 -3
- cuda/cccl/headers/include/thrust/system/detail/generic/count.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/count.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/equal.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/equal.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/extrema.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/extrema.inl +0 -3
- cuda/cccl/headers/include/thrust/system/detail/generic/fill.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/find.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/find.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/for_each.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/gather.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/gather.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/generate.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/inner_product.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/inner_product.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/logical.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/memory.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/memory.inl +0 -3
- cuda/cccl/headers/include/thrust/system/detail/generic/merge.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/merge.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/mismatch.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/mismatch.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/partition.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/partition.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/per_device_resource.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/reduce.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/reduce.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/reduce_by_key.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/reduce_by_key.inl +0 -3
- cuda/cccl/headers/include/thrust/system/detail/generic/remove.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/remove.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/replace.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/replace.inl +0 -3
- cuda/cccl/headers/include/thrust/system/detail/generic/reverse.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/reverse.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/scalar/binary_search.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/scalar/binary_search.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/scan.h +26 -12
- cuda/cccl/headers/include/thrust/system/detail/generic/scan_by_key.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/scan_by_key.inl +0 -3
- cuda/cccl/headers/include/thrust/system/detail/generic/scatter.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/scatter.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/select_system.h +0 -1
- cuda/cccl/headers/include/thrust/system/detail/generic/set_operations.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/set_operations.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/shuffle.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/shuffle.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/sort.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/sort.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/swap_ranges.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/swap_ranges.inl +0 -3
- cuda/cccl/headers/include/thrust/system/detail/generic/tabulate.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/tabulate.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/tag.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/temporary_buffer.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/temporary_buffer.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/transform.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/transform_reduce.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/transform_reduce.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/transform_scan.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/transform_scan.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/uninitialized_copy.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/uninitialized_copy.inl +2 -4
- cuda/cccl/headers/include/thrust/system/detail/generic/uninitialized_fill.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/uninitialized_fill.inl +0 -3
- cuda/cccl/headers/include/thrust/system/detail/generic/unique.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/unique.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/unique_by_key.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/generic/unique_by_key.inl +0 -2
- cuda/cccl/headers/include/thrust/system/detail/internal/decompose.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/sequential/adjacent_difference.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/sequential/assign_value.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/sequential/binary_search.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/sequential/copy.h +76 -5
- cuda/cccl/headers/include/thrust/system/detail/sequential/copy_backward.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/sequential/copy_if.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/sequential/extrema.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/sequential/find.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/sequential/for_each.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/sequential/general_copy.h +0 -3
