cuda-cccl 0.3.2__cp313-cp313-manylinux_2_24_aarch64.whl → 0.3.4__cp313-cp313-manylinux_2_24_aarch64.whl

cuda/cccl/headers/include/cub/block/specializations/block_histogram_atomic.cuh

@@ -1,30 +1,6 @@
-/******************************************************************************
- * Copyright (c) 2011, Duane Merrill.  All rights reserved.
- * Copyright (c) 2011-2018, NVIDIA CORPORATION.  All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of the NVIDIA CORPORATION nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
- * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- ******************************************************************************/
+// SPDX-FileCopyrightText: Copyright (c) 2011, Duane Merrill. All rights reserved.
+// SPDX-FileCopyrightText: Copyright (c) 2011-2018, NVIDIA CORPORATION. All rights reserved.
+// SPDX-License-Identifier: BSD-3
 
 /**
  * @file
@@ -51,7 +27,7 @@ namespace detail
  * @brief The BlockHistogramAtomic class provides atomic-based methods for constructing block-wide
  *        histograms from data samples partitioned across a CUDA thread block.
  */
-template <int BINS>
+template <int Bins>
 struct BlockHistogramAtomic
 {
   /// Shared memory storage layout type
@@ -71,7 +47,7 @@ struct BlockHistogramAtomic
    *   Reference to shared/device-accessible memory histogram
    */
   template <typename T, typename CounterT, int ITEMS_PER_THREAD>
-  _CCCL_DEVICE _CCCL_FORCEINLINE void Composite(T (&items)[ITEMS_PER_THREAD], CounterT histogram[BINS])
+  _CCCL_DEVICE _CCCL_FORCEINLINE void Composite(T (&items)[ITEMS_PER_THREAD], CounterT histogram[Bins])
   {
     // Update histogram
     _CCCL_PRAGMA_UNROLL_FULL()

cuda/cccl/headers/include/cub/block/specializations/block_histogram_sort.cuh

@@ -1,30 +1,6 @@
-/******************************************************************************
- * Copyright (c) 2011, Duane Merrill.  All rights reserved.
- * Copyright (c) 2011-2018, NVIDIA CORPORATION.  All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of the NVIDIA CORPORATION nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
- * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- ******************************************************************************/
+// SPDX-FileCopyrightText: Copyright (c) 2011, Duane Merrill. All rights reserved.
+// SPDX-FileCopyrightText: Copyright (c) 2011-2018, NVIDIA CORPORATION. All rights reserved.
+// SPDX-License-Identifier: BSD-3
 
 /**
  * @file
@@ -58,46 +34,46 @@ namespace detail
  * @tparam T
  *   Sample type
  *
- * @tparam BLOCK_DIM_X
+ * @tparam BlockDimX
  *   The thread block length in threads along the X dimension
  *
- * @tparam ITEMS_PER_THREAD
+ * @tparam ItemsPerThread
  *   The number of samples per thread
  *
- * @tparam BINS
+ * @tparam Bins
  *   The number of bins into which histogram samples may fall
  *
- * @tparam BLOCK_DIM_Y
+ * @tparam BlockDimY
  *   The thread block length in threads along the Y dimension
  *
- * @tparam BLOCK_DIM_Z
+ * @tparam BlockDimZ
  *   The thread block length in threads along the Z dimension
  */
-template <typename T, int BLOCK_DIM_X, int ITEMS_PER_THREAD, int BINS, int BLOCK_DIM_Y, int BLOCK_DIM_Z>
+template <typename T, int BlockDimX, int ItemsPerThread, int Bins, int BlockDimY, int BlockDimZ>
 struct BlockHistogramSort
 {
   /// Constants
   enum
   {
     /// The thread block size in threads
-    BLOCK_THREADS = BLOCK_DIM_X * BLOCK_DIM_Y * BLOCK_DIM_Z,
+    BLOCK_THREADS = BlockDimX * BlockDimY * BlockDimZ,
   };
 
