Rafael Espindola 7372093fcc Two fixes to the vpermilvar optimization. ... The instcomine logic to handle vpermilvar's pd and 256 variants was incorrect. The _256 variants have indexes into the individual 128 bit lanes and in all cases it also has to mask out unused bits. llvm-svn: 207577		2014-04-29 20:41:54 +00:00
..
Hello	[Modules] Fix potential ODR violations by sinking the DEBUG_TYPE	2014-04-22 02:55:47 +00:00
InstCombine	Two fixes to the vpermilvar optimization.	2014-04-29 20:41:54 +00:00
Instrumentation	[C++] Use 'nullptr'.	2014-04-28 04:05:08 +00:00
IPO	Add slp vectorization to LTO passes	2014-04-29 19:35:39 +00:00
ObjCARC	[C++] Use 'nullptr'.	2014-04-28 04:05:08 +00:00
Scalar	Reapply r207271 without the testcase	2014-04-29 18:25:28 +00:00
Utils	Add optimization remarks to the loop unroller and vectorizer.	2014-04-29 14:27:31 +00:00
Vectorize	Fix vectorization remarks.	2014-04-29 20:06:10 +00:00
CMakeLists.txt
LLVMBuild.txt
Makefile