In this paper we present the ongoing effort to implement C structure optimizations in GCC, its current status and future plans. The general idea of this set of optimizations is to adapt the layout of a data-structure to its access patterns in order to better utilize the cache by increasing spatial locality. These optimizations are known to have many variations and require wide va-riety of data to be analyzed. An initial imple-mentation of these optimizations in GCC was presented at the GCC Developers ’ Summit of 2005. Since then many changes took place, in-cluding updating to Tree-SSA and preparation for GCC mainline 4.3. Additional opportuni-ties were revealed after close analysis of mcf benchmark from Spec2000/Spec20061. Struct-reorg reli...
International audienceComputer scientists are always eager to have a powerful, robust and stable com...
International audienceRecent advances in research on compressed caches make them an attractive desig...
This artifact contains the software required to reproduce the experimental findings of the paper "BO...
This paper presents a design and imple-mentation of a whole-program interprocedu-ral optimizer built...
Tree SSA is a new optimization framework based on the Static Single Assignment (SSA) form that opera...
Tree SSA is a new optimization framework for GCC that allows the implementation of ma-chine and lang...
One of the most important compiler passes is a loop optimization. The GCC’s current loop optimizer i...
We describe an important memory optimization that arises in the presence of aggregate data structure...
Several microprocessors, used in digital signal processing and embedded devices, have lim-ited displ...
SSA form is a very important concept in compiler internal code representa- tion. Φ-functions are an ...
As the gap between processor and memory continues to grow Memory performance becomes a key performan...
Though compilers usually focus on optimizing for performance, the size of the generated code has onl...
We explore the use of compiler optimizations, which optimize the layout of instructions in memory. T...
Data structure splicing (DSS) refers to reorganizing data structures by merging or splitting them, r...
In the past decade, processor speed has become significantly faster than memory speed. Small, fast c...
International audienceComputer scientists are always eager to have a powerful, robust and stable com...
International audienceRecent advances in research on compressed caches make them an attractive desig...
This artifact contains the software required to reproduce the experimental findings of the paper "BO...
This paper presents a design and imple-mentation of a whole-program interprocedu-ral optimizer built...
Tree SSA is a new optimization framework based on the Static Single Assignment (SSA) form that opera...
Tree SSA is a new optimization framework for GCC that allows the implementation of ma-chine and lang...
One of the most important compiler passes is a loop optimization. The GCC’s current loop optimizer i...
We describe an important memory optimization that arises in the presence of aggregate data structure...
Several microprocessors, used in digital signal processing and embedded devices, have lim-ited displ...
SSA form is a very important concept in compiler internal code representa- tion. Φ-functions are an ...
As the gap between processor and memory continues to grow Memory performance becomes a key performan...
Though compilers usually focus on optimizing for performance, the size of the generated code has onl...
We explore the use of compiler optimizations, which optimize the layout of instructions in memory. T...
Data structure splicing (DSS) refers to reorganizing data structures by merging or splitting them, r...
In the past decade, processor speed has become significantly faster than memory speed. Small, fast c...
International audienceComputer scientists are always eager to have a powerful, robust and stable com...
International audienceRecent advances in research on compressed caches make them an attractive desig...
This artifact contains the software required to reproduce the experimental findings of the paper "BO...