llvm-project

History

Tobias Grosser 6a2da6b9c8 Add test cases for multi-dimensional variable lengths arrays At the moment we can handle such arrays only by conservatively assuming that each access to such an array may touch any element in the array. It would be great if we could improve Polly/LLVM at some point, such that we can recover the multi-dimensionality of the accesses. llvm-svn: 163619		2012-09-11 14:03:19 +00:00
..
autoconf	autoconf: Only define GPGPU_CODEGEN, if that feature is requested	2012-08-21 12:29:10 +00:00
cmake	Add support for libpluto as the scheduling optimizer.	2012-08-02 07:47:26 +00:00
docs	Add initial version of Polly	2011-04-29 06:27:02 +00:00
include	Remove dead code	2012-09-08 14:00:32 +00:00
lib	ScopInfo: Align parameters when using -polly-allow-nonaffine	2012-09-11 13:50:21 +00:00
test	Add test cases for multi-dimensional variable lengths arrays	2012-09-11 14:03:19 +00:00
tools	Update libGPURuntime to be dual licensed under MIT and UIUC license.	2012-07-06 10:40:15 +00:00
utils	Update isl to a newer version	2012-09-03 07:42:40 +00:00
www	Remove executable bits from html files	2012-08-15 05:50:24 +00:00
CMakeLists.txt	Add preliminary implementation for GPGPU code generation.	2012-08-03 12:50:07 +00:00
CREDITS.txt	(Test commit for polly)	2011-07-16 13:30:03 +00:00
LICENSE.txt	Happy new year 2012!	2012-01-01 08:16:56 +00:00
Makefile	Revert "Fix a bug introduced by r153739: We are not able to provide the correct"	2012-04-11 07:43:13 +00:00
Makefile.common.in	Add initial version of Polly	2011-04-29 06:27:02 +00:00
Makefile.config.in	Add support for libpluto as the scheduling optimizer.	2012-08-02 07:47:26 +00:00
README	Remove some empty lines	2011-10-04 06:56:36 +00:00
configure	autoconf: Only define GPGPU_CODEGEN, if that feature is requested	2012-08-21 12:29:10 +00:00

README

Polly - Polyhedral optimizations for LLVM

Polly uses a mathematical representation, the polyhedral model, to represent and
transform loops and other control flow structures. Using an abstract
representation it is possible to reason about transformations in a more general
way and to use highly optimized linear programming libraries to figure out the
optimal loop structure. These transformations can be used to do constant
propagation through arrays, remove dead loop iterations, optimize loops for
cache locality, optimize arrays, apply advanced automatic parallelization, drive
vectorization, or they can be used to do software pipelining.