Tim Northover f8d927f22b CodeGen: emit IR-level f16 conversion intrinsics as fptrunc/fpext ... This makes the first stage DAG for @llvm.convert.to.fp16 an fptrunc, and correspondingly @llvm.convert.from.fp16 an fpext. The legalisation path is now uniform, regardless of the input IR: fptrunc -> FP_TO_FP16 (if f16 illegal) -> libcall fpext -> FP16_TO_FP (if f16 illegal) -> libcall Each target should be able to select the version that best matches its operations and not be required to duplicate patterns for both fptrunc and FP_TO_FP16 (for example). As a result we can remove some redundant AArch64 patterns. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@213507 91177308-0d34-0410-b5e6-96231b3b80d8		2014-07-21 09:13:56 +00:00
..
CMakeLists.txt
DAGCombiner.cpp	[SDAG,cleanup] Switch the DAG combiner over to use the spelling	2014-07-21 08:56:44 +00:00
FastISel.cpp	[FastISel] Local values shouldn't be alive across an inline asm call with side effects.	2014-07-16 22:20:51 +00:00
FunctionLoweringInfo.cpp	[Modules] Remove potential ODR violations by sinking the DEBUG_TYPE	2014-04-22 02:02:50 +00:00
InstrEmitter.cpp	This reapplies r207235 with an additional bugfixes caught by the msan	2014-04-25 20:49:25 +00:00
InstrEmitter.h
LegalizeDAG.cpp	CodeGen: generate single libcall for fptrunc -> f16 operations.	2014-07-17 11:12:12 +00:00
LegalizeFloatTypes.cpp	CodeGen: emit IR-level f16 conversion intrinsics as fptrunc/fpext	2014-07-21 09:13:56 +00:00
LegalizeIntegerTypes.cpp	CodeGen: extend f16 conversions to permit types > float.	2014-07-17 10:51:23 +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	ARM: support legalisation of "fptrunc ... to half" operations.	2014-07-18 13:01:19 +00:00
LegalizeTypesGeneric.cpp	Fix ppcf128 component access on little-endian systems	2014-07-03 15:06:47 +00:00
LegalizeVectorOps.cpp	[x86,SDAG] Introduce any- and sign-extend-vector-inreg nodes analogous	2014-07-10 12:32:32 +00:00
LegalizeVectorTypes.cpp	[x86,SDAG] Introduce any- and sign-extend-vector-inreg nodes analogous	2014-07-10 12:32:32 +00:00
LLVMBuild.txt
Makefile
ResourcePriorityQueue.cpp	Fix 'platform-specific' hyphenations	2014-06-30 18:57:16 +00:00
ScheduleDAGFast.cpp	[Modules] Remove potential ODR violations by sinking the DEBUG_TYPE	2014-04-22 02:02:50 +00:00
ScheduleDAGRRList.cpp	fixed typo	2014-07-14 18:21:07 +00:00
ScheduleDAGSDNodes.cpp	Convert SelectionDAG::MorphNodeTo to use ArrayRef.	2014-04-27 19:21:16 +00:00
ScheduleDAGSDNodes.h
ScheduleDAGVLIW.cpp	The hazard recognizer only needs a subtarget, not a target machine	2014-06-13 22:38:52 +00:00
SDNodeDbgValue.h	This reapplies r207235 with an additional bugfixes caught by the msan	2014-04-25 20:49:25 +00:00
SelectionDAG.cpp	[DAG] Refactor some logic. No functional change.	2014-07-21 07:28:51 +00:00
SelectionDAGBuilder.cpp	CodeGen: emit IR-level f16 conversion intrinsics as fptrunc/fpext	2014-07-21 09:13:56 +00:00
SelectionDAGBuilder.h	Fix 'platform-specific' hyphenations	2014-06-30 18:57:16 +00:00
SelectionDAGDumper.cpp	CodeGen: extend f16 conversions to permit types > float.	2014-07-17 10:51:23 +00:00
SelectionDAGISel.cpp	[C++11] Add predecessors(BasicBlock *) / successors(BasicBlock *) iterator ranges.	2014-07-20 09:10:11 +00:00
SelectionDAGPrinter.cpp	Revert "Introduce a string_ostream string builder facilty"	2014-06-26 22:52:05 +00:00
TargetLowering.cpp	SelectionDAG: Factor FP_TO_SINT lower code out of DAGLegalizer	2014-07-10 22:40:18 +00:00
TargetSelectionDAGInfo.cpp	Have TargetSelectionDAGInfo take a DataLayout initializer rather than	2014-06-06 19:04:48 +00:00