Dorit Nuzman c4c9199631 [IAI,LV] Add support for vectorizing predicated strided accesses using masked ... interleave-group The vectorizer currently does not attempt to create interleave-groups that contain predicated loads/stores; predicated strided accesses can currently be vectorized only using masked gather/scatter or scalarization. This patch makes predicated loads/stores candidates for forming interleave-groups during the Loop-Vectorizer's analysis, and adds the proper support for masked-interleave- groups to the Loop-Vectorizer's planning and transformation stages. The patch also extends the TTI API to allow querying the cost of masked interleave groups (which each target can control); Targets that support masked vector loads/ stores may choose to enable this feature and allow vectorizing predicated strided loads/stores using masked wide loads/stores and shuffles. Reviewers: Ayal, hsaito, dcaballe, fhahn, javed.absar Reviewed By: Ayal Differential Revision: https://reviews.llvm.org/D53011 llvm-svn: 344472		2018-10-14 07:06:16 +00:00
..
ADCE
AddDiscriminators
AggressiveInstCombine
AlignmentFromAssumptions
ArgumentPromotion
AtomicExpand	AMDGPU: Expand atomicrmw nand in IR	2018-10-02 03:50:56 +00:00
BDCE
BlockExtractor
BranchFolding
CalledValuePropagation
CallSiteSplitting	[CallSiteSplitting] Add debug location to created PHI nodes.	2018-09-11 17:55:58 +00:00
CodeExtractor	[CodeExtractor] Use 'normal destination' BB as insert point to store invoke results.	2018-08-21 20:07:46 +00:00
CodeGenPrepare	[CodeGen] Enable tail calls for functions with NonNull attributes.	2018-09-26 10:46:18 +00:00
ConstantHoisting	X86/TargetTransformInfo: Report div/rem constant immediate costs as TCC_Free	2018-10-11 23:14:35 +00:00
ConstantMerge
ConstProp
Coroutines
CorrelatedValuePropagation	[CVP] Extend tests to illustrate an old patch isn't needed	2018-08-24 21:56:43 +00:00
CrossDSOCFI
DCE
DeadArgElim	[deadargelim] Update dbg.value of 'unused' parameters	2018-09-24 10:01:24 +00:00
DeadStoreElimination
DivRemPairs
EarlyCSE	[EarlyCSEwMemorySSA] Add MSSA verification and tests to make EarlyCSE failures easier to track.	2018-09-17 22:35:21 +00:00
EliminateAvailableExternally
EntryExitInstrumenter
ExpandMemCmp/X86
Float2Int
ForcedFunctionAttrs
FunctionAttrs	[FuncAttrs] Remove "access range attributes" for read-none functions	2018-09-11 11:51:29 +00:00
FunctionImport	[ThinLTO] Keep non-prevailing (linkonce|weak)_odr symbols live	2018-10-08 15:12:48 +00:00
GCOVProfiling
GlobalDCE
GlobalMerge
GlobalOpt	[PPC] Remove Darwin support from POWER backend.	2018-08-28 01:18:29 +00:00
GlobalSplit
GuardWidening	[GuardWidening] Ignore guards with trivial conditions	2018-08-22 02:40:49 +00:00
GVN	Make YAML quote forward slashes.	2018-10-12 16:31:20 +00:00
GVNHoist	[MemorySSAUpdater] Avoid creating self-referencing MemoryDefs	2018-09-11 14:29:59 +00:00
GVNSink	[Local] Make DoesKMove required for combineMetadata.	2018-08-24 11:40:04 +00:00
HotColdSplit	Add support for new pass manager	2018-10-03 05:55:20 +00:00
IndirectBrExpand
IndVarSimplify	[IndVars] Drop "exact" flag from lshr and udiv when substituting their args	2018-10-11 07:22:26 +00:00
InferAddressSpaces	AMDGPU: Fix tests using old number for constant address space	2018-09-10 02:54:25 +00:00
InferFunctionAttrs	[LibCalls] Added returned attribute to libcalls	2018-08-23 05:18:23 +00:00
Inline	Make YAML quote forward slashes.	2018-10-12 16:31:20 +00:00
InstCombine	[InstCombine] fix complexity canonicalization with fake unary vector ops	2018-10-13 16:15:37 +00:00
InstMerge
InstNamer
InstSimplify	[FPEnv] PatternMatcher support for checking FNEG ignoring signed zeros	2018-10-09 21:48:00 +00:00
InterleavedAccess
Internalize
