cuda-cccl 0.3.0__cp311-cp311-manylinux_2_24_aarch64.whl → 0.3.2__cp311-cp311-manylinux_2_24_aarch64.whl

cuda/{cccl/cooperative/experimental → coop}/block/_block_load_store.py

@@ -5,12 +5,12 @@
 
 import numba
 
-from cuda.cccl.cooperative.experimental._common import (
+from .._common import (
     make_binary_tempfile,
     normalize_dim_param,
     normalize_dtype_param,
 )
-from cuda.cccl.cooperative.experimental._types import (
+from .._types import (
     Algorithm,
     Dependency,
     DependentArray,
@@ -70,13 +70,13 @@ def load(dtype, threads_per_block, items_per_thread=1, algorithm="direct"):
         The code snippet below illustrates a striped load and store of 128 integer items by 32 threads, with
         each thread handling 4 integers.
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_block_load_store_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_block_load_store_api.py
             :language: python
             :dedent:
             :start-after: example-begin imports
             :end-before: example-end imports
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_block_load_store_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_block_load_store_api.py
             :language: python
             :dedent:
             :start-after: example-begin load_store
@@ -158,13 +158,13 @@ def store(dtype, threads_per_block, items_per_thread=1, algorithm="direct"):
         The code snippet below illustrates a striped load and store of 128 integer items by 32 threads, with
         each thread handling 4 integers.
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_block_load_store_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_block_load_store_api.py
             :language: python
             :dedent:
             :start-after: example-begin imports
             :end-before: example-end imports
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_block_load_store_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_block_load_store_api.py
             :language: python
             :dedent:
             :start-after: example-begin load_store

cuda/{cccl/cooperative/experimental → coop}/block/_block_merge_sort.py

@@ -6,12 +6,12 @@ from typing import TYPE_CHECKING, Callable, Literal, Union
 
 import numba
 
-from cuda.cccl.cooperative.experimental._common import (
+from .._common import (
     make_binary_tempfile,
     normalize_dim_param,
     normalize_dtype_param,
 )
-from cuda.cccl.cooperative.experimental._types import (
+from .._types import (
     Algorithm,
     Constant,
     Dependency,
@@ -41,7 +41,7 @@ def merge_sort_keys(
         are partitioned in a :ref:`blocked arrangement <flexible-data-arrangement>` across 128 threads
         where each thread owns 4 consecutive keys. We start by importing necessary modules:
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_block_merge_sort_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_block_merge_sort_api.py
             :language: python
             :dedent:
             :start-after: example-begin imports
@@ -49,7 +49,7 @@ def merge_sort_keys(
 
         Below is the code snippet that demonstrates the usage of the ``merge_sort_keys`` API:
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_block_merge_sort_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_block_merge_sort_api.py
             :language: python
             :dedent:
             :start-after: example-begin merge-sort

cuda/{cccl/cooperative/experimental → coop}/block/_block_radix_sort.py

@@ -6,7 +6,7 @@ from typing import TYPE_CHECKING, Tuple, Union
 
 import numba
 
-from cuda.cccl.cooperative.experimental._common import (
+from .._common import (
     CUB_BLOCK_SCAN_ALGOS,
     CudaSharedMemConfig,
     dim3,
@@ -14,7 +14,7 @@ from cuda.cccl.cooperative.experimental._common import (
     normalize_dim_param,
     normalize_dtype_param,
 )
-from cuda.cccl.cooperative.experimental._types import (
+from .._types import (
     Algorithm,
     Dependency,
     DependentArray,
@@ -140,7 +140,7 @@ def radix_sort_keys(dtype, threads_per_block, items_per_thread):
         are partitioned in a :ref:`blocked arrangement <flexible-data-arrangement>` across 128 threads
         where each thread owns 4 consecutive keys. We start by importing necessary modules:
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_block_radix_sort_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_block_radix_sort_api.py
             :language: python
             :dedent:
             :start-after: example-begin imports
@@ -148,7 +148,7 @@ def radix_sort_keys(dtype, threads_per_block, items_per_thread):
 
