NVIDIA Corporation, 2007. - 143 p.
This document is organized into the following chapters:
contains a general introduction to CUDA.
outlines the programming model.
describes its hardware implementation.
describes the CUDA API and runtime.
gives some guidance on how to achieve maximum performance.
illustrates the previous chapters by walking through the code of some simple example.
Appendixsgives the technical specifications of various devices.
lists the mathematical functions supported in CUDA.
lists the atomic functions supported in CUDA.
is the CUDA runtime API reference.
is the CUDA driver API reference.
gives more details on texture fetching.