DAGDA Decoupling Address Generation from Loads and Stores

DAGDA exposes some of the hidden operations that the hardware uses when performing loads and stores to the compiler to save energy and increase performance. We decouple the micro-operations for loads and stores into two operations: the first, the "prepare to access memory" instruction, or...

Full description

Bibliographic Details
Other Authors: Stokes, Michael (author)
Format: Others
Language:English
English
Published: Florida State University
Subjects:
Online Access:http://purl.flvc.org/fsu/fd/2018_Su_Stokes_fsu_0071N_14269
id ndltd-fsu.edu-oai-fsu.digital.flvc.org-fsu_647307
record_format oai_dc
spelling ndltd-fsu.edu-oai-fsu.digital.flvc.org-fsu_6473072019-04-05T00:06:09Z DAGDA Decoupling Address Generation from Loads and Stores Stokes, Michael (author) Whalley, David B. (professor directing thesis) Liu, Xiuwen, 1966- (committee member) Tyson, Gary Scott (committee member) Florida State University (degree granting institution) College of Arts and Sciences (degree granting college) Department of Computer Science (degree granting departmentdgg) Text text master thesis Florida State University English eng 1 online resource (33 pages) computer application/pdf DAGDA exposes some of the hidden operations that the hardware uses when performing loads and stores to the compiler to save energy and increase performance. We decouple the micro-operations for loads and stores into two operations: the first, the "prepare to access memory" instruction, or "pam", checks to see if a line is resident in the L1 DC and determines its way in the L1 DC data array, if it exists. The second operations performs the actual data access. This allows us to both save energy using compiler optimization techniques and improve performance because "pam" operations are a natural way of prefetching data into the L1 DC A Thesis submitted to the Department of Computer Science in partial fulfillment of the requirements for the degree of Master of Science. Summer Semester 2018. May 4, 2018. Includes bibliographical references. David B. Whalley, Professor Directing Thesis; Xiuwen Liu, Committee Member; Gary Tyson, Committee Member. Computer science 2018_Su_Stokes_fsu_0071N_14269 http://purl.flvc.org/fsu/fd/2018_Su_Stokes_fsu_0071N_14269 http://diginole.lib.fsu.edu/islandora/object/fsu%3A647307/datastream/TN/view/DAGDA%20Decoupling%20Address%20Generation%20from%20Loads%20and%20Stores.jpg
collection NDLTD
language English
English
format Others
sources NDLTD
topic Computer science
spellingShingle Computer science
DAGDA Decoupling Address Generation from Loads and Stores
description DAGDA exposes some of the hidden operations that the hardware uses when performing loads and stores to the compiler to save energy and increase performance. We decouple the micro-operations for loads and stores into two operations: the first, the "prepare to access memory" instruction, or "pam", checks to see if a line is resident in the L1 DC and determines its way in the L1 DC data array, if it exists. The second operations performs the actual data access. This allows us to both save energy using compiler optimization techniques and improve performance because "pam" operations are a natural way of prefetching data into the L1 DC === A Thesis submitted to the Department of Computer Science in partial fulfillment of the requirements for the degree of Master of Science. === Summer Semester 2018. === May 4, 2018. === Includes bibliographical references. === David B. Whalley, Professor Directing Thesis; Xiuwen Liu, Committee Member; Gary Tyson, Committee Member.
author2 Stokes, Michael (author)
author_facet Stokes, Michael (author)
title DAGDA Decoupling Address Generation from Loads and Stores
title_short DAGDA Decoupling Address Generation from Loads and Stores
title_full DAGDA Decoupling Address Generation from Loads and Stores
title_fullStr DAGDA Decoupling Address Generation from Loads and Stores
title_full_unstemmed DAGDA Decoupling Address Generation from Loads and Stores
title_sort dagda decoupling address generation from loads and stores
publisher Florida State University
url http://purl.flvc.org/fsu/fd/2018_Su_Stokes_fsu_0071N_14269
_version_ 1719015943254835200