IPConstantProp	Recommit r333268: [IPSCCP] Use PredicateInfo to propagate facts from cmp instructions.	2018-08-23 11:04:00 +00:00
IRCE	Revert "[SCEV][NFC] Check NoWrap flags before lexicographical comparison of SCEVs"	2018-08-27 21:41:37 +00:00
JumpThreading	[Local] Make DoesKMove required for combineMetadata.	2018-08-24 11:40:04 +00:00
LCSSA
LICM	[Analysis] Make LocationSizes carry an 'imprecise' bit	2018-10-10 06:39:40 +00:00
LoadStoreVectorizer	AMDGPU: Fix private handling for allowsMisalignedMemoryAccesses	2018-09-24 13:18:15 +00:00
LoopDataPrefetch
LoopDeletion
LoopDistribute
LoopIdiom
LoopInstSimplify	Update MemorySSA in LoopInstSimplify.	2018-08-22 20:05:21 +00:00
LoopInterchange	[LoopInterchange] Preserve LCSSA.	2018-09-26 19:34:25 +00:00
LoopLoadElim
LoopPredication
LoopReroll
LoopRotate	[New PM] Introducing PassInstrumentation framework	2018-09-20 17:08:45 +00:00
LoopSimplify
LoopSimplifyCFG	Update MemorySSA in LoopSimplifyCFG.	2018-08-22 20:10:21 +00:00
LoopStrengthReduce	AMDGPU: Fix some outdated datalayouts in tests	2018-09-13 11:56:28 +00:00
LoopUnroll	Remove LoopID metadata from the branch instruction	2018-09-26 01:03:21 +00:00
LoopUnrollAndJam
LoopUnswitch	Update MemorySSA in LoopUnswitch.	2018-09-11 19:19:21 +00:00
LoopVectorize	[IAI,LV] Add support for vectorizing predicated strided accesses using masked	2018-10-14 07:06:16 +00:00
LoopVersioning
LoopVersioningLICM
LowerAtomic
LowerExpectIntrinsic
LowerGuardIntrinsic
LowerInvoke
LowerSwitch
LowerTypeTests
Mem2Reg
MemCpyOpt	Merge clang's isRepeatedBytePattern with LLVM's isBytewiseValue	2018-09-21 05:17:42 +00:00
MergeFunc
MergeICmps	Re-submitting changes in D51550 because it failed to patch.	2018-09-24 20:47:12 +00:00
MetaRenamer
NameAnonGlobals
NaryReassociate
NewGVN	[NewGVN] Mark function as changed if we erase instructions.	2018-09-07 11:41:34 +00:00
ObjCARC
PartiallyInlineLibCalls
PGOProfile	Add a flag to remap manglings when reading profile data information.	2018-10-10 23:13:47 +00:00
PhaseOrdering
PlaceSafepoints
PreISelIntrinsicLowering
PruneEH
Reassociate	revert r341288 - [Reassociate] swap binop operands to increase factoring potential	2018-09-12 21:29:11 +00:00
Reg2Mem
RewriteStatepointsForGC
SafeStack	SafeStack: Prevent OOB reads with mem intrinsics	2018-08-30 20:44:51 +00:00
SampleProfile	Add a flag to remap manglings when reading profile data information.	2018-10-10 23:13:47 +00:00
ScalarizeMaskedMemIntrin/X86	[ScalarizeMaskedMemIntrin] Use MinAlign to calculate alignment for the scalar load/stores to handle element types that are byte-sized but not powers of 2.	2018-09-28 03:35:37 +00:00
Scalarizer	Fix an ordering bug in the scalarizer.	2018-10-10 09:27:45 +00:00
SCCP	SCCP: avoid caching DenseMap entry that might be invalidated.	2018-10-12 09:01:59 +00:00
SeparateConstOffsetFromGEP
SimpleLoopUnswitch	[SimpleLoopUnswitch] remove a chain of dead blocks at once	2018-09-04 20:19:41 +00:00
SimplifyCFG	[SimplifyCFG] Put an alignment on generated switch tables	2018-09-12 09:54:17 +00:00
Sink
SLPVectorizer	[SLPVectorizer] Check that lowered type is floating point before calling isFabsFree	2018-10-09 18:41:17 +00:00
SpeculateAroundPHIs
SpeculativeExecution
SROA	[SROA] Fix alignment for uses of PHI nodes.	2018-08-30 18:59:24 +00:00
StraightLineStrengthReduce
StripDeadPrototypes
StripSymbols
StructurizeCFG	AMDGPU: Fix tests using old number for constant address space	2018-09-10 02:54:25 +00:00
SyntheticCountsPropagation
TailCallElim	[TailCallElim] Enable marking of calls with byval as tails	2018-10-08 18:03:40 +00:00
ThinLTOBitcodeWriter
Util	Relax trivial cast requirements in CallPromotionUtils	2018-10-10 16:35:47 +00:00
WholeProgramDevirt	[ThinLTO] Efficiency fix for writing type id records in per-module indexes	2018-09-25 20:14:40 +00:00