Guang-Huei Lin

  • Citations Per Year
Learn More
With mostly non-sequential reference, memory access becomes the bottleneck of SIMD processor performance. In this paper, we present an approach for optimal H.264 motion estimation code generation over a SIMD platform with the objective to minimize memory access overhead. Specifically, we formulate the code generation task as a constrained optimization(More)
PLX is a native SIMD instruction set architecture, which integrates scalar instructions into a multimedia extension SIMD core to run OS and multimedia functions. It changes the typical CPU-DSP dual core architecture to a SIMDI/OCTRL architecture. We have developed a PLXbased system developing platform, including a PLX core, a RISC-like 8051 as I/O(More)
A SIMD platform provides higher computing power by executing multiple data simultaneously, but this feature is making the design and verification harder. This paper aims at helping designer with an efficient methodology for verifying and evaluating the performance of a SIMD PLX platform design. Using Mathematica, a computer algebra system (CAS), and reverse(More)
  • 1