Analyzing CUDA's Compiler through the Visualization of Decoded GPU Binaries

Provided by: Eindhoven University of Technology
Topic: Hardware
Format: PDF
With GPU architectures becoming increasingly important due to their large number of parallel processors, NVIDIA's CUDA environment is becoming widely used to support general purpose applications. To efficiently use the parallel processing power, programmers need to efficiently parallelize and map their algorithms. The difficulty of this task leads to the idea to investigate CUDA's compiler. Part of the compiler in the CUDA tool-chain is entirely un-documented, as is its output. To draw conclusions on the behavior of this compiler, the resulting object code is re-verse engineered.

Find By Topic