NVIDIA CUDA 13.3 adds native carryless multiplication support

NVIDIA has introduced a new PTX instruction for carryless multiplication in CUDA 13.3, addressing a long-standing gap where GPUs lacked native support for this operation despite its presence on x86 CPUs for over fifteen years.

This addition enables more efficient implementation of authenticated encryption, error-correcting codes, and modern zero-knowledge proofs on NVIDIA hardware.

By providing this primitive directly in the PTX instruction set, CUDA 13.3 allows developers to build faster cryptographic applications that leverage GPU acceleration.