   // Parameterize BlockRadixSort type for our thread block
   using BlockRadixSortT =
     BlockRadixSort<T,
-                   BLOCK_DIM_X,
-                   ITEMS_PER_THREAD,
+                   BlockDimX,
+                   ItemsPerThread,
                    NullType,
                    4,
                    true,
                    BLOCK_SCAN_WARP_SCANS,
                    cudaSharedMemBankSizeFourByte,
-                   BLOCK_DIM_Y,
-                   BLOCK_DIM_Z>;
+                   BlockDimY,
+                   BlockDimZ>;
 
   // Parameterize BlockDiscontinuity type for our thread block
-  using BlockDiscontinuityT = BlockDiscontinuity<T, BLOCK_DIM_X, BLOCK_DIM_Y, BLOCK_DIM_Z>;
+  using BlockDiscontinuityT = BlockDiscontinuity<T, BlockDimX, BlockDimY, BlockDimZ>;
 
   /// Shared memory
   union _TempStorage
@@ -111,8 +87,8 @@ struct BlockHistogramSort
       typename BlockDiscontinuityT::TempStorage flag;
 
       // Storage for noting begin/end offsets of bin runs in the tile of sorted bin values
-      unsigned int run_begin[BINS];
-      unsigned int run_end[BINS];
+      unsigned int run_begin[Bins];
+      unsigned int run_end[Bins];
     } discontinuities;
   };
 
@@ -127,7 +103,7 @@ struct BlockHistogramSort
   /// Constructor
   _CCCL_DEVICE _CCCL_FORCEINLINE BlockHistogramSort(TempStorage& temp_storage)
       : temp_storage(temp_storage.Alias())
-      , linear_tid(RowMajorTid(BLOCK_DIM_X, BLOCK_DIM_Y, BLOCK_DIM_Z))
+      , linear_tid(RowMajorTid(BlockDimX, BlockDimY, BlockDimZ))
   {}
 
   // Discontinuity functor
@@ -169,11 +145,11 @@ struct BlockHistogramSort
    *   Reference to shared/device-accessible memory histogram
    */
   template <typename CounterT>
-  _CCCL_DEVICE _CCCL_FORCEINLINE void Composite(T (&items)[ITEMS_PER_THREAD], CounterT histogram[BINS])
+  _CCCL_DEVICE _CCCL_FORCEINLINE void Composite(T (&items)[ItemsPerThread], CounterT histogram[Bins])
   {
     enum
     {
-      TILE_SIZE = BLOCK_THREADS * ITEMS_PER_THREAD
+      TILE_SIZE = BLOCK_THREADS * ItemsPerThread
     };
 
     // Sort bytes in blocked arrangement
@@ -185,13 +161,13 @@ struct BlockHistogramSort
     int histo_offset = 0;
 
     _CCCL_PRAGMA_UNROLL_FULL()
-    for (; histo_offset + BLOCK_THREADS <= BINS; histo_offset += BLOCK_THREADS)
+    for (; histo_offset + BLOCK_THREADS <= Bins; histo_offset += BLOCK_THREADS)
     {
       temp_storage.discontinuities.run_begin[histo_offset + linear_tid] = TILE_SIZE;
       temp_storage.discontinuities.run_end[histo_offset + linear_tid]   = TILE_SIZE;
     }
     // Finish up with guarded initialization if necessary
-    if ((BINS % BLOCK_THREADS != 0) && (histo_offset + linear_tid < BINS))
+    if ((Bins % BLOCK_THREADS != 0) && (histo_offset + linear_tid < Bins))
     {
       temp_storage.discontinuities.run_begin[histo_offset + linear_tid] = TILE_SIZE;
       temp_storage.discontinuities.run_end[histo_offset + linear_tid]   = TILE_SIZE;
@@ -199,7 +175,7 @@ struct BlockHistogramSort
 
     __syncthreads();
 
-    int flags[ITEMS_PER_THREAD]; // unused
+    int flags[ItemsPerThread]; // unused
 