- cuda/cccl/headers/include/thrust/system/detail/sequential/get_value.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/sequential/insertion_sort.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/sequential/iter_swap.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/sequential/malloc_and_free.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/sequential/merge.h +78 -6
- cuda/cccl/headers/include/thrust/system/detail/sequential/partition.h +0 -4
- cuda/cccl/headers/include/thrust/system/detail/sequential/reduce.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/sequential/reduce_by_key.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/sequential/remove.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/sequential/scan.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/sequential/scan_by_key.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/sequential/set_operations.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/sequential/sort.h +67 -6
- cuda/cccl/headers/include/thrust/system/detail/sequential/stable_merge_sort.h +310 -11
- cuda/cccl/headers/include/thrust/system/detail/sequential/stable_primitive_sort.h +78 -5
- cuda/cccl/headers/include/thrust/system/detail/sequential/stable_radix_sort.h +543 -7
- cuda/cccl/headers/include/thrust/system/detail/sequential/trivial_copy.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/sequential/unique.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/sequential/unique_by_key.h +0 -2
- cuda/cccl/headers/include/thrust/system/detail/system_error.inl +0 -2
- cuda/cccl/headers/include/thrust/system/error_code.h +0 -4
- cuda/cccl/headers/include/thrust/system/omp/detail/adjacent_difference.h +5 -25
- cuda/cccl/headers/include/thrust/system/omp/detail/assign_value.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/binary_search.h +5 -25
- cuda/cccl/headers/include/thrust/system/omp/detail/copy.h +40 -29
- cuda/cccl/headers/include/thrust/system/omp/detail/copy_if.h +11 -28
- cuda/cccl/headers/include/thrust/system/omp/detail/count.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/default_decomposition.h +26 -28
- cuda/cccl/headers/include/thrust/system/omp/detail/equal.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/execution_policy.h +18 -13
- cuda/cccl/headers/include/thrust/system/omp/detail/extrema.h +5 -25
- cuda/cccl/headers/include/thrust/system/omp/detail/fill.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/find.h +5 -25
- cuda/cccl/headers/include/thrust/system/omp/detail/for_each.h +47 -30
- cuda/cccl/headers/include/thrust/system/omp/detail/gather.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/generate.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/get_value.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/inner_product.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/iter_swap.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/logical.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/malloc_and_free.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/merge.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/mismatch.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/partition.h +26 -31
- cuda/cccl/headers/include/thrust/system/omp/detail/per_device_resource.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/pragma_omp.h +2 -26
- cuda/cccl/headers/include/thrust/system/omp/detail/reduce.h +35 -27
- cuda/cccl/headers/include/thrust/system/omp/detail/reduce_by_key.h +13 -28
- cuda/cccl/headers/include/thrust/system/omp/detail/reduce_intervals.h +56 -28
- cuda/cccl/headers/include/thrust/system/omp/detail/remove.h +26 -31
- cuda/cccl/headers/include/thrust/system/omp/detail/replace.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/reverse.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/scan.h +176 -17
- cuda/cccl/headers/include/thrust/system/omp/detail/scan_by_key.h +8 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/scatter.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/sequence.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/set_operations.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/sort.h +213 -28
- cuda/cccl/headers/include/thrust/system/omp/detail/swap_ranges.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/tabulate.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/temporary_buffer.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/transform.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/transform_reduce.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/transform_scan.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/uninitialized_copy.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/uninitialized_fill.h +2 -15
- cuda/cccl/headers/include/thrust/system/omp/detail/unique.h +21 -30
- cuda/cccl/headers/include/thrust/system/omp/detail/unique_by_key.h +17 -29
- cuda/cccl/headers/include/thrust/system/omp/memory.h +51 -9
- cuda/cccl/headers/include/thrust/system/omp/memory_resource.h +3 -7
- cuda/cccl/headers/include/thrust/system/omp/pointer.h +3 -7
- cuda/cccl/headers/include/thrust/system/omp/vector.h +3 -6
- cuda/cccl/headers/include/thrust/system/system_error.h +0 -2
- cuda/cccl/headers/include/thrust/system/tbb/detail/adjacent_difference.h +4 -25
- cuda/cccl/headers/include/thrust/system/tbb/detail/assign_value.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/binary_search.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/copy.h +38 -29
- cuda/cccl/headers/include/thrust/system/tbb/detail/copy_if.h +91 -24
- cuda/cccl/headers/include/thrust/system/tbb/detail/count.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/equal.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/execution_policy.h +17 -13
- cuda/cccl/headers/include/thrust/system/tbb/detail/extrema.h +4 -25
