llvm-mirror

mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-26 12:43:36 +01:00

History

Ta-Wei Tu df760e157d [LoopInterchange] Replace tightly-nesting-ness check with the one from `LoopNest` The check `tightlyNested()` in `LoopInterchange` is similar to the one in `LoopNest`. In fact, the former misses some cases where loop-interchange is not feasible and results in incorrect behaviour. Replacing it with the much robust version provided by `LoopNest` reduces code duplications and fixes https://bugs.llvm.org/show_bug.cgi?id=48113. `LoopInterchange` has a weaker definition of tightly or perfectly nesting-ness than the one implemented in `LoopNest::arePerfectlyNested()`. Therefore, `tightlyNested()` is instead implemented with `LoopNest::checkLoopsStructure` and additional checks for unsafe instructions. Reviewed By: Whitney Differential Revision: https://reviews.llvm.org/D97290		2021-03-08 11:36:08 +08:00
..
models/inliner
AliasAnalysis.cpp	[AA] Cache (optionally) estimated PartialAlias offsets.	2021-03-02 19:04:15 +03:00
AliasAnalysisEvaluator.cpp
AliasAnalysisSummary.cpp
AliasAnalysisSummary.h
AliasSetTracker.cpp	[NFC][AliasSetTracker] Remove implicit conversion AliasResult to integer.	2021-03-05 00:53:27 +03:00
Analysis.cpp
AssumeBundleQueries.cpp
AssumptionCache.cpp
BasicAliasAnalysis.cpp	[basicaa] Recurse through a single phi input	2021-03-04 13:07:06 -08:00
BlockFrequencyInfo.cpp
BlockFrequencyInfoImpl.cpp
BranchProbabilityInfo.cpp
CallGraph.cpp
CallGraphSCCPass.cpp
CallPrinter.cpp
CaptureTracking.cpp
CFG.cpp
CFGPrinter.cpp
CFLAndersAliasAnalysis.cpp
CFLGraph.h
CFLSteensAliasAnalysis.cpp
CGSCCPassManager.cpp
CMakeLists.txt
CmpInstAnalysis.cpp
CodeMetrics.cpp
ConstantFolding.cpp	[ConstantFold] Handle vectors in ConstantFoldLoadThroughBitcast()	2021-03-06 12:17:56 +01:00
ConstraintSystem.cpp
CostModel.cpp
DDG.cpp
DDGPrinter.cpp
Delinearization.cpp
DemandedBits.cpp
DependenceAnalysis.cpp
DependenceGraphBuilder.cpp
DevelopmentModeInlineAdvisor.cpp
DivergenceAnalysis.cpp
DominanceFrontier.cpp
DomPrinter.cpp
DomTreeUpdater.cpp
EHPersonalities.cpp
FunctionPropertiesAnalysis.cpp
GlobalsModRef.cpp
GuardUtils.cpp
HeatUtils.cpp
ImportedFunctionsInliningStatistics.cpp
IndirectCallPromotionAnalysis.cpp	[SampleFDO] Another fix to prevent repeated indirect call promotion in	2021-03-04 18:44:12 -08:00
InlineAdvisor.cpp
InlineCost.cpp
InlineSizeEstimatorAnalysis.cpp
InstCount.cpp
InstructionPrecedenceTracking.cpp
InstructionSimplify.cpp	[InstCombine] Add simplification of two logical and/ors	2021-03-08 02:38:43 +09:00
Interval.cpp
IntervalPartition.cpp
IRSimilarityIdentifier.cpp
IVDescriptors.cpp	[Analysis][LoopVectorize] rename "Unsafe" variables/methods; NFC	2021-03-04 08:53:04 -05:00
IVUsers.cpp
LazyBlockFrequencyInfo.cpp
LazyBranchProbabilityInfo.cpp
LazyCallGraph.cpp