     // Compute head flags to demarcate contiguous runs of the same bin in the sorted tile
     DiscontinuityOp flag_op(temp_storage);
@@ -217,7 +193,7 @@ struct BlockHistogramSort
     histo_offset = 0;
 
     _CCCL_PRAGMA_UNROLL_FULL()
-    for (; histo_offset + BLOCK_THREADS <= BINS; histo_offset += BLOCK_THREADS)
+    for (; histo_offset + BLOCK_THREADS <= Bins; histo_offset += BLOCK_THREADS)
     {
       int thread_offset = histo_offset + linear_tid;
       CounterT count =
@@ -226,7 +202,7 @@ struct BlockHistogramSort
     }
 
     // Finish up with guarded composition if necessary
-    if ((BINS % BLOCK_THREADS != 0) && (histo_offset + linear_tid < BINS))
+    if ((Bins % BLOCK_THREADS != 0) && (histo_offset + linear_tid < Bins))
     {
       int thread_offset = histo_offset + linear_tid;
       CounterT count =

cuda/cccl/headers/include/cub/block/specializations/block_reduce_raking.cuh

@@ -1,30 +1,6 @@
-/******************************************************************************
- * Copyright (c) 2011, Duane Merrill.  All rights reserved.
- * Copyright (c) 2011-2018, NVIDIA CORPORATION.  All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of the NVIDIA CORPORATION nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
- * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- ******************************************************************************/
+// SPDX-FileCopyrightText: Copyright (c) 2011, Duane Merrill. All rights reserved.
+// SPDX-FileCopyrightText: Copyright (c) 2011-2018, NVIDIA CORPORATION. All rights reserved.
+// SPDX-License-Identifier: BSD-3
 
 /**
  * @file
@@ -49,6 +25,8 @@
 #include <cub/util_ptx.cuh>
 #include <cub/warp/warp_reduce.cuh>
 
+#include <cuda/__cmath/pow2.h>
+
 CUB_NAMESPACE_BEGIN
 namespace detail
 {
@@ -69,20 +47,20 @@ namespace detail
  * @tparam T
  *   Data type being reduced
  *
- * @tparam BLOCK_DIM_X
+ * @tparam BlockDimX
  *   The thread block length in threads along the X dimension
  *
- * @tparam BLOCK_DIM_Y
+ * @tparam BlockDimY
  *   The thread block length in threads along the Y dimension
  *
- * @tparam BLOCK_DIM_Z
+ * @tparam BlockDimZ
  *   The thread block length in threads along the Z dimension
  */
-template <typename T, int BLOCK_DIM_X, int BLOCK_DIM_Y, int BLOCK_DIM_Z>
+template <typename T, int BlockDimX, int BlockDimY, int BlockDimZ>
 struct BlockReduceRaking
 {
   /// The thread block size in threads
-  static constexpr int BLOCK_THREADS = BLOCK_DIM_X * BLOCK_DIM_Y * BLOCK_DIM_Z;
+  static constexpr int BLOCK_THREADS = BlockDimX * BlockDimY * BlockDimZ;
 
   /// Layout type for padded thread block raking grid
   using BlockRakingLayout = BlockRakingLayout<T, BLOCK_THREADS>;
@@ -102,7 +80,7 @@ struct BlockReduceRaking
 
   /// Whether or not warp-synchronous reduction should be unguarded (i.e., the warp-reduction elements is a power of
   /// two
-  static constexpr int WARP_SYNCHRONOUS_UNGUARDED = PowerOfTwo<RAKING_THREADS>::VALUE;
+  static constexpr int WARP_SYNCHRONOUS_UNGUARDED = ::cuda::is_power_of_two(RAKING_THREADS);
 