- cuda/cccl/headers/include/thrust/system/tbb/detail/fill.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/find.h +4 -25
- cuda/cccl/headers/include/thrust/system/tbb/detail/for_each.h +47 -28
- cuda/cccl/headers/include/thrust/system/tbb/detail/gather.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/generate.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/get_value.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/inner_product.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/iter_swap.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/logical.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/malloc_and_free.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/merge.h +254 -29
- cuda/cccl/headers/include/thrust/system/tbb/detail/mismatch.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/partition.h +25 -31
- cuda/cccl/headers/include/thrust/system/tbb/detail/per_device_resource.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/reduce.h +95 -29
- cuda/cccl/headers/include/thrust/system/tbb/detail/reduce_by_key.h +345 -28
- cuda/cccl/headers/include/thrust/system/tbb/detail/reduce_intervals.h +4 -26
- cuda/cccl/headers/include/thrust/system/tbb/detail/remove.h +32 -42
- cuda/cccl/headers/include/thrust/system/tbb/detail/replace.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/reverse.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/scan.h +265 -30
- cuda/cccl/headers/include/thrust/system/tbb/detail/scan_by_key.h +7 -17
- cuda/cccl/headers/include/thrust/system/tbb/detail/scatter.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/sequence.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/set_operations.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/sort.h +244 -32
- cuda/cccl/headers/include/thrust/system/tbb/detail/swap_ranges.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/tabulate.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/temporary_buffer.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/transform.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/transform_reduce.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/transform_scan.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/uninitialized_copy.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/uninitialized_fill.h +2 -15
- cuda/cccl/headers/include/thrust/system/tbb/detail/unique.h +23 -33
- cuda/cccl/headers/include/thrust/system/tbb/detail/unique_by_key.h +16 -29
- cuda/cccl/headers/include/thrust/system/tbb/memory.h +52 -24
- cuda/cccl/headers/include/thrust/system/tbb/memory_resource.h +4 -22
- cuda/cccl/headers/include/thrust/system/tbb/pointer.h +4 -22
- cuda/cccl/headers/include/thrust/system/tbb/vector.h +4 -21
- cuda/cccl/headers/include/thrust/transform.h +14 -3
- cuda/cccl/headers/include/thrust/type_traits/integer_sequence.h +0 -4
- cuda/cccl/headers/include/thrust/type_traits/is_contiguous_iterator.h +0 -1
- cuda/cccl/headers/include/thrust/type_traits/is_operator_less_or_greater_function_object.h +0 -4
- cuda/cccl/headers/include/thrust/type_traits/is_operator_plus_function_object.h +0 -4
- cuda/cccl/headers/include/thrust/type_traits/is_trivially_relocatable.h +0 -4
- cuda/cccl/headers/include/thrust/universal_allocator.h +8 -0
- cuda/cccl/headers/include/thrust/universal_vector.h +9 -0
- cuda/cccl/headers/include/thrust/zip_function.h +2 -28
- cuda/compute/__init__.py +4 -0
- cuda/compute/_bindings.pyi +26 -3
- cuda/compute/_bindings_impl.pyx +143 -1
- cuda/compute/algorithms/__init__.py +9 -5
- cuda/compute/algorithms/_sort/__init__.py +23 -0
- cuda/compute/algorithms/{_merge_sort.py → _sort/_merge_sort.py} +10 -10
- cuda/compute/algorithms/{_radix_sort.py → _sort/_radix_sort.py} +9 -58
- cuda/compute/algorithms/_sort/_segmented_sort.py +288 -0
- cuda/compute/algorithms/_sort/_sort_common.py +52 -0
- cuda/compute/cu12/_bindings_impl.cpython-313-aarch64-linux-gnu.so +0 -0
- cuda/compute/cu12/cccl/libcccl.c.parallel.so +0 -0
- cuda/compute/cu13/_bindings_impl.cpython-313-aarch64-linux-gnu.so +0 -0
- cuda/compute/cu13/cccl/libcccl.c.parallel.so +0 -0
- cuda_cccl-0.3.4.dist-info/METADATA +78 -0
- {cuda_cccl-0.3.2.dist-info → cuda_cccl-0.3.4.dist-info}/RECORD +830 -867
- cuda/cccl/headers/include/cuda/__memory_resource/resource_ref.h +0 -652
- cuda/cccl/headers/include/cuda/std/detail/libcxx/include/tuple +0 -1365
- cuda/cccl/headers/include/cuda/std/detail/libcxx/include/variant +0 -2144
- cuda/cccl/headers/include/thrust/detail/integer_math.h +0 -113
- cuda/cccl/headers/include/thrust/system/detail/adl/adjacent_difference.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/assign_value.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/binary_search.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/copy.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/copy_if.h +0 -52