         Below is the code snippet that demonstrates the usage of the ``radix_sort_keys`` API:
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_block_radix_sort_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_block_radix_sort_api.py
             :language: python
             :dedent:
             :start-after: example-begin radix-sort
@@ -181,7 +181,7 @@ def radix_sort_keys_descending(dtype, threads_per_block, items_per_thread):
         are partitioned in a :ref:`blocked arrangement <flexible-data-arrangement>` across 128 threads
         where each thread owns 4 consecutive keys. We start by importing necessary modules:
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_block_radix_sort_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_block_radix_sort_api.py
             :language: python
             :dedent:
             :start-after: example-begin imports
@@ -189,7 +189,7 @@ def radix_sort_keys_descending(dtype, threads_per_block, items_per_thread):
 
         Below is the code snippet that demonstrates the usage of the ``radix_sort_keys`` API:
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_block_radix_sort_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_block_radix_sort_api.py
             :language: python
             :dedent:
             :start-after: example-begin radix-sort-descending

cuda/{cccl/cooperative/experimental → coop}/block/_block_reduce.py

@@ -6,13 +6,13 @@ from typing import TYPE_CHECKING, Callable, Literal, Tuple, Union
 
 import numba
 
-from cuda.cccl.cooperative.experimental._common import (
+from .._common import (
     CUB_BLOCK_REDUCE_ALGOS,
     make_binary_tempfile,
     normalize_dim_param,
     normalize_dtype_param,
 )
-from cuda.cccl.cooperative.experimental._types import (
+from .._types import (
     Algorithm,
     Dependency,
     DependentArray,
@@ -208,13 +208,13 @@ def reduce(
         The code snippet below illustrates a max reduction of 128 integer items that are
         partitioned across 128 threads.
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_block_reduce_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_block_reduce_api.py
             :language: python
             :dedent:
             :start-after: example-begin imports
             :end-before: example-end imports
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_block_reduce_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_block_reduce_api.py
             :language: python
             :dedent:
             :start-after: example-begin reduce
@@ -269,13 +269,13 @@ def sum(
         The code snippet below illustrates a sum of 128 integer items that are partitioned
         across 128 threads.
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_block_reduce_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_block_reduce_api.py
             :language: python
             :dedent:
             :start-after: example-begin imports
             :end-before: example-end imports
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_block_reduce_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_block_reduce_api.py
             :language: python
             :dedent:
             :start-after: example-begin sum

cuda/{cccl/cooperative/experimental → coop}/block/_block_scan.py

@@ -3,7 +3,7 @@
 # SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 
 """
-cuda.cccl.cooperative.block_scan
+cuda.coop.block_scan
 ===========================
 
 This module provides a set of :ref:`collective <collective-primitives>`
@@ -73,16 +73,16 @@ from typing import Any, Callable, Literal
 
 import numba
 
-from cuda.cccl.cooperative.experimental._common import (
+from .._common import (
     CUB_BLOCK_SCAN_ALGOS,
     make_binary_tempfile,
     normalize_dim_param,
     normalize_dtype_param,
 )
-from cuda.cccl.cooperative.experimental._scan_op import (
+from .._scan_op import (
     ScanOp,
 )
-from cuda.cccl.cooperative.experimental._types import (
+from .._types import (
     Algorithm,
     Dependency,
     DependentArray,
@@ -94,7 +94,7 @@ from cuda.cccl.cooperative.experimental._types import (
     TemplateParameter,
     numba_type_to_wrapper,
 )
-from cuda.cccl.cooperative.experimental._typing import (
+from .._typing import (
     DimType,
     DtypeType,
     ScanOpType,
@@ -669,7 +669,7 @@ def exclusive_sum(
         :ref:`blocked arrangement <flexible-data-arrangement>` across 128
         threads where each thread owns 4 consecutive items.
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_block_scan_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_block_scan_api.py
             :language: python
             :dedent:
             :start-after: example-begin imports
@@ -678,7 +678,7 @@ def exclusive_sum(
         Below is the code snippet that demonstrates the usage of the
         ``exclusive_sum`` API:
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_block_scan_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_block_scan_api.py
             :language: python
             :dedent:
             :start-after: example-begin exclusive-sum