   /// Whether or not accesses into smem are unguarded
   static constexpr bool RAKING_UNGUARDED = BlockRakingLayout::UNGUARDED;
@@ -128,7 +106,7 @@ struct BlockReduceRaking
   /// Constructor
   _CCCL_DEVICE _CCCL_FORCEINLINE BlockReduceRaking(TempStorage& temp_storage)
       : temp_storage(temp_storage.Alias())
-      , linear_tid(RowMajorTid(BLOCK_DIM_X, BLOCK_DIM_Y, BLOCK_DIM_Z))
+      , linear_tid(RowMajorTid(BlockDimX, BlockDimY, BlockDimZ))
   {}
 
   /**

cuda/cccl/headers/include/cub/block/specializations/block_reduce_raking_commutative_only.cuh

@@ -1,30 +1,6 @@
-/******************************************************************************
- * Copyright (c) 2011, Duane Merrill.  All rights reserved.
- * Copyright (c) 2011-2018, NVIDIA CORPORATION.  All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of the NVIDIA CORPORATION nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
- * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- ******************************************************************************/
+// SPDX-FileCopyrightText: Copyright (c) 2011, Duane Merrill. All rights reserved.
+// SPDX-FileCopyrightText: Copyright (c) 2011-2018, NVIDIA CORPORATION. All rights reserved.
+// SPDX-License-Identifier: BSD-3
 
 /**
  * @file
@@ -62,28 +38,28 @@ namespace detail
  * @tparam T
  *   Data type being reduced
  *
- * @tparam BLOCK_DIM_X
+ * @tparam BlockDimX
  *   The thread block length in threads along the X dimension
  *
- * @tparam BLOCK_DIM_Y
+ * @tparam BlockDimY
  *   The thread block length in threads along the Y dimension
  *
- * @tparam BLOCK_DIM_Z
+ * @tparam BlockDimZ
  *   The thread block length in threads along the Z dimension
  */
-template <typename T, int BLOCK_DIM_X, int BLOCK_DIM_Y, int BLOCK_DIM_Z>
+template <typename T, int BlockDimX, int BlockDimY, int BlockDimZ>
 struct BlockReduceRakingCommutativeOnly
 {
   /// Constants
   enum
   {
     /// The thread block size in threads
-    BLOCK_THREADS = BLOCK_DIM_X * BLOCK_DIM_Y * BLOCK_DIM_Z,
+    BLOCK_THREADS = BlockDimX * BlockDimY * BlockDimZ,
   };
 
   // The fall-back implementation to use when BLOCK_THREADS is not a multiple of the warp size or not all threads have
   // valid values
-  using FallBack = detail::BlockReduceRaking<T, BLOCK_DIM_X, BLOCK_DIM_Y, BLOCK_DIM_Z>;
+  using FallBack = detail::BlockReduceRaking<T, BlockDimX, BlockDimY, BlockDimZ>;
 
   /// Constants
   enum
@@ -137,7 +113,7 @@ struct BlockReduceRakingCommutativeOnly
   /// Constructor
   _CCCL_DEVICE _CCCL_FORCEINLINE BlockReduceRakingCommutativeOnly(TempStorage& temp_storage)
       : temp_storage(temp_storage.Alias())
-      , linear_tid(RowMajorTid(BLOCK_DIM_X, BLOCK_DIM_Y, BLOCK_DIM_Z))
+      , linear_tid(RowMajorTid(BlockDimX, BlockDimY, BlockDimZ))
   {}
 
   /**

cuda/cccl/headers/include/cub/block/specializations/block_reduce_warp_reductions.cuh

@@ -1,30 +1,6 @@
-/******************************************************************************
- * Copyright (c) 2011, Duane Merrill.  All rights reserved.
- * Copyright (c) 2011-2025, NVIDIA CORPORATION.  All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of the NVIDIA CORPORATION nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
- * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- ******************************************************************************/
+// SPDX-FileCopyrightText: Copyright (c) 2011, Duane Merrill. All rights reserved.
+// SPDX-FileCopyrightText: Copyright (c) 2011-2025, NVIDIA CORPORATION. All rights reserved.
+// SPDX-License-Identifier: BSD-3
 
