Matthew Simpson 383f076bfc [AArch64] Consider widening instructions in cost calculations ... The AArch64 instruction set has a few "widening" instructions (e.g., uaddl, saddl, uaddw, etc.) that take one or more doubleword operands and produce quadword results. The operands are automatically sign- or zero-extended as appropriate. However, in LLVM IR, these extends are explicit. This patch updates TTI to consider these widening instructions as single operations whose cost is attached to the arithmetic instruction. It marks extends that are part of a widening operation "free" and applies a sub-target specified overhead (zero by default) to the arithmetic instructions. Differential Revision: https://reviews.llvm.org/D32706 llvm-svn: 302582		2017-05-09 20:18:12 +00:00
..
BBVectorize.cpp	[SystemZ] TargetTransformInfo cost functions implemented.	2017-04-12 11:49:08 +00:00
CMakeLists.txt	[CMake] NFC. Updating CMake dependency specifications	2016-11-17 04:36:50 +00:00
LLVMBuild.txt
LoadStoreVectorizer.cpp	[ValueTracking] Introduce a KnownBits struct to wrap the two APInts for computeKnownBits	2017-04-26 16:39:58 +00:00
LoopVectorize.cpp	[LV] Fix insertion point for shuffle vectors in first order recurrence	2017-05-09 14:29:33 +00:00
SLPVectorizer.cpp	[AArch64] Consider widening instructions in cost calculations	2017-05-09 20:18:12 +00:00
Vectorize.cpp	Add LoadStoreVectorizer pass	2016-06-30 23:11:38 +00:00