b7c5481c55
This is the release of the CUDA 13.3 samples, which include additions for CUDA Tile C++, and updated CCCL and Python samples.
libcuxxRandom - libcu++ Random Distributions
Description
This sample demonstrates the random-number facilities added to libcu++ in CCCL. <cuda/std/random> now offers host- and device-compatible implementations of the standard C++ distributions (uniform, normal, Poisson, Bernoulli, and more) and backports the C++26 cuda::std::philox4x32 / philox4x64 engines. <cuda/random> adds cuda::pcg64 as an NVIDIA extension (the same generator NumPy uses by default). A kernel draws samples on each thread and the host computes empirical statistics, comparing them to the theoretical mean / variance / probability.
Key Concepts
CCCL 3.3, libcu++ Random, PCG, Philox, Device-Side PRNG
Supported SM Architectures
SM 7.0 SM 7.5 SM 8.0 SM 8.6 SM 8.9 SM 9.0 SM 10.0 SM 11.0 SM 12.0
Supported OSes
Linux, Windows
Supported CPU Architecture
x86_64, aarch64
CUDA APIs involved
CCCL libcu++
cuda::pcg64, cuda::std::philox4x32, cuda::std::uniform_real_distribution, cuda::std::normal_distribution, cuda::std::poisson_distribution, cuda::std::bernoulli_distribution
CUDA Runtime API
cudaMalloc, cudaFree, cudaMemcpy, cudaDeviceSynchronize, cudaGetDeviceProperties
Dependencies needed to build/run
CCCL 3.3+. Fetched automatically via CPM at configure time (pinned to v3.3.3). Override with -DCCCL_SOURCE_DIR=/path/to/cccl to use a local checkout.
Prerequisites
Download and install the CUDA Toolkit for your corresponding platform. Make sure the dependencies mentioned in Dependencies section above are installed.