Andrea Di Biagio c0edcf7de8 [X86] Improved target combine rules for selecting horizontal add/sub. ... This patch slightly changes the algorithm introduced at revision 210477 to fix a problem where the algorithm was producing incorrect code for the VEX.256 encoded versions of horizontal add/sub. For these cases, we now try to split the two 256-bit vectors into 128-bit chunks before emitting horizontal add/sub dag nodes. Added a new test case into haddsub-2.ll. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@210545 91177308-0d34-0410-b5e6-96231b3b80d8		2014-06-10 16:42:57 +00:00
..
Analysis	Removing an "if (!this)" check from two print methods. The condition will	2014-06-09 22:53:16 +00:00
AsmParser
Bitcode
CodeGen	SelectionDAG: Don't use MVT::Other to determine legality of ISD::SELECT_CC	2014-06-10 16:01:29 +00:00
DebugInfo
ExecutionEngine
IR	Reorder Value and User fields to save 8 bytes of padding on 64-bit	2014-06-09 23:32:20 +00:00
IRReader
LineEditor
Linker
LTO	Build fix: remove initializeJumpInstrTablesPass() call from LTO	2014-06-07 20:39:53 +00:00
MC	Fix gcc warning (enumeral and non-enumeral type in conditional expression)	2014-06-09 07:35:07 +00:00
Object
Option
ProfileData
Support	APFloat: x - NaN needs to flip the signbit of NaN when x is a number.	2014-06-08 16:53:31 +00:00
TableGen	Anonymous definitions in foreach blocks triggered a 'def already exists'	2014-06-10 12:41:14 +00:00
Target	[X86] Improved target combine rules for selecting horizontal add/sub.	2014-06-10 16:42:57 +00:00
Transforms	Removing an "if (!this)" check from two print methods. The condition will	2014-06-09 22:53:16 +00:00
CMakeLists.txt
LLVMBuild.txt	ProfileData: Introduce the InstrProfReader interface and a text reader	2014-03-21 17:24:48 +00:00
Makefile