Publicación (Conferencias y Seminarios)

Runtime multi-versioning and specialization inside a memoized speculative loop optimizer

Lazcano López, Raquel; Madroñal Quintín, Daniel; Juárez Martínez, Eduardo; Clauss, P.

Resumen:

In this paper, we propose a runtime framework that implements code multi-versioning and specialization to optimize and parallelize loop kernels that are invoked many times with varying parameters. These parameters may influence the code structure, the touched memory locations, the workload, and the runtime performance. They may also impact the validity of the parallelizing and optimizing polyhedral transformations that are applied on-the-fly. For a target loop kernel and its associated parameters, a different optimizing and parallelizing transformation is evaluated at each invocation, among a finite set of transformations (multi-versioning and specialization). The best performing transformed code version is stored and indexed using its associated parameters. When every optimizing transformation has been evaluated, the best performing code version regarding the current parameters, which has been stored, is relaunched at next invocations (memoization).

Áreas de investigación:

• Tecnologías de Vídeo e Imagen

Año:

2020

Tipo de publicación:

Conferencias y Seminarios

Palabras clave:

runtime speculative optimization and parallelization; multi-versioning; memoization; specialization; polyhedral model

Editor:

ACM

Dirección:

San Diego, Estados Unidos de América

Organización:

International Conference on Compiler Construction (CC)

Mes:

Febrero

DOI:

10.1145/3377555.3377886