Release v13.3 of the CUDA samples with CUDA 13.3 Toolkit (#435)

This is the release of the CUDA 13.3 samples, which include additions for CUDA Tile C++, and updated CCCL and Python samples.

python/2_CoreConcepts/prefixSum/README.md

@@ -8,7 +8,7 @@ Demonstrates parallel prefix sum (scan) algorithms using cuda.compute with cuda.
 - Exclusive scan: `output[i] = init_value + input[0] + input[1] + ... + input[i-1]`
 - Uses cuda.compute APIs for optimized CUB-based implementations
 - Uses cuda.core APIs for device and stream management
-- Demonstrates CuPy integration via `ExternalStream`
+- Demonstrates CuPy integration via `Stream.from_external`
 
 ## Requirements
 
@@ -21,9 +21,9 @@ Demonstrates parallel prefix sum (scan) algorithms using cuda.compute with cuda.
 - CUDA Toolkit 13.0+
 - Python 3.10+
 - `cuda-python` (13.0.0+)
-- `cuda-core` (>=0.6.0)
+- `cuda-core` (>=1.0.0)
 - `cuda-cccl` (1.0.0+)
-- `cupy-cuda13x` (13.0.0+)
+- `cupy-cuda13x` (14.0.0+)
 - `numpy` (>=2.3.2)
 
 ## Usage
@@ -56,8 +56,8 @@ This sample demonstrates proper stream usage across libraries:
 # Create stream with cuda.core
 stream = device.create_stream()
 
-# Wrap for CuPy compatibility (requires int handle)
-cp_stream = cp.cuda.ExternalStream(int(stream.handle))
+# Wrap for CuPy compatibility (cuda.core Stream implements the __cuda_stream__ protocol)
+cp_stream = cp.cuda.Stream.from_external(stream)
 
 # Use with CuPy operations
 with cp_stream:

python/2_CoreConcepts/prefixSum/prefixSum.py

@@ -62,7 +62,7 @@ def main() -> bool:
     device = Device(0)
     device.set_current()
     stream = device.create_stream()
-    cp_stream = cp.cuda.ExternalStream(int(stream.handle))
+    cp_stream = cp.cuda.Stream.from_external(stream)
 
     ok = True
     try:
@@ -153,7 +153,7 @@ def main() -> bool:
         )
         stream.sync()
 
-        event_opts = EventOptions(enable_timing=True)
+        event_opts = EventOptions(timing_enabled=True)
         start_event = device.create_event(options=event_opts)
         end_event = device.create_event(options=event_opts)
 
@@ -184,7 +184,7 @@ def main() -> bool:
         print("• Inclusive: output[i] includes input[i]")
         print("• Exclusive: output[i] excludes input[i], starts with init_value")
         print("• cuda.compute provides CUB-based optimized implementations")
-        print("• cuda.core Stream integrates with CuPy via ExternalStream")
+        print("• cuda.core Stream integrates with CuPy via Stream.from_external")
         print("• Applications: stream compaction, radix sort, histograms")
         print("=" * 60)
         return ok

python/2_CoreConcepts/prefixSum/requirements.txt

@@ -2,7 +2,7 @@
 # Requires Python 3.10+, CUDA Toolkit 13.0+
 
 cuda-python>=13.0.0
-cuda-core>=0.6.0
+cuda-core>=1.0.0
 cuda-cccl>=1.0.0
-cupy-cuda13x>=13.0.0
+cupy-cuda13x>=14.0.0
 numpy>=2.3.2