- cuda/cccl/headers/include/thrust/system/detail/adl/count.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/equal.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/extrema.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/fill.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/find.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/for_each.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/gather.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/generate.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/get_value.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/inner_product.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/iter_swap.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/logical.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/malloc_and_free.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/merge.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/mismatch.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/partition.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/per_device_resource.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/reduce.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/reduce_by_key.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/remove.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/replace.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/reverse.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/scan.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/scan_by_key.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/scatter.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/sequence.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/set_operations.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/sort.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/swap_ranges.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/tabulate.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/temporary_buffer.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/transform.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/transform_reduce.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/transform_scan.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/uninitialized_copy.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/uninitialized_fill.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/unique.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/adl/unique_by_key.h +0 -51
- cuda/cccl/headers/include/thrust/system/detail/generic/scan.inl +0 -85
- cuda/cccl/headers/include/thrust/system/detail/sequential/copy.inl +0 -119
- cuda/cccl/headers/include/thrust/system/detail/sequential/merge.inl +0 -145
- cuda/cccl/headers/include/thrust/system/detail/sequential/sort.inl +0 -116
- cuda/cccl/headers/include/thrust/system/detail/sequential/stable_merge_sort.inl +0 -356
- cuda/cccl/headers/include/thrust/system/detail/sequential/stable_primitive_sort.inl +0 -124
- cuda/cccl/headers/include/thrust/system/detail/sequential/stable_radix_sort.inl +0 -586
- cuda/cccl/headers/include/thrust/system/omp/detail/copy.inl +0 -74
- cuda/cccl/headers/include/thrust/system/omp/detail/copy_if.inl +0 -59
- cuda/cccl/headers/include/thrust/system/omp/detail/default_decomposition.inl +0 -65
- cuda/cccl/headers/include/thrust/system/omp/detail/for_each.inl +0 -87
- cuda/cccl/headers/include/thrust/system/omp/detail/memory.inl +0 -93
- cuda/cccl/headers/include/thrust/system/omp/detail/partition.inl +0 -102
- cuda/cccl/headers/include/thrust/system/omp/detail/reduce.inl +0 -78
- cuda/cccl/headers/include/thrust/system/omp/detail/reduce_by_key.inl +0 -65
- cuda/cccl/headers/include/thrust/system/omp/detail/reduce_intervals.inl +0 -103
- cuda/cccl/headers/include/thrust/system/omp/detail/remove.inl +0 -87
- cuda/cccl/headers/include/thrust/system/omp/detail/sort.inl +0 -265
- cuda/cccl/headers/include/thrust/system/omp/detail/unique.inl +0 -71
- cuda/cccl/headers/include/thrust/system/omp/detail/unique_by_key.inl +0 -75
- cuda/cccl/headers/include/thrust/system/tbb/detail/copy.inl +0 -73
- cuda/cccl/headers/include/thrust/system/tbb/detail/copy_if.inl +0 -136
- cuda/cccl/headers/include/thrust/system/tbb/detail/for_each.inl +0 -91
- cuda/cccl/headers/include/thrust/system/tbb/detail/memory.inl +0 -94
- cuda/cccl/headers/include/thrust/system/tbb/detail/merge.inl +0 -327
- cuda/cccl/headers/include/thrust/system/tbb/detail/partition.inl +0 -98
- cuda/cccl/headers/include/thrust/system/tbb/detail/reduce.inl +0 -137
- cuda/cccl/headers/include/thrust/system/tbb/detail/reduce_by_key.inl +0 -400
- cuda/cccl/headers/include/thrust/system/tbb/detail/remove.inl +0 -87
- cuda/cccl/headers/include/thrust/system/tbb/detail/scan.inl +0 -312
- cuda/cccl/headers/include/thrust/system/tbb/detail/sort.inl +0 -295
- cuda/cccl/headers/include/thrust/system/tbb/detail/unique.inl +0 -71
- cuda/cccl/headers/include/thrust/system/tbb/detail/unique_by_key.inl +0 -75
- cuda_cccl-0.3.2.dist-info/METADATA +0 -42
- {cuda_cccl-0.3.2.dist-info → cuda_cccl-0.3.4.dist-info}/WHEEL +0 -0
- {cuda_cccl-0.3.2.dist-info → cuda_cccl-0.3.4.dist-info}/licenses/LICENSE +0 -0
|
@@ -1,145 +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/copy.h>
|
|
29
|
-
#include <thrust/detail/function.h>
|
|
30
|
-
#include <thrust/iterator/iterator_traits.h>
|
|
31
|
-
#include <thrust/system/detail/sequential/merge.h>
|
|
32
|
-
|
|
33
|
-
THRUST_NAMESPACE_BEGIN
|
|
34
|
-
namespace system::detail::sequential
|
|
35
|
-
{
|
|
36
|
-
|
|
37
|
-
_CCCL_EXEC_CHECK_DISABLE
|
|
38
|
-
template <typename DerivedPolicy,
|
|
39
|
-
typename InputIterator1,
|
|
40
|
-
typename InputIterator2,
|
|
41
|
-
typename OutputIterator,
|
|
42
|
-
typename StrictWeakOrdering>
|
|
43
|
-
_CCCL_HOST_DEVICE OutputIterator merge(
|
|
44
|
-
sequential::execution_policy<DerivedPolicy>& exec,
|
|
45
|
-
InputIterator1 first1,
|
|
46
|
-
InputIterator1 last1,
|
|
47
|
-
InputIterator2 first2,
|
|
48
|
-
InputIterator2 last2,
|
|
49
|
-
OutputIterator result,
|
|
50
|
-
StrictWeakOrdering comp)
|
|
51
|
-
{
|
|
52
|
-
// wrap comp
|
|
53
|
-
thrust::detail::wrapped_function<StrictWeakOrdering, bool> wrapped_comp{comp};
|
|
54
|
-
|
|
55
|
-
while (first1 != last1 && first2 != last2)
|
|
56
|
-
{
|
|
57
|
-
if (wrapped_comp(*first2, *first1))
|
|
58
|
-
{
|
|
59
|
-
*result = *first2;
|
|
60
|
-
++first2;
|
|
61
|
-
} // end if
|
|
62
|
-
else
|
|
63
|
-
{
|
|
64
|
-