 //! @rst
 //! @file

cuda/cccl/headers/include/cub/block/specializations/block_scan_raking.cuh

@@ -1,30 +1,6 @@
-/******************************************************************************
- * Copyright (c) 2011, Duane Merrill.  All rights reserved.
- * Copyright (c) 2011-2018, NVIDIA CORPORATION.  All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of the NVIDIA CORPORATION nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
- * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- ******************************************************************************/
+// SPDX-FileCopyrightText: Copyright (c) 2011, Duane Merrill. All rights reserved.
+// SPDX-FileCopyrightText: Copyright (c) 2011-2018, NVIDIA CORPORATION. All rights reserved.
+// SPDX-License-Identifier: BSD-3
 
 /**
  * @file
@@ -61,20 +37,20 @@ namespace detail
  * @tparam T
  *   Data type being scanned
  *
- * @tparam BLOCK_DIM_X
+ * @tparam BlockDimX
  *   The thread block length in threads along the X dimension
  *
- * @tparam BLOCK_DIM_Y
+ * @tparam BlockDimY
  *   The thread block length in threads along the Y dimension
  *
- * @tparam BLOCK_DIM_Z
+ * @tparam BlockDimZ
  *   The thread block length in threads along the Z dimension
  *
- * @tparam MEMOIZE
+ * @tparam Memoize
  *   Whether or not to buffer outer raking scan partials to incur fewer shared memory reads at the
  * expense of higher register pressure
  */
-template <typename T, int BLOCK_DIM_X, int BLOCK_DIM_Y, int BLOCK_DIM_Z, bool MEMOIZE>
+template <typename T, int BlockDimX, int BlockDimY, int BlockDimZ, bool Memoize>
 struct BlockScanRaking
 {
   //---------------------------------------------------------------------
@@ -82,7 +58,7 @@ struct BlockScanRaking
   //---------------------------------------------------------------------
 
   /// The thread block size in threads
-  static constexpr int BLOCK_THREADS = BLOCK_DIM_X * BLOCK_DIM_Y * BLOCK_DIM_Z;
+  static constexpr int BLOCK_THREADS = BlockDimX * BlockDimY * BlockDimZ;
 
   /// Layout type for padded thread block raking grid
   using BlockRakingLayout = BlockRakingLayout<T, BLOCK_THREADS>;
@@ -222,7 +198,7 @@ struct BlockScanRaking
     T* smem_raking_ptr = BlockRakingLayout::RakingPtr(temp_storage.raking_grid, linear_tid);
 
     // Read data back into registers
-    if (!MEMOIZE)
+    if constexpr (!Memoize)
     {
       CopySegment(cached_segment, smem_raking_ptr, constant_v<0>);
     }
@@ -240,7 +216,7 @@ struct BlockScanRaking
     T* smem_raking_ptr = BlockRakingLayout::RakingPtr(temp_storage.raking_grid, linear_tid);
 
     // Read data back into registers
-    if (!MEMOIZE)
+    if constexpr (!Memoize)
     {
       CopySegment(cached_segment, smem_raking_ptr, constant_v<0>);
     }
@@ -258,7 +234,7 @@ struct BlockScanRaking
   /// Constructor
   _CCCL_DEVICE _CCCL_FORCEINLINE BlockScanRaking(TempStorage& temp_storage)
       : temp_storage(temp_storage.Alias())
-      , linear_tid(RowMajorTid(BLOCK_DIM_X, BLOCK_DIM_Y, BLOCK_DIM_Z))
+      , linear_tid(RowMajorTid(BlockDimX, BlockDimY, BlockDimZ))
   {}
 
   //---------------------------------------------------------------------

cuda/cccl/headers/include/cub/block/specializations/block_scan_warp_scans.cuh

@@ -1,30 +1,6 @@
-/******************************************************************************
- * Copyright (c) 2011, Duane Merrill.  All rights reserved.
- * Copyright (c) 2011-2018, NVIDIA CORPORATION.  All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of the NVIDIA CORPORATION nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
- * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- ******************************************************************************/
+// SPDX-FileCopyrightText: Copyright (c) 2011, Duane Merrill. All rights reserved.
+// SPDX-FileCopyrightText: Copyright (c) 2011-2018, NVIDIA CORPORATION. All rights reserved.
+// SPDX-License-Identifier: BSD-3
 
