Guozhi Wei 8bfc7351e0 [SLP] Pass in correct alignment when query memory access cost ... This patch fixes bug https://llvm.org/bugs/show_bug.cgi?id=27897. When query memory access cost, current SLP always passes in alignment value of 1 (unaligned), so it gets a very high cost of scalar memory access, and wrongly vectorize memory loads in the test case. It can be fixed by simply giving correct alignment. llvm-svn: 271333		2016-05-31 20:41:19 +00:00
..
ADCE
AddDiscriminators	Revert http://reviews.llvm.org/D19926 as it breaks tests.	2016-05-05 20:47:53 +00:00
AlignmentFromAssumptions
ArgumentPromotion
AtomicExpand
BBVectorize	[BBVectorize] Don't vectorize selects with a scalar condition and vector operands.	2016-05-26 18:43:57 +00:00
BDCE	[PM] Port BDCE to the new pass manager.	2016-05-25 01:57:04 +00:00
BranchFolding
CodeExtractor
CodeGenPrepare
ConstantHoisting
ConstantMerge	[PM] Port ConstantMerge to the new pass manager.	2016-05-05 00:51:09 +00:00
ConstProp	Revert "[SCCP] Partially propagate informations when the input is not fully defined."	2016-05-11 23:06:10 +00:00
CorrelatedValuePropagation	Remove extra whitespace. NFC.	2016-05-02 16:45:00 +00:00
CrossDSOCFI
DCE
DeadArgElim	[IR] Copy comdats in GlobalObject::copyAttributesFrom	2016-05-25 18:36:22 +00:00
DeadStoreElimination	[PM] Port DSE to the new pass manager	2016-05-17 21:38:13 +00:00
EarlyCSE
EliminateAvailableExternally	[PM] Port EliminateAvailableExternally pass to the new pass manager.	2016-05-05 02:37:32 +00:00
Float2Int
ForcedFunctionAttrs
FunctionAttrs	[CaptureTracking] Volatile operations capture their memory location	2016-05-26 17:36:22 +00:00
FunctionImport	ThinLTO: do not import function whose linkage prevents inlining.	2016-05-03 00:27:28 +00:00
GCOVProfiling
GlobalDCE	[GlobalDCE, Misc] Don't remove functions referenced by ifuncs	2016-05-04 00:20:48 +00:00
GlobalMerge	CodeGen: Make the global-merge pass independently testable, and add a test.	2016-05-19 04:38:56 +00:00
GlobalOpt	Make "@name =" mandatory for globals in .ll files.	2016-05-10 18:22:45 +00:00
GuardWidening	[GuardWidening] Fix incorrect use of remove_if	2016-05-21 02:24:44 +00:00
GVN	[GVN] Preserve !range metadata when PRE'ing loads	2016-05-27 19:03:10 +00:00
IndVarSimplify	[IndVars] Eliminate op.with.overflow when possible (re-apply)	2016-05-29 00:36:25 +00:00
InferFunctionAttrs	[TLI] Also cover Linux 64 libfunc (stat64, ...) prototype checking.	2016-05-25 21:16:33 +00:00
Inline	Allow -inline-threshold to override default threshold.	2016-05-19 23:02:09 +00:00
InstCombine	[X86][SSE] (Reapplied) Replace (V)PMOVSX and (V)PMOVZX integer extension intrinsics with generic IR (llvm)	2016-05-28 18:03:41 +00:00
InstMerge	[MergedLoadStoreMotion] Don't transform across may-throw calls	2016-05-26 07:11:09 +00:00
InstSimplify	[ValueTracking, InstSimplify] extend isKnownNonZero() to handle vector constants	2016-05-24 14:18:49 +00:00
Internalize
IPConstantProp	[PM] Port Interprocedural SCCP to the new pass manager.	2016-05-05 21:05:36 +00:00
IRCE	[IRCE] Optimize conjunctions of range checks	2016-05-26 00:09:02 +00:00
