Sciweavers

CC
2008
Springer
14 years 1 months ago
Generating SIMD Vectorized Permutations
Abstract. This paper introduces a method to generate efficient vectorized implementations of small stride permutations using only vector load and vector shuffle instructions. These...
Franz Franchetti, Markus Püschel
MICRO
2005
IEEE
130views Hardware» more  MICRO 2005»
14 years 5 months ago
Exploiting Vector Parallelism in Software Pipelined Loops
An emerging trend in processor design is the addition of short vector instructions to general-purpose and embedded ISAs. Frequently, these extensions are employed using traditiona...
Samuel Larsen, Rodric M. Rabbah, Saman P. Amarasin...