*result = *first1;
|
|
65
|
-
++first1;
|
|
66
|
-
} // end else
|
|
67
|
-
|
|
68
|
-
++result;
|
|
69
|
-
} // end while
|
|
70
|
-
|
|
71
|
-
return thrust::copy(exec, first2, last2, thrust::copy(exec, first1, last1, result));
|
|
72
|
-
} // end merge()
|
|
73
|
-
|
|
74
|
-
_CCCL_EXEC_CHECK_DISABLE
|
|
75
|
-
template <typename DerivedPolicy,
|
|
76
|
-
typename InputIterator1,
|
|
77
|
-
typename InputIterator2,
|
|
78
|
-
typename InputIterator3,
|
|
79
|
-
typename InputIterator4,
|
|
80
|
-
typename OutputIterator1,
|
|
81
|
-
typename OutputIterator2,
|
|
82
|
-
typename StrictWeakOrdering>
|
|
83
|
-
_CCCL_HOST_DEVICE thrust::pair<OutputIterator1, OutputIterator2> merge_by_key(
|
|
84
|
-
sequential::execution_policy<DerivedPolicy>&,
|
|
85
|
-
InputIterator1 keys_first1,
|
|
86
|
-
InputIterator1 keys_last1,
|
|
87
|
-
InputIterator2 keys_first2,
|
|
88
|
-
InputIterator2 keys_last2,
|
|
89
|
-
InputIterator3 values_first1,
|
|
90
|
-
InputIterator4 values_first2,
|
|
91
|
-
OutputIterator1 keys_result,
|
|
92
|
-
OutputIterator2 values_result,
|
|
93
|
-
StrictWeakOrdering comp)
|
|
94
|
-
{
|
|
95
|
-
// wrap comp
|
|
96
|
-
thrust::detail::wrapped_function<StrictWeakOrdering, bool> wrapped_comp{comp};
|
|
97
|
-
|
|
98
|
-
while (keys_first1 != keys_last1 && keys_first2 != keys_last2)
|
|
99
|
-
{
|
|
100
|
-
if (!wrapped_comp(*keys_first2, *keys_first1))
|
|
101
|
-
{
|
|
102
|
-
// *keys_first1 <= *keys_first2
|
|
103
|
-
*keys_result = *keys_first1;
|
|
104
|
-
*values_result = *values_first1;
|
|
105
|
-
++keys_first1;
|
|
106
|
-
++values_first1;
|
|
107
|
-
}
|
|
108
|
-
else
|
|
109
|
-
{
|
|
110
|
-
// *keys_first1 > keys_first2
|
|
111
|
-
*keys_result = *keys_first2;
|
|
112
|
-
*values_result = *values_first2;
|
|
113
|
-
++keys_first2;
|
|
114
|
-
++values_first2;
|
|
115
|
-
}
|
|
116
|
-
|
|
117
|
-
++keys_result;
|
|
118
|
-
++values_result;
|
|
119
|
-
}
|
|
120
|
-
|
|
121
|
-
while (keys_first1 != keys_last1)
|
|
122
|
-
{
|
|
123
|
-
*keys_result = *keys_first1;
|
|
124
|
-
*values_result = *values_first1;
|
|
125
|
-
++keys_first1;
|
|
126
|
-
++values_first1;
|
|
127
|
-
++keys_result;
|
|
128
|
-
++values_result;
|
|
129
|
-
}
|
|
130
|
-
|
|
131
|
-
while (keys_first2 != keys_last2)
|
|
132
|
-
{
|
|
133
|
-
*keys_result = *keys_first2;
|
|
134
|
-
*values_result = *values_first2;
|
|
135
|
-
++keys_first2;
|
|
136
|
-
++values_first2;
|
|
137
|
-
++keys_result;
|
|
138
|
-
++values_result;
|
|
139
|
-
}
|
|
140
|
-
|
|
141
|
-
return thrust::make_pair(keys_result, values_result);
|
|
142
|
-
}
|
|
143
|
-
|
|
144
|
-
} // namespace system::detail::sequential
|
|
145
|
-
THRUST_NAMESPACE_END
|
|
@@ -1,116 +0,0 @@
|
|
|
1
|
-
/*
|
|
2
|
-
* Copyright 2008-2021 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
|
-
#include <thrust/detail/type_traits.h>
|
|
30
|
-
#include <thrust/iterator/iterator_traits.h>
|
|
31
|
-
#include <thrust/reverse.h>
|
|
32
|
-
#include <thrust/system/detail/sequential/stable_merge_sort.h>
|
|
33
|
-
#include <thrust/system/detail/sequential/stable_primitive_sort.h>
|
|
34
|
-
|
|
35
|
-
#include <nv/target>
|
|
36
|
-
|
|
37
|
-
THRUST_NAMESPACE_BEGIN
|
|
38
|
-
namespace system::detail::sequential
|
|
39
|
-
{
|
|
40
|
-
namespace sort_detail
|
|
41
|
-
{
|
|
42
|
-
template <typename KeyType, typename Compare>
|
|
43
|
-
inline constexpr bool use_primitive_sort =
|
|
44
|
-
::cuda::std::is_arithmetic_v<KeyType>
|
|
45
|
-
&& (::cuda::std::is_same_v<Compare, ::cuda::std::less<KeyType>>
|
|
46
|
-
|| ::cuda::std::is_same_v<Compare, ::cuda::std::greater<KeyType>>);
|
|
47
|
-
|
|
48
|
-
template <typename KeyType, typename Compare>
|
|
49
|
-
inline constexpr bool needs_reverse = ::cuda::std::is_same_v<Compare, ::cuda::std::greater<KeyType>>;
|
|
50
|
-
} // end namespace sort_detail
|
|
51
|
-
|
|
52
|
-
template <typename DerivedPolicy, typename RandomAccessIterator, typename StrictWeakOrdering>
|
|
53
|
-
_CCCL_HOST_DEVICE void stable_sort(
|
|
54
|
-
sequential::execution_policy<DerivedPolicy>& exec,
|
|
55
|
-
RandomAccessIterator first,
|
|
56
|
-
RandomAccessIterator last,
|
|
57
|
-
[[maybe_unused]] StrictWeakOrdering comp) // GCC 7-9 warn that comp is unused
|
|
58
|
-
{
|
|
59
|
-
NV_IF_TARGET(
|
|
60
|
-
NV_IS_HOST,
|
|
61
|
-
(
|
|
62
|
-
using KeyType = thrust::detail::it_value_t<RandomAccessIterator>;
|
|
63
|
-
if constexpr (sort_detail::use_primitive_sort<KeyType, StrictWeakOrdering>) {
|
|
64
|
-
thrust::system::detail::sequential::stable_primitive_sort(exec, first, last);
|
|
65
|
-
|
|
66
|
-
// if comp is greater<T> then reverse the keys
|
|
67
|
-
if constexpr (sort_detail::needs_reverse<KeyType, StrictWeakOrdering>)
|
|
68
|
-
{
|
|
69
|
-
thrust::reverse(exec, first, last);
|
|
70
|
-
}
|
|
71
|
-
} else { thrust::system::detail::sequential::stable_merge_sort(exec, first, last, comp); }),
|
|
72
|
-
( // NV_IS_DEVICE:
|
|
73
|
-
// the compilation time of stable_primitive_sort is too expensive to use within a single CUDA thread
|
|
74
|
-
thrust::system::detail::sequential::stable_merge_sort(exec, first, last, comp);));
|
|
75
|
-
#if _CCCL_COMPILER(GCC, <, 10)
|
|
76
|
-
(void) comp; // GCC 7-9 warn that comp is unused
|
|
77
|
-
#endif // _CCCL_COMPILER(GCC, <, 10)
|
|
78
|
-
}
|
|
79
|
-
|
|
80
|
-
template <typename DerivedPolicy,
|
|
81
|
-
typename RandomAccessIterator1,
|
|
82
|
-
typename RandomAccessIterator2,
|
|
83
|
-
typename StrictWeakOrdering>
|
|
84
|
-
_CCCL_HOST_DEVICE void stable_sort_by_key(
|
|
85
|
-
sequential::execution_policy<DerivedPolicy>& exec,
|
|
86
|
-
RandomAccessIterator1 first1,
|
|
87
|
-
