Abstract : In this paper we put forward an annotation system for specifying a sequence of data layout transformations for loop vectorization. We propose four basic primitives for data layout transformations that programmers can compose to achieve complex data layout transformations. Our system automatically modifies all loops and other code operating on the transformed arrays. In addition, we propose data layout aware loop transformations to reduce the overhead of address computation and help vectorization. Taking the Scalar Penta-diagonal (SP) solver, from the NAS Parallel Benchmarks as a case study, we show that the programmer can achieve significant speedups using our annotations.
https://hal.inria.fr/hal-01403118 Contributor : Hal IfipConnect in order to contact the contributor Submitted on : Friday, November 25, 2016 - 2:38:49 PM Last modification on : Thursday, March 5, 2020 - 5:40:15 PM Long-term archiving on: : Tuesday, March 21, 2017 - 2:19:24 AM
Shixiong Xu, David Gregg. Semi-automatic Composition of Data Layout Transformations for Loop Vectorization. 11th IFIP International Conference on Network and Parallel Computing (NPC), Sep 2014, Ilan, Taiwan. pp.485-496, ⟨10.1007/978-3-662-44917-2_40⟩. ⟨hal-01403118⟩