scratch pad memory, affine reference This paper considers compiler management of fast, local memories for loop nests that contain affine array references. We show how to reduce the local memory footprint of such arrays by doing an affine reindexing of the array elements. We approximate the best reindexing by a two-step algorithm. The first step uses a generalized Hermite factorization, and the second uses a one-norm lattice basis reduction technique. We improve on earlier work in which a local memory array stores the smallest rectangular subarray of the elements referenced in the loop nest. Our techniques apply directly to nests with families of uniformly generated affine references, and we propose ways to extend them to more general situat...
AbstractÐExploiting locality of references has become extremely important in realizing the potential...
This paper adresses the problem of efficient mappings of nested loops, and more generally of system...
We have been investigating problems which arise in compiling single assignment labguages (in which m...
(eng) We investigate the technique of storing multiple array elements in the same memory cell, with ...
[[abstract]]This paper presents an efficient compilation technique to generate the local memory acce...
[[abstract]]Address generation for compiling programs, written in HPF, to executable SPMD code is an...
Existing techniques can enhance the locality of arrays indexed by affine functions of induction vari...
Efficient memory allocation is crucial for data-intensive applications, as a smaller memory footprin...
Scratch-Pad memory (SPM) allocators that exploit the presence of affine references to arrays are imp...
Given a regular application described by a system of uniform recurrence equations, systolic arrays a...
This paper presents a data layout optimization technique based on the theory of hyperplanes from lin...
Minimizing communication overhead when mapping affine loop nests onto distributed memory parallel co...
Global locality optimization is a technique for improving the cache performance of a sequence of loo...
Efficient memory usage is crucial for data-intensive applications as a smaller memory footprint ensu...
This thesis investigates compiler algorithms to transform program and data to utilize efficiently th...
AbstractÐExploiting locality of references has become extremely important in realizing the potential...
This paper adresses the problem of efficient mappings of nested loops, and more generally of system...
We have been investigating problems which arise in compiling single assignment labguages (in which m...
(eng) We investigate the technique of storing multiple array elements in the same memory cell, with ...
[[abstract]]This paper presents an efficient compilation technique to generate the local memory acce...
[[abstract]]Address generation for compiling programs, written in HPF, to executable SPMD code is an...
Existing techniques can enhance the locality of arrays indexed by affine functions of induction vari...
Efficient memory allocation is crucial for data-intensive applications, as a smaller memory footprin...
Scratch-Pad memory (SPM) allocators that exploit the presence of affine references to arrays are imp...
Given a regular application described by a system of uniform recurrence equations, systolic arrays a...
This paper presents a data layout optimization technique based on the theory of hyperplanes from lin...
Minimizing communication overhead when mapping affine loop nests onto distributed memory parallel co...
Global locality optimization is a technique for improving the cache performance of a sequence of loo...
Efficient memory usage is crucial for data-intensive applications as a smaller memory footprint ensu...
This thesis investigates compiler algorithms to transform program and data to utilize efficiently th...
AbstractÐExploiting locality of references has become extremely important in realizing the potential...
This paper adresses the problem of efficient mappings of nested loops, and more generally of system...
We have been investigating problems which arise in compiling single assignment labguages (in which m...