RandomAccessIterator1 last1,
|
|
88
|
-
RandomAccessIterator2 first2,
|
|
89
|
-
[[maybe_unused]] StrictWeakOrdering comp) // GCC 7-9 warn that comp is unused
|
|
90
|
-
{
|
|
91
|
-
NV_IF_TARGET(
|
|
92
|
-
NV_IS_HOST,
|
|
93
|
-
(
|
|
94
|
-
using KeyType = thrust::detail::it_value_t<RandomAccessIterator1>;
|
|
95
|
-
if constexpr (sort_detail::use_primitive_sort<KeyType, StrictWeakOrdering>) {
|
|
96
|
-
// if comp is greater<T> then reverse the keys and values
|
|
97
|
-
// note, we also have to reverse the (unordered) input to preserve stability
|
|
98
|
-
if constexpr (sort_detail::needs_reverse<KeyType, StrictWeakOrdering>)
|
|
99
|
-
{
|
|
100
|
-
thrust::reverse(exec, first1, last1);
|
|
101
|
-
thrust::reverse(exec, first2, first2 + (last1 - first1));
|
|
102
|
-
}
|
|
103
|
-
thrust::system::detail::sequential::stable_primitive_sort_by_key(exec, first1, last1, first2);
|
|
104
|
-
if constexpr (sort_detail::needs_reverse<KeyType, StrictWeakOrdering>)
|
|
105
|
-
{
|
|
106
|
-
thrust::reverse(exec, first1, last1);
|
|
107
|
-
thrust::reverse(exec, first2, first2 + (last1 - first1));
|
|
108
|
-
}
|
|
109
|
-
} else { thrust::system::detail::sequential::stable_merge_sort_by_key(exec, first1, last1, first2, comp); }),
|
|
110
|
-
( // NV_IS_DEVICE:
|
|
111
|
-
// the compilation time of stable_primitive_sort is too expensive to use within a single CUDA thread
|
|
112
|
-
thrust::system::detail::sequential::stable_merge_sort_by_key(exec, first1, last1, first2, comp);));
|
|
113
|
-
}
|
|
114
|
-
|
|
115
|
-
} // namespace system::detail::sequential
|
|
116
|
-
THRUST_NAMESPACE_END
|
|
@@ -1,356 +0,0 @@
|
|
|
1
|
-
/*
|
|
2
|
-
* Copyright 2008-2021 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
|
-
#include <thrust/detail/temporary_array.h>
|
|
30
|
-
#include <thrust/iterator/iterator_traits.h>
|
|
31
|
-
#include <thrust/merge.h>
|
|
32
|
-
#include <thrust/system/detail/sequential/insertion_sort.h>
|
|
33
|
-
|
|
34
|
-
#include <cuda/std/__algorithm/min.h>
|
|
35
|
-
#include <cuda/std/iterator>
|
|
36
|
-
|
|
37
|
-
#include <nv/target>
|
|
38
|
-
|
|
39
|
-
THRUST_NAMESPACE_BEGIN
|
|
40
|
-
namespace system::detail::sequential
|
|
41
|
-
{
|
|
42
|
-
namespace stable_merge_sort_detail
|
|
43
|
-
{
|
|
44
|
-
|
|
45
|
-
template <typename DerivedPolicy, typename RandomAccessIterator, typename StrictWeakOrdering>
|
|
46
|
-
_CCCL_HOST_DEVICE void inplace_merge(
|
|
47
|
-
sequential::execution_policy<DerivedPolicy>& exec,
|
|
48
|
-
RandomAccessIterator first,
|
|
49
|
-
RandomAccessIterator middle,
|
|
50
|
-
RandomAccessIterator last,
|
|
51
|
-
StrictWeakOrdering comp)
|
|
52
|
-
{
|
|
53
|
-
using value_type = thrust::detail::it_value_t<RandomAccessIterator>;
|
|
54
|
-
|
|
55
|
-
thrust::detail::temporary_array<value_type, DerivedPolicy> a(exec, first, middle);
|
|
56
|
-
thrust::detail::temporary_array<value_type, DerivedPolicy> b(exec, middle, last);
|
|
57
|
-
|
|
58
|
-
thrust::merge(exec, a.begin(), a.end(), b.begin(), b.end(), first, comp);
|
|
59
|
-
}
|
|
60
|
-
|
|
61
|
-
template <typename DerivedPolicy,
|
|
62
|
-
typename RandomAccessIterator1,
|
|
63
|
-
typename RandomAccessIterator2,
|
|
64
|
-
typename StrictWeakOrdering>
|
|
65
|
-
_CCCL_HOST_DEVICE void inplace_merge_by_key(
|
|
66
|
-
sequential::execution_policy<DerivedPolicy>& exec,
|
|
67
|
-
RandomAccessIterator1 first1,
|
|
68
|
-
RandomAccessIterator1 middle1,
|
|
69
|
-
RandomAccessIterator1 last1,
|
|
70
|
-
RandomAccessIterator2 first2,
|
|
71
|
-
StrictWeakOrdering comp)
|
|
72
|
-
{
|
|
73
|
-
using value_type1 = thrust::detail::it_value_t<RandomAccessIterator1>;
|
|
74
|
-
using value_type2 = thrust::detail::it_value_t<RandomAccessIterator2>;
|
|
75
|
-
|
|
76
|
-
RandomAccessIterator2 middle2 = first2 + (middle1 - first1);
|
|
77
|
-
RandomAccessIterator2 last2 = first2 + (last1 - first1);
|
|
78
|
-
|
|
79
|
-
thrust::detail::temporary_array<value_type1, DerivedPolicy> lhs1(exec, first1, middle1);
|
|
80
|
-
thrust::detail::temporary_array<value_type1, DerivedPolicy> rhs1(exec, middle1, last1);
|
|
81
|
-
thrust::detail::temporary_array<value_type2, DerivedPolicy> lhs2(exec, first2, middle2);
|
|
82
|
-
thrust::detail::temporary_array<value_type2, DerivedPolicy> rhs2(exec, middle2, last2);
|
|
83
|
-
|
|
84
|
-
thrust::merge_by_key(
|
|
85
|
-
exec, lhs1.begin(), lhs1.end(), rhs1.begin(), rhs1.end(), lhs2.begin(), rhs2.begin(), first1, first2, comp);
|
|
86
|
-
}
|
|
87
|
-
|
|
88
|
-
template <typename RandomAccessIterator, typename Size, typename StrictWeakOrdering>
|
|
89
|
-
_CCCL_HOST_DEVICE void
|
|
90
|
-
insertion_sort_each(RandomAccessIterator first, RandomAccessIterator last, Size partition_size, StrictWeakOrdering comp)
|
|
91
|
-
{
|
|
92
|
-
if (partition_size > 1)
|
|
93
|
-
{
|
|
94
|
-
for (; first < last; first += partition_size)
|
|
95
|
-
{
|
|
96
|
-
RandomAccessIterator partition_last = (::cuda::std::min) (last, first + partition_size);
|
|
97
|
-
|
|
98
|
-
thrust::system::detail::sequential::insertion_sort(first, partition_last, comp);
|
|
99
|
-
} // end for
|
|
100
|
-
} // end if
|
|
101
|
-
} // end insertion_sort_each()
|
|
102
|
-
|
|
103
|
-
template <typename RandomAccessIterator1, typename RandomAccessIterator2, typename Size, typename StrictWeakOrdering>
|
|
104
|
-
_CCCL_HOST_DEVICE void insertion_sort_each_by_key(
|
|
105
|
-
RandomAccessIterator1 keys_first,
|
|
106
|
-
RandomAccessIterator1 keys_last,
|
|
107
|
-
RandomAccessIterator2 values_first,
|
|
108
|
-
Size partition_size,
|
|
109
|
-
StrictWeakOrdering comp)
|
|
110
|
-
{
|
|
111
|
-
if (partition_size > 1)
|
|
112
|
-
{
|
|
113