JumpThreading
LCSSA
LICM	MemorySSA: Revert r269678 and r268068; replace with special casing in MemorySSA.	2016-05-26 04:58:46 +00:00
LoadCombine
LoopDataPrefetch	[LoopDataPrefetch] Add optimization remark	2016-05-05 00:08:15 +00:00
LoopDeletion	Use all_of instead of a raw loop; NFC	2016-05-03 17:50:06 +00:00
LoopDistribute
LoopIdiom	AMDGPU: Other sizes of popcnt are fast	2016-05-18 16:10:19 +00:00
LoopInterchange
LoopLoadElim
LoopReroll	Enable loopreroll for sext of loop control only IV	2016-05-10 21:16:49 +00:00
LoopRotate	LPM: Drop require<loops> from these tests, it's redundant. NFC	2016-05-10 18:28:10 +00:00
LoopSimplify
LoopSimplifyCFG	LPM: Drop require<loops> from these tests, it's redundant. NFC	2016-05-10 18:28:10 +00:00
LoopStrengthReduce	AMDGPU: Fix a few slightly broken tests	2016-05-18 15:48:44 +00:00
LoopUnroll	The patch refactors unroll pass.	2016-05-27 23:15:06 +00:00
LoopUnswitch
LoopVectorize	Move test to X86 directory: I think it depends on X86 TTI.	2016-05-27 16:56:54 +00:00
LoopVersioning
LoopVersioningLICM
LowerAtomic	[PM] Port LowerAtomic to the new pass manager.	2016-05-13 22:52:35 +00:00
LowerBitSets
LowerExpectIntrinsic
LowerGuardIntrinsic	[Guards] Add branch metadata when lowering	2016-05-17 17:51:19 +00:00
LowerInvoke
LowerSwitch
Mem2Reg
MemCpyOpt	[MemCpyOpt] Don't perform callslot optimization across may-throw calls	2016-05-26 19:24:24 +00:00
MergeFunc	Fix a crash in MergeFunctions related to ordering of weak/strong functions	2016-05-31 17:20:23 +00:00
MetaRenamer
NameAnonFunctions
NaryReassociate
ObjCARC	Form objc_storeStrong in the presence of bitcasts.	2016-05-27 02:13:53 +00:00
PartiallyInlineLibCalls	[PM] Port PartiallyInlineLibCalls to the new pass manager.	2016-05-25 23:38:53 +00:00
PGOProfile	Bring back r271090 in a way that doesn't depend on r271089.	2016-05-28 04:05:36 +00:00
PhaseOrdering	Mark that SpeculativeExecution preserves Globals Alias Analysis.	2016-05-03 08:33:26 +00:00
PlaceSafepoints
PreISelIntrinsicLowering
PruneEH
Reassociate
Reg2Mem
RewriteStatepointsForGC	[RewriteStatepointsForGC] All constant should have null base pointer	2016-05-27 13:13:59 +00:00
SafeStack
SampleProfile	[PM] Port the Sample FDO to new PM (part-2)	2016-05-27 23:20:16 +00:00
Scalarizer
ScalarRepl
SCCP	[PM] Port per-function SCCP to the new pass manager.	2016-05-18 15:18:25 +00:00
SeparateConstOffsetFromGEP
SimplifyCFG	[SimplifyCFG] Remove cleanuppads which are empty except for calls to lifetime.end	2016-05-21 05:12:32 +00:00
Sink
SLPVectorizer	[SLP] Pass in correct alignment when query memory access cost	2016-05-31 20:41:19 +00:00
SpeculativeExecution
SROA	[SROA] Function canConvertValue needs to check whether both NewTy and OldTy pointers are	2016-05-03 19:30:48 +00:00
StraightLineStrengthReduce
StripDeadPrototypes
StripSymbols	Refactor stripDebugInfo(Function) to handle intrinsic	2016-05-07 04:10:52 +00:00
StructurizeCFG
TailCallElim
Util	MemorySSA: Revert r269678 and r268068; replace with special casing in MemorySSA.	2016-05-26 04:58:46 +00:00
WholeProgramDevirt