Tim Northover 44697f3fc1 X86 CodeGenPrep: sink shufflevectors before shifts ... On x86, shifting a vector by a scalar is significantly cheaper than shifting a vector by another fully general vector. Unfortunately, because SelectionDAG operates on just one basic block at a time, the shufflevector instruction that reveals whether the right-hand side of a shift *is* really a scalar is often not visible to CodeGen when it's needed. This adds another handler to CodeGenPrepare, to sink any useful shufflevector instructions down to the basic block where they're used, predicated on a target hook (since on other architectures, doing so will often just introduce extra real work). rdar://problem/16063505 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201655 91177308-0d34-0410-b5e6-96231b3b80d8		2014-02-19 10:02:43 +00:00
..
Hello	Set LLVM_EXPORTED_SYMBOL_FILE in CMakeLists whose corresponding Makefiles do so.	2013-12-29 23:06:49 +00:00
InstCombine	Do more addrspacecast transforms that happen for bitcast.	2014-02-14 00:49:12 +00:00
Instrumentation	Now that llvm always does the right thing with private, use it.	2014-02-19 02:08:39 +00:00
IPO	GlobalOpt: Aliases don't have sections, don't copy them when replacing	2014-02-13 02:18:36 +00:00
ObjCARC	Update optimization passes to handle inalloca arguments	2014-01-28 02:38:36 +00:00
Scalar	X86 CodeGenPrep: sink shufflevectors before shifts	2014-02-19 10:02:43 +00:00
Utils	Rename some member variables from TD to DL.	2014-02-18 15:33:12 +00:00
Vectorize	fix for null VectorizedValue assertion in the SLP Vectorizer (in function vectorizeTree()). radar://16064178	2014-02-17 03:06:16 +00:00
CMakeLists.txt
LLVMBuild.txt
Makefile