mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-11-24 11:42:57 +01:00
77d21dcd3f
Arm MVE has multiple instructions such as VMLAVA.s8, which (in this case) can take two 128bit vectors, sign extend the inputs to i32, multiplying them together and sum the result into a 32bit general purpose register. So taking 16 i8's as inputs, they can multiply and accumulate the result into a single i32 without any rounding/truncating along the way. There are also reduction instructions for plain integer add and min/max, and operations that sum into a pair of 32bit registers together treated as a 64bit integer (even though MVE does not have a plain 64bit addition instruction). So giving the vectorizer the ability to use these instructions both enables us to vectorize at higher bitwidths, and to vectorize things we previously could not. In order to do that we need a way to represent that the reduction operation, specified with a llvm.experimental.vector.reduce when vectorizing for Arm, occurs inside the loop not after it like most reductions. This patch attempts to do that, teaching the vectorizer about in-loop reductions. It does this through a vplan recipe representing the reductions that the original chain of reduction operations is replaced by. Cost modelling is currently just done through a prefersInloopReduction TTI hook (which follows in a later patch). Differential Revision: https://reviews.llvm.org/D75069 |
||
|---|---|---|
| .. | ||
| models/inliner | ||
| AliasAnalysis.cpp | ||
| AliasAnalysisEvaluator.cpp | ||
| AliasAnalysisSummary.cpp | ||
| AliasAnalysisSummary.h | ||
| AliasSetTracker.cpp | ||
| Analysis.cpp | ||
| AssumeBundleQueries.cpp | ||
| AssumptionCache.cpp | ||
| BasicAliasAnalysis.cpp | ||
| 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 | ||
| CostModel.cpp | ||
| DDG.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 | ||
| IndirectCallPromotionAnalysis.cpp | ||
| InlineAdvisor.cpp | ||
| InlineCost.cpp | ||
| InlineSizeEstimatorAnalysis.cpp | ||
| InstCount.cpp | ||
| InstructionPrecedenceTracking.cpp | ||
| InstructionSimplify.cpp | ||
| Interval.cpp | ||
| IntervalPartition.cpp | ||
| IVDescriptors.cpp | ||
| IVUsers.cpp | ||
| LazyBlockFrequencyInfo.cpp | ||
| LazyBranchProbabilityInfo.cpp | ||
| LazyCallGraph.cpp | ||
| LazyValueInfo.cpp | ||
| LegacyDivergenceAnalysis.cpp | ||
| Lint.cpp | ||
| LLVMBuild.txt | ||
| Loads.cpp | ||
| LoopAccessAnalysis.cpp | ||
| LoopAnalysisManager.cpp | ||
| LoopCacheAnalysis.cpp | ||
| LoopInfo.cpp | ||
| LoopNestAnalysis.cpp | ||
| LoopPass.cpp | ||
| LoopUnrollAnalyzer.cpp | ||
| MemDepPrinter.cpp | ||
| MemDerefPrinter.cpp | ||
| MemoryBuiltins.cpp | ||
| MemoryDependenceAnalysis.cpp | ||
| MemoryLocation.cpp | ||
| MemorySSA.cpp | ||
| MemorySSAUpdater.cpp | ||
| MLInlineAdvisor.cpp | ||
| ModuleDebugInfoPrinter.cpp | ||
| ModuleSummaryAnalysis.cpp | ||
| MustExecute.cpp | ||
| ObjCARCAliasAnalysis.cpp | ||
| ObjCARCAnalysisUtils.cpp | ||
| ObjCARCInstKind.cpp | ||
| OptimizationRemarkEmitter.cpp | ||
| PHITransAddr.cpp | ||
| PhiValues.cpp | ||
| PostDominators.cpp | ||
| ProfileSummaryInfo.cpp | ||
| PtrUseVisitor.cpp | ||
| README.txt | ||
| RegionInfo.cpp | ||
| RegionPass.cpp | ||
| RegionPrinter.cpp | ||
| ReleaseModeModelRunner.cpp | ||
| ScalarEvolution.cpp | ||
| ScalarEvolutionAliasAnalysis.cpp | ||
| ScalarEvolutionDivision.cpp | ||
| ScalarEvolutionNormalization.cpp | ||
| ScopedNoAliasAA.cpp | ||
| 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 | ||
| VectorUtils.cpp | ||
| VFABIDemangling.cpp | ||
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))
//===---------------------------------------------------------------------===//