mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-11-24 03:33:20 +01:00
bf4b959200
Some SIMD implementations are not IEEE-754 compliant, for example ARM's NEON. This patch teaches the loop vectorizer to only allow transformations of loops that either contain no floating-point operations or have enough allowance flags supporting lack of precision (ex. -ffast-math, Darwin). For that, the target description now has a method which tells us if the vectorizer is allowed to handle FP math without falling into unsafe representations, plus a check on every FP instruction in the candidate loop to check for the safety flags. This commit makes LLVM behave like GCC with respect to ARM NEON support, but it stops short of fixing the underlying problem: sub-normals. Neither GCC nor LLVM have a flag for allowing sub-normal operations. Before this patch, GCC only allows it using unsafe-math flags and LLVM allows it by default with no way to turn it off (short of not using NEON at all). As a first step, we push this change to make it safe and in sync with GCC. The second step is to discuss a new sub-normal's flag on both communitues and come up with a common solution. The third step is to improve the FastMath flags in LLVM to encode sub-normals and use those flags to restrict NEON FP. Fixes PR16275. llvm-svn: 266363 |
||
|---|---|---|
| .. | ||
| AliasAnalysis.cpp | ||
| AliasAnalysisEvaluator.cpp | ||
| AliasSetTracker.cpp | ||
| Analysis.cpp | ||
| AssumptionCache.cpp | ||
| BasicAliasAnalysis.cpp | ||
| BlockFrequencyInfo.cpp | ||
| BlockFrequencyInfoImpl.cpp | ||
| BranchProbabilityInfo.cpp | ||
| CallGraph.cpp | ||
| CallGraphSCCPass.cpp | ||
| CallPrinter.cpp | ||
| CaptureTracking.cpp | ||
| CFG.cpp | ||
| CFGPrinter.cpp | ||
| CFLAliasAnalysis.cpp | ||
| CGSCCPassManager.cpp | ||
| CMakeLists.txt | ||
| CodeMetrics.cpp | ||
| ConstantFolding.cpp | ||
| CostModel.cpp | ||
| Delinearization.cpp | ||
| DemandedBits.cpp | ||
| DependenceAnalysis.cpp | ||
| DivergenceAnalysis.cpp | ||
| DominanceFrontier.cpp | ||
| DomPrinter.cpp | ||
| EHPersonalities.cpp | ||
| GlobalsModRef.cpp | ||
| InlineCost.cpp | ||
| InstCount.cpp | ||
| InstructionSimplify.cpp | ||
| Interval.cpp | ||
| IntervalPartition.cpp | ||
| IteratedDominanceFrontier.cpp | ||
| IVUsers.cpp | ||
| LazyCallGraph.cpp | ||
| LazyValueInfo.cpp | ||
| Lint.cpp | ||
| LLVMBuild.txt | ||
| Loads.cpp | ||
| LoopAccessAnalysis.cpp | ||
| LoopInfo.cpp | ||
| LoopPass.cpp | ||
| LoopPassManager.cpp | ||
| LoopUnrollAnalyzer.cpp | ||
| MemDepPrinter.cpp | ||
| MemDerefPrinter.cpp | ||
| MemoryBuiltins.cpp | ||
| MemoryDependenceAnalysis.cpp | ||
| MemoryLocation.cpp | ||
| ModuleDebugInfoPrinter.cpp | ||
| ModuleSummaryAnalysis.cpp | ||
| ObjCARCAliasAnalysis.cpp | ||
| ObjCARCAnalysisUtils.cpp | ||
| ObjCARCInstKind.cpp | ||
| OrderedBasicBlock.cpp | ||
| PHITransAddr.cpp | ||
| PostDominators.cpp | ||
| PtrUseVisitor.cpp | ||
| README.txt | ||
| RegionInfo.cpp | ||
| RegionPass.cpp | ||
| RegionPrinter.cpp | ||
| ScalarEvolution.cpp | ||
| ScalarEvolutionAliasAnalysis.cpp | ||
| ScalarEvolutionExpander.cpp | ||
| ScalarEvolutionNormalization.cpp | ||
| ScopedNoAliasAA.cpp | ||
| SparsePropagation.cpp | ||
| StratifiedSets.h | ||
| TargetLibraryInfo.cpp | ||
| TargetTransformInfo.cpp | ||
| Trace.cpp | ||
| TypeBasedAliasAnalysis.cpp | ||
| ValueTracking.cpp | ||
| VectorUtils.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))
//===---------------------------------------------------------------------===//