RUS  ENG
Full version
JOURNALS // Program Systems: Theory and Applications // Archive

Program Systems: Theory and Applications, 2020 Volume 11, Issue 3, Pages 17–31 (Mi ps368)

This article is cited in 2 papers

Mathematical Foundations of Programming

The loop fusion for data localization

B. Ya. Steinberg, O. B. Shteinberg, A. A. Vasilenko

Southern Federal University

Abstract: The loop merging can be used to improve data localization. Merging loops that share variables can speed up execution by reducing the number of cache misses. This transformation has been known for a long time, but compilers perform it only for the simplest cases.
Our improved algorithms use pre-transformations to correctly merge loops with different numbers of iterations and informational dependencies.

Key words and phrases: loop fusion, optimizing compiler, program transformations, data locality, memory access optimization.

UDC: 519.681.3:004.415.3
BBK: 32.972.11:32.971.32-043

MSC: Primary 97P30; Secondary 97P20, 97R40

Received: 29.04.2020
02.07.2020

DOI: 10.25209/2079-3316-2020-11-3-17-31



© Steklov Math. Inst. of RAS, 2025