 /**
  * @file
@@ -57,16 +33,16 @@ namespace detail
  * @brief BlockScanWarpScans provides warpscan-based variants of parallel prefix scan across a CUDA
  *        thread block.
  *
- * @tparam BLOCK_DIM_X
+ * @tparam BlockDimX
  *   The thread block length in threads along the X dimension
  *
- * @tparam BLOCK_DIM_Y
+ * @tparam BlockDimY
  *   The thread block length in threads along the Y dimension
  *
- * @tparam BLOCK_DIM_Z
+ * @tparam BlockDimZ
  *   The thread block length in threads along the Z dimension
  */
-template <typename T, int BLOCK_DIM_X, int BLOCK_DIM_Y, int BLOCK_DIM_Z>
+template <typename T, int BlockDimX, int BlockDimY, int BlockDimZ>
 struct BlockScanWarpScans
 {
   //---------------------------------------------------------------------
@@ -78,7 +54,7 @@ struct BlockScanWarpScans
   static constexpr int WARP_THREADS = warp_threads;
 
   /// The thread block size in threads
-  static constexpr int BLOCK_THREADS = BLOCK_DIM_X * BLOCK_DIM_Y * BLOCK_DIM_Z;
+  static constexpr int BLOCK_THREADS = BlockDimX * BlockDimY * BlockDimZ;
 
   /// Number of active warps
   static constexpr int WARPS = ::cuda::ceil_div(BLOCK_THREADS, WARP_THREADS);
@@ -123,7 +99,7 @@ struct BlockScanWarpScans
   /// Constructor
   _CCCL_DEVICE _CCCL_FORCEINLINE BlockScanWarpScans(TempStorage& temp_storage)
       : temp_storage(temp_storage.Alias())
-      , linear_tid(RowMajorTid(BLOCK_DIM_X, BLOCK_DIM_Y, BLOCK_DIM_Z))
+      , linear_tid(RowMajorTid(BlockDimX, BlockDimY, BlockDimZ))
       , warp_id((WARPS == 1) ? 0 : linear_tid / WARP_THREADS)
       , lane_id(::cuda::ptx::get_sreg_laneid())
   {}

cuda/cccl/headers/include/cub/config.cuh

@@ -1,29 +1,5 @@
-/******************************************************************************
- * Copyright (c) 2020, NVIDIA CORPORATION.  All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of the NVIDIA CORPORATION nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
- * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- ******************************************************************************/
+// SPDX-FileCopyrightText: Copyright (c) 2020, NVIDIA CORPORATION. All rights reserved.
+// SPDX-License-Identifier: BSD-3
 
 /**
  * \file

cuda/cccl/headers/include/cub/cub.cuh

@@ -1,30 +1,6 @@
-/******************************************************************************
- * Copyright (c) 2011, Duane Merrill.  All rights reserved.
- * Copyright (c) 2011-2018, NVIDIA CORPORATION.  All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions are met:
- *     * Redistributions of source code must retain the above copyright
- *       notice, this list of conditions and the following disclaimer.
- *     * Redistributions in binary form must reproduce the above copyright
- *       notice, this list of conditions and the following disclaimer in the
- *       documentation and/or other materials provided with the distribution.
- *     * Neither the name of the NVIDIA CORPORATION nor the
- *       names of its contributors may be used to endorse or promote products
- *       derived from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
- * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
- * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
- * DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE FOR ANY
- * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
- * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
- * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
- * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- ******************************************************************************/
+// SPDX-FileCopyrightText: Copyright (c) 2011, Duane Merrill. All rights reserved.
+// SPDX-FileCopyrightText: Copyright (c) 2011-2018, NVIDIA CORPORATION. All rights reserved.
+// SPDX-License-Identifier: BSD-3
 
 /**
  * \file