Add examples/cccl_cooperative_block_reduce.py

Author: oleksandr-pavlyk

python/examples/cccl_cooperative_block_reduce.py

@@ -0,0 +1,103 @@
+# Copyright 2025 NVIDIA Corporation
+#
+#  Licensed under the Apache License, Version 2.0 with the LLVM exception
+#  (the "License"); you may not use this file except in compliance with
+#  the License.
+#
+#  You may obtain a copy of the License at
+#
+#      http://llvm.org/foundation/relicensing/LICENSE.txt
+#
+#  Unless required by applicable law or agreed to in writing, software
+#  distributed under the License is distributed on an "AS IS" BASIS,
+#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#  See the License for the specific language governing permissions and
+#  limitations under the License.
+
+import sys
+
+import cuda.cccl.cooperative.experimental as coop
+import cuda.nvbench as nvbench
+import numba
+import numpy as np
+from numba import cuda
+from pynvjitlink import patch
+
+
+class BitsetRing:
+    """
+    Addition operation over ring fixed width unsigned integers
+    with ring_plus = bitwise_or and ring_mul = bitwise_and,
+         ring_zero = 0, ring_one = -1
+    """
+
+    def __init__(self):
+        self.dt = np.uint64
+        self.zero = self.dt(0)
+        self.one = np.bitwise_invert(self.zero)
+
+    @staticmethod
+    def add(op1, op2):
+        return op1 | op2
+
+    @staticmethod
+    def mul(op1, op2):
+        return op1 & op2
+
+
+def as_cuda_Stream(cs: nvbench.CudaStream) -> cuda.cudadrv.driver.Stream:
+    return cuda.external_stream(cs.addressof())
+
+
+def multi_block_bench(state: nvbench.State):
+    threads_per_block = state.get_int64("ThreadsPerBlock")
+    num_blocks = state.get_int64("NumBlocks")
+    total_elements = threads_per_block * num_blocks
+
+    if total_elements > 2**26:
+        state.skip(reason="Memory footprint over threshold")
+        return
+
+    ring = BitsetRing()
+    block_reduce = coop.block.reduce(numba.uint64, threads_per_block, BitsetRing.add)
+
+    @cuda.jit(link=block_reduce.files)
+    def kernel(inp_arr, out_arr):
+        # Each thread contributes one element
+        block_idx = cuda.blockIdx.x
+        thread_idx = cuda.threadIdx.x
+        global_idx = block_idx * threads_per_block + thread_idx
+
+        block_output = block_reduce(inp_arr[global_idx])
+
+        # Only thread 0 of each block writes the result
+        if thread_idx == 0:
+            out_arr[block_idx] = block_output
+
+    h_inp = np.arange(1, total_elements + 1, dtype=ring.dt)
+    d_inp = cuda.to_device(h_inp)
+    d_out = cuda.device_array(num_blocks, dtype=ring.dt)
+
+    cuda_s = as_cuda_Stream(state.get_stream())
+    # warmup
+    kernel[num_blocks, threads_per_block, cuda_s, 0](d_inp, d_out)
+
+    state.add_element_count(total_elements)
+    state.add_global_memory_reads(total_elements * h_inp.itemsize)
+    state.add_global_memory_writes(num_blocks * h_inp.itemsize)
+
+    def launcher(launch: nvbench.Launch):
+        cuda_s = as_cuda_Stream(launch.get_stream())
+        kernel[num_blocks, threads_per_block, cuda_s, 0](d_inp, d_out)
+
+    state.exec(launcher)
+
+
+if __name__ == "__main__":
+    patch.patch_numba_linker(lto=True)
+
+    b = nvbench.register(multi_block_bench)
+    b.add_int64_axis("ThreadsPerBlock", [64, 128, 192, 256])
+    b.add_int64_power_of_two_axis("NumBlocks", [10, 11, 12, 14, 16])
+
+    nvbench.run_all_benchmarks(sys.argv)