|
-
for (; keys_first < keys_last; keys_first += partition_size, values_first += partition_size)
|
|
114
|
-
{
|
|
115
|
-
RandomAccessIterator1 keys_partition_last = (::cuda::std::min) (keys_last, keys_first + partition_size);
|
|
116
|
-
|
|
117
|
-
thrust::system::detail::sequential::insertion_sort_by_key(keys_first, keys_partition_last, values_first, comp);
|
|
118
|
-
} // end for
|
|
119
|
-
} // end if
|
|
120
|
-
} // end insertion_sort_each()
|
|
121
|
-
|
|
122
|
-
template <typename DerivedPolicy,
|
|
123
|
-
typename RandomAccessIterator1,
|
|
124
|
-
typename Size,
|
|
125
|
-
typename RandomAccessIterator2,
|
|
126
|
-
typename StrictWeakOrdering>
|
|
127
|
-
_CCCL_HOST_DEVICE void merge_adjacent_partitions(
|
|
128
|
-
sequential::execution_policy<DerivedPolicy>& exec,
|
|
129
|
-
RandomAccessIterator1 first,
|
|
130
|
-
RandomAccessIterator1 last,
|
|
131
|
-
Size partition_size,
|
|
132
|
-
RandomAccessIterator2 result,
|
|
133
|
-
StrictWeakOrdering comp)
|
|
134
|
-
{
|
|
135
|
-
for (; first < last; first += 2 * partition_size, result += 2 * partition_size)
|
|
136
|
-
{
|
|
137
|
-
RandomAccessIterator1 interval_middle = (::cuda::std::min) (last, first + partition_size);
|
|
138
|
-
RandomAccessIterator1 interval_last = (::cuda::std::min) (last, interval_middle + partition_size);
|
|
139
|
-
|
|
140
|
-
thrust::merge(exec, first, interval_middle, interval_middle, interval_last, result, comp);
|
|
141
|
-
} // end for
|
|
142
|
-
} // end merge_adjacent_partitions()
|
|
143
|
-
|
|
144
|
-
template <typename DerivedPolicy,
|
|
145
|
-
typename RandomAccessIterator1,
|
|
146
|
-
typename RandomAccessIterator2,
|
|
147
|
-
typename Size,
|
|
148
|
-
typename RandomAccessIterator3,
|
|
149
|
-
typename RandomAccessIterator4,
|
|
150
|
-
typename StrictWeakOrdering>
|
|
151
|
-
_CCCL_HOST_DEVICE void merge_adjacent_partitions_by_key(
|
|
152
|
-
sequential::execution_policy<DerivedPolicy>& exec,
|
|
153
|
-
RandomAccessIterator1 keys_first,
|
|
154
|
-
RandomAccessIterator1 keys_last,
|
|
155
|
-
RandomAccessIterator2 values_first,
|
|
156
|
-
Size partition_size,
|
|
157
|
-
RandomAccessIterator3 keys_result,
|
|
158
|
-
RandomAccessIterator4 values_result,
|
|
159
|
-
StrictWeakOrdering comp)
|
|
160
|
-
{
|
|
161
|
-
Size stride = 2 * partition_size;
|
|
162
|
-
|
|
163
|
-
for (; keys_first < keys_last;
|
|
164
|
-
keys_first += stride, values_first += stride, keys_result += stride, values_result += stride)
|
|
165
|
-
{
|
|
166
|
-
RandomAccessIterator1 keys_interval_middle = (::cuda::std::min) (keys_last, keys_first + partition_size);
|
|
167
|
-
RandomAccessIterator1 keys_interval_last = (::cuda::std::min) (keys_last, keys_interval_middle + partition_size);
|
|
168
|
-
|
|
169
|
-
RandomAccessIterator2 values_first2 = values_first + (keys_interval_middle - keys_first);
|
|
170
|
-
|
|
171
|
-
thrust::merge_by_key(
|
|
172
|
-
exec,
|
|
173
|
-
keys_first,
|
|
174
|
-
keys_interval_middle,
|
|
175
|
-
keys_interval_middle,
|
|
176
|
-
keys_interval_last,
|
|
177
|
-
values_first,
|
|
178
|
-
values_first2,
|
|
179
|
-
keys_result,
|
|
180
|
-
values_result,
|
|
181
|
-
comp);
|
|
182
|
-
} // end for
|
|
183
|
-
} // end merge_adjacent_partitions()
|
|
184
|
-
|
|
185
|
-
template <typename DerivedPolicy, typename RandomAccessIterator, typename StrictWeakOrdering>
|
|
186
|
-
_CCCL_HOST_DEVICE void iterative_stable_merge_sort(
|
|
187
|
-
sequential::execution_policy<DerivedPolicy>& exec,
|
|
188
|
-
RandomAccessIterator first,
|
|
189
|
-
RandomAccessIterator last,
|
|
190
|
-
StrictWeakOrdering comp)
|
|
191
|
-
{
|
|
192
|
-
using value_type = thrust::detail::it_value_t<RandomAccessIterator>;
|
|
193
|
-
using difference_type = thrust::detail::it_difference_t<RandomAccessIterator>;
|
|
194
|
-
|
|
195
|
-
difference_type n = last - first;
|
|
196
|
-
|
|
197
|
-
thrust::detail::temporary_array<value_type, DerivedPolicy> temp(exec, n);
|
|
198
|
-
|
|
199
|
-
// insertion sort each 32 element partition
|
|
200
|
-
difference_type partition_size = 32;
|
|
201
|
-
insertion_sort_each(first, last, partition_size, comp);
|
|
202
|
-
|
|
203
|
-
// ping indicates whether or not the latest data is in the source range [first, last)
|
|
204
|
-
bool ping = true;
|
|
205
|
-
|
|
206
|
-
// merge adjacent partitions until the partition size covers the entire range
|
|
207
|
-
for (; partition_size < n; partition_size *= 2, ping = !ping)
|
|
208
|
-
{
|
|
209
|
-
if (ping)
|
|
210
|
-
{
|
|
211
|
-
merge_adjacent_partitions(exec, first, last, partition_size, temp.begin(), comp);
|
|
212
|
-
} // end if
|
|
213
|
-
else
|
|
214
|
-
{
|
|
215
|
-
merge_adjacent_partitions(exec, temp.begin(), temp.end(), partition_size, first, comp);
|
|
216
|
-
} // end else
|
|
217
|
-
} // end for m
|
|
218
|
-
|
|
219
|
-
if (!ping)
|
|
220
|
-
{
|
|
221
|
-
thrust::copy(exec, temp.begin(), temp.end(), first);
|
|
222
|
-
} // end if
|
|
223
|
-
} // end iterative_stable_merge_sort()
|
|
224
|
-
|
|
225
|
-
template <typename DerivedPolicy,
|
|
226
|
-
typename RandomAccessIterator1,
|
|
227
|
-
typename RandomAccessIterator2,
|
|
228
|
-
typename StrictWeakOrdering>
|
|
229
|
-
_CCCL_HOST_DEVICE void iterative_stable_merge_sort_by_key(
|
|
230
|
-
sequential::execution_policy<DerivedPolicy>& exec,
|
|
231
|
-
RandomAccessIterator1 keys_first,
|
|
232
|
-
RandomAccessIterator1 keys_last,
|
|
233
|
-
RandomAccessIterator2 values_first,
|
|
234
|
-
StrictWeakOrdering comp)
|
|
235
|
-
{
|
|
236
|
-
using value_type1 = thrust::detail::it_value_t<RandomAccessIterator1>;
|
|
237
|
-
using value_type2 = thrust::detail::it_value_t<RandomAccessIterator2>;