cuda/coop/warp/__init__.py

@@ -0,0 +1,9 @@
+# Copyright (c) 2024, NVIDIA CORPORATION & AFFILIATES. ALL RIGHTS RESERVED.
+#
+# SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+
+from ._warp_merge_sort import merge_sort_keys
+from ._warp_reduce import reduce, sum
+from ._warp_scan import exclusive_sum
+
+__all__ = ["exclusive_sum", "reduce", "sum", "merge_sort_keys"]

cuda/{cccl/cooperative/experimental → coop}/warp/_warp_merge_sort.py

@@ -4,8 +4,8 @@
 
 import numba
 
-from cuda.cccl.cooperative.experimental._common import make_binary_tempfile
-from cuda.cccl.cooperative.experimental._types import (
+from .._common import make_binary_tempfile
+from .._types import (
     Algorithm,
     Constant,
     Dependency,
@@ -30,7 +30,7 @@ def merge_sort_keys(
 
         Below is the code snippet that demonstrates the usage of the ``merge_sort_keys`` API:
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_warp_merge_sort_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_warp_merge_sort_api.py
             :language: python
             :dedent:
             :start-after: example-begin merge-sort

cuda/{cccl/cooperative/experimental → coop}/warp/_warp_reduce.py

@@ -4,8 +4,8 @@
 
 import numba
 
-from cuda.cccl.cooperative.experimental._common import make_binary_tempfile
-from cuda.cccl.cooperative.experimental._types import (
+from .._common import make_binary_tempfile
+from .._types import (
     Algorithm,
     Dependency,
     DependentPythonOperator,
@@ -28,7 +28,7 @@ def reduce(dtype, binary_op, threads_in_warp=32, methods=None):
         The code snippet below illustrates a max reduction of 32 integer items that
         are partitioned across a warp of threads.
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_warp_reduce_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_warp_reduce_api.py
             :language: python
             :dedent:
             :start-after: example-begin imports
@@ -36,7 +36,7 @@ def reduce(dtype, binary_op, threads_in_warp=32, methods=None):
 
         Below is the code snippet that demonstrates the usage of the ``reduce`` API:
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_warp_reduce_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_warp_reduce_api.py
             :language: python
             :dedent:
             :start-after: example-begin reduce
@@ -100,7 +100,7 @@ def sum(dtype, threads_in_warp=32):
         The code snippet below illustrates a reduction of 32 integer items that
         are partitioned across a warp of threads.
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_warp_reduce_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_warp_reduce_api.py
             :language: python
             :dedent:
             :start-after: example-begin imports
@@ -108,7 +108,7 @@ def sum(dtype, threads_in_warp=32):
 
         Below is the code snippet that demonstrates the usage of the ``reduce`` API:
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_warp_reduce_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_warp_reduce_api.py
             :language: python
             :dedent:
             :start-after: example-begin sum

cuda/{cccl/cooperative/experimental → coop}/warp/_warp_scan.py

@@ -5,8 +5,8 @@
 
 import numba
 
-from cuda.cccl.cooperative.experimental._common import make_binary_tempfile
-from cuda.cccl.cooperative.experimental._types import (
+from .._common import make_binary_tempfile
+from .._types import (
     Algorithm,
     Dependency,
     DependentReference,
@@ -23,7 +23,7 @@ def exclusive_sum(dtype, threads_in_warp=32):
     Example:
         The code snippet below illustrates an exclusive prefix sum of 32 integer items:
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_warp_scan_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_warp_scan_api.py
             :language: python
             :dedent:
             :start-after: example-begin imports
@@ -31,7 +31,7 @@ def exclusive_sum(dtype, threads_in_warp=32):
 
         Below is the code snippet that demonstrates the usage of the ``exclusive_sum`` API:
 
-        .. literalinclude:: ../../python/cuda_cccl/tests/cooperative/test_warp_scan_api.py
+        .. literalinclude:: ../../python/cuda_cccl/tests/coop/test_warp_scan_api.py
             :language: python
             :dedent:
             :start-after: example-begin exclusive-sum

{cuda_cccl-0.3.0.dist-info → cuda_cccl-0.3.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: cuda-cccl
-Version: 0.3.0
+Version: 0.3.2
 Summary: CUDA Core Library for Python
 Author: NVIDIA Corporation
 Classifier: Programming Language :: Python :: 3 :: Only