LazyValueInfo.cpp	[LVI] Simplify and generalize handling of clamp patterns	2021-03-06 10:42:41 +01:00
LegacyDivergenceAnalysis.cpp
Lint.cpp
Loads.cpp	[Loads] Restructure getAvailableLoadStore implementation (NFC)	2021-03-06 16:58:11 +01:00
LoopAccessAnalysis.cpp
LoopAnalysisManager.cpp
LoopCacheAnalysis.cpp
LoopInfo.cpp
LoopNestAnalysis.cpp	[LoopInterchange] Replace tightly-nesting-ness check with the one from `LoopNest`	2021-03-08 11:36:08 +08:00
LoopPass.cpp
LoopUnrollAnalyzer.cpp
MemDepPrinter.cpp
MemDerefPrinter.cpp
MemoryBuiltins.cpp
MemoryDependenceAnalysis.cpp
MemoryLocation.cpp
MemorySSA.cpp
MemorySSAUpdater.cpp
MLInlineAdvisor.cpp
ModuleDebugInfoPrinter.cpp
ModuleSummaryAnalysis.cpp	[ModuleSummaryAnalysis] Avoid duplicate elements in Worklist. NFC	2021-03-06 14:19:22 -08:00
MustExecute.cpp
ObjCARCAliasAnalysis.cpp
ObjCARCAnalysisUtils.cpp
ObjCARCInstKind.cpp	[ObjC][ARC] Use operand bundle 'clang.arc.attachedcall' instead of	2021-03-04 11:22:30 -08:00
OptimizationRemarkEmitter.cpp
PHITransAddr.cpp
PhiValues.cpp
PostDominators.cpp
ProfileSummaryInfo.cpp
PtrUseVisitor.cpp
README.txt
RegionInfo.cpp
RegionPass.cpp
RegionPrinter.cpp
ReleaseModeModelRunner.cpp
ReplayInlineAdvisor.cpp
ScalarEvolution.cpp
ScalarEvolutionAliasAnalysis.cpp
ScalarEvolutionDivision.cpp
ScalarEvolutionNormalization.cpp
ScopedNoAliasAA.cpp	[llvm] Use set_is_subset (NFC)	2021-02-28 10:59:20 -08:00
StackLifetime.cpp
StackSafetyAnalysis.cpp
StratifiedSets.h
SyncDependenceAnalysis.cpp
SyntheticCountsUtils.cpp
TargetLibraryInfo.cpp
TargetTransformInfo.cpp
TFUtils.cpp
Trace.cpp
TypeBasedAliasAnalysis.cpp
TypeMetadataUtils.cpp
ValueLattice.cpp
ValueLatticeUtils.cpp
ValueTracking.cpp	[ValueTracking] update directlyImpliesPoison to look into select's condition	2021-03-07 23:16:44 +09:00
VectorUtils.cpp
VFABIDemangling.cpp

README.txt

Analysis Opportunities:

//===---------------------------------------------------------------------===//

In test/Transforms/LoopStrengthReduce/quadradic-exit-value.ll, the
ScalarEvolution expression for %r is this:

  {1,+,3,+,2}<loop>

Outside the loop, this could be evaluated simply as (%n * %n), however
ScalarEvolution currently evaluates it as

  (-2 + (2 * (trunc i65 (((zext i64 (-2 + %n) to i65) * (zext i64 (-1 + %n) to i65)) /u 2) to i64)) + (3 * %n))

In addition to being much more complicated, it involves i65 arithmetic,
which is very inefficient when expanded into code.

//===---------------------------------------------------------------------===//

In formatValue in test/CodeGen/X86/lsr-delayed-fold.ll,

ScalarEvolution is forming this expression:

((trunc i64 (-1 * %arg5) to i32) + (trunc i64 %arg5 to i32) + (-1 * (trunc i64 undef to i32)))

This could be folded to

(-1 * (trunc i64 undef to i32))

//===---------------------------------------------------------------------===//