Sanjay Patel a46f06efe2 Use rsqrt (X86) to speed up reciprocal square root calcs ... This is a first step for generating SSE rsqrt instructions for reciprocal square root calcs when fast-math is allowed. For now, be conservative and only enable this for AMD btver2 where performance improves significantly - for example, 29% on llvm/projects/test-suite/SingleSource/Benchmarks/BenchmarkGame/n-body.c (if we convert the data type to single-precision float). This patch adds a two constant version of the Newton-Raphson refinement algorithm to DAGCombiner that can be selected by any target via a parameter returned by getRsqrtEstimate().. See PR20900 for more details: http://llvm.org/bugs/show_bug.cgi?id=20900 Differential Revision: http://reviews.llvm.org/D5658 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220570 91177308-0d34-0410-b5e6-96231b3b80d8		2014-10-24 17:02:16 +00:00
..
CMakeLists.txt
DAGCombiner.cpp	Use rsqrt (X86) to speed up reciprocal square root calcs	2014-10-24 17:02:16 +00:00
FastISel.cpp	Introduce enum values for previously defined metadata types. (NFC)	2014-10-21 00:13:20 +00:00
FunctionLoweringInfo.cpp	Remove uses of the TargetMachine from FunctionLoweringInfo	2014-10-09 00:57:31 +00:00
InstrEmitter.cpp	Remove the uses of getSubtargetImpl from InstrEmitter and remove	2014-10-09 01:35:29 +00:00
InstrEmitter.h	Remove the uses of getSubtargetImpl from InstrEmitter and remove	2014-10-09 01:35:29 +00:00
LegalizeDAG.cpp	Add minnum / maxnum codegen	2014-10-21 23:01:01 +00:00
LegalizeFloatTypes.cpp	Add minnum / maxnum codegen	2014-10-21 23:01:01 +00:00
LegalizeIntegerTypes.cpp	Optimize sext/zext insertion algorithm in back-end.	2014-09-19 05:30:35 +00:00
LegalizeTypes.cpp	Make it possible for ints/floats to return different values from getBooleanContents()	2014-07-10 10:18:12 +00:00
LegalizeTypes.h	Add minnum / maxnum codegen	2014-10-21 23:01:01 +00:00
LegalizeTypesGeneric.cpp	AA metadata refactoring (introduce AAMDNodes)	2014-07-24 12:16:19 +00:00
LegalizeVectorOps.cpp	Add minnum / maxnum codegen	2014-10-21 23:01:01 +00:00
LegalizeVectorTypes.cpp	[SelectionDAG] Teach the vector scalarizer about FP conversions.	2014-10-23 22:49:25 +00:00
LLVMBuild.txt
Makefile
ResourcePriorityQueue.cpp	Remove unused argument to CreateTargetScheduleState and change	2014-10-09 01:59:35 +00:00
ScheduleDAGFast.cpp	ScheduleDAG: record PhysReg dependencies represented by CopyFromReg nodes	2014-10-23 22:31:48 +00:00
ScheduleDAGRRList.cpp	ScheduleDAG: record PhysReg dependencies represented by CopyFromReg nodes	2014-10-23 22:31:48 +00:00
ScheduleDAGSDNodes.cpp	ScheduleDAG: record PhysReg dependencies represented by CopyFromReg nodes	2014-10-23 22:31:48 +00:00
ScheduleDAGSDNodes.h	Canonicalize header guards into a common format.	2014-08-13 16:26:38 +00:00
ScheduleDAGVLIW.cpp	Remove more calls to getSubtargetImpl from the schedulers and	2014-10-09 06:28:06 +00:00
SDNodeDbgValue.h	constify the getters in SDNodeDbgValue.	2014-10-13 20:43:47 +00:00
SelectionDAG.cpp	Modernize old-style static asserts. NFC.	2014-10-12 17:56:40 +00:00
SelectionDAGBuilder.cpp	Add minnum / maxnum codegen	2014-10-21 23:01:01 +00:00
SelectionDAGBuilder.h	Add minnum / maxnum codegen	2014-10-21 23:01:01 +00:00
SelectionDAGDumper.cpp	Add minnum / maxnum codegen	2014-10-21 23:01:01 +00:00
SelectionDAGISel.cpp	Remove uses of the TargetMachine from FunctionLoweringInfo	2014-10-09 00:57:31 +00:00
SelectionDAGPrinter.cpp	Revert "Introduce a string_ostream string builder facilty"	2014-06-26 22:52:05 +00:00
TargetLowering.cpp	Eliminate some deep std::vector copies. NFC.	2014-10-03 18:33:16 +00:00
TargetSelectionDAGInfo.cpp	Have TargetSelectionDAGInfo take a DataLayout initializer rather than	2014-06-06 19:04:48 +00:00