[llvm-commits] CVS: llvm-www/pubs/2009-03-ACMSE-Superpage.html 2009-03-ACMSE-Superpage.pdf pubs.js

Sun Oct 25 22:01:09 PDT 2009

Changes in directory llvm-www/pubs:

2009-03-ACMSE-Superpage.html added (r1.1)
2009-03-ACMSE-Superpage.pdf added (r1.1)
pubs.js updated: 1.51 -> 1.52
---
Log message:

add another paper


---
Diffs of the changes:  (+56 -0)

 2009-03-ACMSE-Superpage.html |   49 +++++++++++++++++++++++++++++++++++++++++++
 2009-03-ACMSE-Superpage.pdf  |    0 
 pubs.js                      |    7 ++++++
 3 files changed, 56 insertions(+)


Index: llvm-www/pubs/2009-03-ACMSE-Superpage.html
diff -c /dev/null llvm-www/pubs/2009-03-ACMSE-Superpage.html:1.1
*** /dev/null	Mon Oct 26 00:01:04 2009

--- llvm-www/pubs/2009-03-ACMSE-Superpage.html	Mon Oct 26 00:00:54 2009
***************
*** 0 ****
--- 1,49 ----
+ <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
+ <html>
+ <head>
+   <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
+   <link rel="stylesheet" href="../llvm.css" type="text/css" media="screen">
+   <title>A Case for Compiler-driven Superpage Allocation</title>
+ </head>
+ <body>
+ 
+ <div class="pub_title">
+   A Case for Compiler-driven Superpage Allocation
+ </div>
+ <div class="pub_author">
+   Joshua Magee and Apan Qasem
+ </div>
+ 
+ <h2>Abstract:</h2>
+ <blockquote>
+ Most modern microprocessor-based systems provide support for superpages both at the hardware and software level. Judicious use of superpages can significantly cut down the number of TLB misses and improve overall system performance. However, indiscriminate superpage allocation results in page fragmentation and increased application footprint, which often outweigh the benefits of reduced TLB misses. Previous research has explored policies for smart allocation of superpages from an operating systems perspective. This paper presents a compiler-based strategy for automatic and profitable memory allocation via superpages. A significant advantage of a compiler-based approach is the availability of data-reuse information within an application. Our strategy employs data-locality analysis to estimate the TLB demands of a program and uses this metric to determine if the program will benefit from superpage allocation. Apart from its obvious utility in improving TLB performance, this !
 strategy can be used to improve the effectiveness of certain data-layout transformations and can be a useful tool in benchmarking and empirical tuning. We demonstrate the effectiveness of this strategy with experiments on an Intel Core 2 Duo with a two-level TLB.
+ </blockquote>
+ 
+ <h2>Published:</h2>
+ <blockquote>
+   "A Case for Compiler-driven Superpage Allocation"
+   <br>
+   Joshua Magee and Apan Qasem.
+   <br>
+ <i>
+ Proceedings of the 47th ACM Southeast Regional Conference (ACMSE09)
+ </i>, Mar 2009.
+ </blockquote>
+ <h2>Download:</h2>
+ <h3>Paper:</h3>
+ <ul>
+   <li><a href="2009-03-ACMSE-Superpage.pdf">
+   A Case for Compiler-driven Superpage Allocation
+   </a> (PDF)</li>
+ </ul>
+ 
+ 
+ <!-- *********************************************************************** -->
+ <hr>
+   <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
+   src="http://jigsaw.w3.org/css-validator/images/vcss" alt="Valid CSS!"></a>
+   <a href="http://validator.w3.org/check/referer"><img
+   src="http://www.w3.org/Icons/valid-html401" alt="Valid HTML 4.01!" /></a>
+ 
+ </body>
+ </html>


Index: llvm-www/pubs/2009-03-ACMSE-Superpage.pdf


Index: llvm-www/pubs/pubs.js
diff -u llvm-www/pubs/pubs.js:1.51 llvm-www/pubs/pubs.js:1.52
--- llvm-www/pubs/pubs.js:1.51	Sun Oct 25 23:51:40 2009
+++ llvm-www/pubs/pubs.js	Mon Oct 26 00:00:54 2009
@@ -110,6 +110,13 @@
    month: 3,
    year: 2009},
 
+   {url: '2009-03-ACMSE-Superpage.html',
+    title: 'A Case for Compiler-driven Superpage Allocation',
+    author: 'Joshua Magee and Apan Qasem',
+    published: "Proc. of the 47th ACM Southeast Regional Conference (ACMSE09)",
+    month: 3,
+    year: 2009},
+
    {url: '2009-02-PPoPP-MappingParallelism.html',
     title: 'Mapping parallelism to multi-cores: a machine learning based approach',
     author: "Zheng Wang and Michael F.P. O'Boyle",




