Cost effective memory disambiguation for multimedia codes

Salamí, Esther; Corbal, Jesús; Álvarez, Carlos; Valero, Mateo

doi:10.1145/581630.581649

Cited by 3 publications

(3 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The basic technique is to introduce checks in the source code to disambiguate memory references at runtime. [SCAV02] extend this technique to disambiguate an entire loop instead of disambiguating at the level of iterations. This was proposed in the context of multimedia applications.…”

Section: C2 Software Runtime Dependency Analysismentioning

confidence: 99%

Runtime dependency analysis for loop pipelining in high-level synthesis

Alle

Morvan

Derrien

2013

Proceedings of the 50th Annual Design Automation Conference

View full text Add to dashboard Cite

Research on High-Level Synthesis has mainly focused on applications with statically determinable characteristics and current tools often perform poorly in presence of datadependent memory accesses. The reason is that they rely on conservative static scheduling strategies, which lead to inefficient implementations. In this work, we propose to address this issue by leveraging well-known techniques used in superscalar processors to perform runtime memory disambiguation. Our approach, implemented as a source-tosource transformation at the C level, demonstrates significant performance improvements for a moderate increase in area while retaining portability among HLS tools.

show abstract

Section: C2 Software Runtime Dependency Analysismentioning

confidence: 99%

Runtime dependency analysis for loop pipelining in high-level synthesis

Alle

Morvan

Derrien

2013

Proceedings of the 50th Annual Design Automation Conference

View full text Add to dashboard Cite

show abstract

“…The Dynamic Memory Interval Test (DMIT) [Salamí et al 2002] is a runtime memory-disambiguation technique that makes sense in the context of media applications, or other kind of programs where input and output data streams are usually disjointed. Disambiguation can be easily determined by dynamically analyzing the region domain of every load and store on each invocation of a loop.…”

Section: Introductionmentioning

confidence: 99%

Dynamic memory interval test vs. interprocedural pointer analysis in multimedia applications

Salamí

Valero

2005

ACM Trans. Archit. Code Optim.

Self Cite

View full text Add to dashboard Cite

Techniques to detect aliasing between access patterns of array elements are quite effective for many numeric applications. However, although multimedia codes usually follow very regular memory access patterns, current commercial compilers remain unsuccessful in disambiguating them due mainly to complex pointer references. The Dynamic Memory Interval Test is a runtime memory disambiguation technique that takes advantage of the specific behavior of multimedia memory access patterns. It evaluates whether or not the full loop is disambiguated by analyzing the region domain of each load or store before each invocation of the loop.This paper provides a detailed evaluation of the approach, compares it against an advanced interprocedural pointer analysis framework, and analyzes the possibility of using both techniques at the same time. Both techniques achieve similar speedups separately (1.25X in average for a 8-issue width architecture). Furthermore, they can be used together to improve performance (reaching an average speed-up of 1.32X). Results also confirm that memory disambiguation is a key optimization to exploit the available parallelism in multimedia codes, especially for wide-issue architectures (1.50X average speed-up when scaling from 4-to 12-issue width in contrast to a low 1.10X for the baseline compiler).

show abstract

“…Our internal release of the compiler also includes Pcode Interprocedural Pointer Analysis [11] and Cost Effective Memory Disambiguation [32]. Therefore, our scalar versions of code include memory disambiguation (inherent in the vector versions), which introduces an average performance speed-up of 1.32X (for a 8-issue width architecture) over the same codes compiled with the public release of Trimaran.…”

Section: Compilation and Simulation Frameworkmentioning

confidence: 99%