|
|
238
|
-
using difference_type = thrust::detail::it_difference_t<RandomAccessIterator1>;
|
|
239
|
-
|
|
240
|
-
difference_type n = keys_last - keys_first;
|
|
241
|
-
|
|
242
|
-
thrust::detail::temporary_array<value_type1, DerivedPolicy> keys_temp(exec, n);
|
|
243
|
-
thrust::detail::temporary_array<value_type2, DerivedPolicy> values_temp(exec, n);
|
|
244
|
-
|
|
245
|
-
// insertion sort each 32 element partition
|
|
246
|
-
difference_type partition_size = 32;
|
|
247
|
-
insertion_sort_each_by_key(keys_first, keys_last, values_first, partition_size, comp);
|
|
248
|
-
|
|
249
|
-
// ping indicates whether or not the latest data is in the source range [first, last)
|
|
250
|
-
bool ping = true;
|
|
251
|
-
|
|
252
|
-
// merge adjacent partitions until the partition size covers the entire range
|
|
253
|
-
for (; partition_size < n; partition_size *= 2, ping = !ping)
|
|
254
|
-
{
|
|
255
|
-
if (ping)
|
|
256
|
-
{
|
|
257
|
-
merge_adjacent_partitions_by_key(
|
|
258
|
-
exec, keys_first, keys_last, values_first, partition_size, keys_temp.begin(), values_temp.begin(), comp);
|
|
259
|
-
} // end if
|
|
260
|
-
else
|
|
261
|
-
{
|
|
262
|
-
merge_adjacent_partitions_by_key(
|
|
263
|
-
exec, keys_temp.begin(), keys_temp.end(), values_temp.begin(), partition_size, keys_first, values_first, comp);
|
|
264
|
-
} // end else
|
|
265
|
-
} // end for m
|
|
266
|
-
|
|
267
|
-
if (!ping)
|
|
268
|
-
{
|
|
269
|
-
thrust::copy(exec, keys_temp.begin(), keys_temp.end(), keys_first);
|
|
270
|
-
thrust::copy(exec, values_temp.begin(), values_temp.end(), values_first);
|
|
271
|
-
} // end if
|
|
272
|
-
} // end iterative_stable_merge_sort()
|
|
273
|
-
|
|
274
|
-
template <typename DerivedPolicy, typename RandomAccessIterator, typename StrictWeakOrdering>
|
|
275
|
-
_CCCL_HOST_DEVICE void recursive_stable_merge_sort(
|
|
276
|
-
sequential::execution_policy<DerivedPolicy>& exec,
|
|
277
|
-
RandomAccessIterator first,
|
|
278
|
-
RandomAccessIterator last,
|
|
279
|
-
StrictWeakOrdering comp)
|
|
280
|
-
{
|
|
281
|
-
if (last - first <= 32)
|
|
282
|
-
{
|
|
283
|
-
thrust::system::detail::sequential::insertion_sort(first, last, comp);
|
|
284
|
-
} // end if
|
|
285
|
-
else
|
|
286
|
-
{
|
|
287
|
-
RandomAccessIterator middle = first + (last - first) / 2;
|
|
288
|
-
|
|
289
|
-
stable_merge_sort_detail::recursive_stable_merge_sort(exec, first, middle, comp);
|
|
290
|
-
stable_merge_sort_detail::recursive_stable_merge_sort(exec, middle, last, comp);
|
|
291
|
-
stable_merge_sort_detail::inplace_merge(exec, first, middle, last, comp);
|
|
292
|
-
} // end else
|
|
293
|
-
} // end recursive_stable_merge_sort()
|
|
294
|
-
|
|
295
|
-
template <typename DerivedPolicy,
|
|
296
|
-
typename RandomAccessIterator1,
|
|
297
|
-
typename RandomAccessIterator2,
|
|
298
|
-
typename StrictWeakOrdering>
|
|
299
|
-
_CCCL_HOST_DEVICE void recursive_stable_merge_sort_by_key(
|
|
300
|
-
sequential::execution_policy<DerivedPolicy>& exec,
|
|
301
|
-
RandomAccessIterator1 first1,
|
|
302
|
-
RandomAccessIterator1 last1,
|
|
303
|
-
RandomAccessIterator2 first2,
|
|
304
|
-
StrictWeakOrdering comp)
|
|
305
|
-
{
|
|
306
|
-
if (last1 - first1 <= 32)
|
|
307
|
-
{
|
|
308
|
-
thrust::system::detail::sequential::insertion_sort_by_key(first1, last1, first2, comp);
|
|
309
|
-
} // end if
|
|
310
|
-
else
|
|
311
|
-
{
|
|
312
|
-
RandomAccessIterator1 middle1 = first1 + (last1 - first1) / 2;
|
|
313
|
-
RandomAccessIterator2 middle2 = first2 + (last1 - first1) / 2;
|
|
314
|
-
|
|
315
|
-
stable_merge_sort_detail::recursive_stable_merge_sort_by_key(exec, first1, middle1, first2, comp);
|
|
316
|
-
stable_merge_sort_detail::recursive_stable_merge_sort_by_key(exec, middle1, last1, middle2, comp);
|
|
317
|
-
stable_merge_sort_detail::inplace_merge_by_key(exec, first1, middle1, last1, first2, comp);
|
|
318
|
-
} // end else
|
|
319
|
-
} // end recursive_stable_merge_sort_by_key()
|
|
320
|
-
|
|
321
|
-
} // end namespace stable_merge_sort_detail
|
|
322
|
-
|
|
323
|
-
template <typename DerivedPolicy, typename RandomAccessIterator, typename StrictWeakOrdering>
|
|
324
|
-
_CCCL_HOST_DEVICE void stable_merge_sort(
|
|
325
|
-
sequential::execution_policy<DerivedPolicy>& exec,
|
|
326
|
-
RandomAccessIterator first,
|
|
327
|
-
RandomAccessIterator last,
|
|
328
|
-
StrictWeakOrdering comp)
|
|
329
|
-
{
|
|
330
|
-
NV_IF_TARGET(NV_IS_DEVICE,
|
|
331
|
-
(
|
|
332
|
-
// avoid recursion in CUDA threads
|
|
333
|
-
stable_merge_sort_detail::iterative_stable_merge_sort(exec, first, last, comp);),
|
|
334
|
-
(stable_merge_sort_detail::recursive_stable_merge_sort(exec, first, last, comp);));
|
|
335
|
-
}
|
|
336
|
-
|
|
337
|
-
template <typename DerivedPolicy,
|
|
338
|
-
typename RandomAccessIterator1,
|
|
339
|
-
typename RandomAccessIterator2,
|
|
340
|
-
typename StrictWeakOrdering>
|
|
341
|
-
_CCCL_HOST_DEVICE void stable_merge_sort_by_key(
|
|
342
|
-
sequential::execution_policy<DerivedPolicy>& exec,
|
|
343
|
-
RandomAccessIterator1 first1,
|
|
344
|
-
RandomAccessIterator1 last1,
|
|
345
|
-
RandomAccessIterator2 first2,
|
|
346
|
-
StrictWeakOrdering comp)
|
|
347
|
-
{
|
|
348
|
-
NV_IF_TARGET(NV_IS_DEVICE,
|
|
349
|
-
(
|
|
350
|
-
// avoid recursion in CUDA threads
|
|
351
|
-
stable_merge_sort_detail::iterative_stable_merge_sort_by_key(exec, first1, last1, first2, comp);),
|
|
352
|
-
(stable_merge_sort_detail::recursive_stable_merge_sort_by_key(exec, first1, last1, first2, comp);));
|
|
353
|
-
}
|
|
354
|
-
|
|
355
|
-
} // namespace system::detail::sequential
|
|
356
|
-
THRUST_NAMESPACE_END
|