llvm-6502/lib/Analysis
Hal Finkel f208398528 Revert: r189565 - Add getUnrollingPreferences to TTI
Revert unintentional commit (of an unreviewed change).

Original commit message:

Add getUnrollingPreferences to TTI

Allow targets to customize the default behavior of the generic loop unrolling
transformation. This will be used by the PowerPC backend when targeting the A2
core (which is in-order with a deep pipeline), and using more aggressive
defaults is important.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@189566 91177308-0d34-0410-b5e6-96231b3b80d8
2013-08-29 03:33:15 +00:00
..
IPA Disable inlining between sanitized and non-sanitized functions. 2013-08-08 08:22:39 +00:00
AliasAnalysis.cpp Reimplement isPotentiallyReachable to make nocapture deduction much stronger. 2013-07-27 01:24:00 +00:00
AliasAnalysisCounter.cpp
AliasAnalysisEvaluator.cpp
AliasDebugger.cpp
AliasSetTracker.cpp
Analysis.cpp
BasicAliasAnalysis.cpp Remove trailing spaces. 2013-08-24 14:16:00 +00:00
BlockFrequencyInfo.cpp
BranchProbabilityInfo.cpp
CaptureTracking.cpp
CFG.cpp Add some constantness. 2013-08-20 23:04:15 +00:00
CFGPrinter.cpp
CMakeLists.txt Also update CMakeLists.txt for r187283. 2013-07-27 01:25:51 +00:00
CodeMetrics.cpp
ConstantFolding.cpp Teach ConstantFolding about pointer address spaces 2013-08-20 21:20:04 +00:00
CostModel.cpp
DependenceAnalysis.cpp Remove extraneous semicolon. 2013-08-06 16:40:40 +00:00
DominanceFrontier.cpp
DomPrinter.cpp
InstCount.cpp
InstructionSimplify.cpp Add ISD::FROUND for libm round() 2013-08-07 22:49:12 +00:00
Interval.cpp
IntervalPartition.cpp
IVUsers.cpp
LazyValueInfo.cpp
LibCallAliasAnalysis.cpp
LibCallSemantics.cpp
Lint.cpp Fix lint assert on integer vector division 2013-08-26 23:29:33 +00:00
LLVMBuild.txt
Loads.cpp
LoopInfo.cpp
LoopPass.cpp Comment: try to clarify loop iteration order. 2013-07-20 23:10:31 +00:00
Makefile
MemDepPrinter.cpp
MemoryBuiltins.cpp Treat nothrow forms of ::operator delete and ::operator delete[] as 2013-07-21 23:11:42 +00:00
MemoryDependenceAnalysis.cpp
ModuleDebugInfoPrinter.cpp
NoAliasAnalysis.cpp
PathNumbering.cpp
PathProfileInfo.cpp
PathProfileVerifier.cpp
PHITransAddr.cpp
PostDominators.cpp
ProfileDataLoader.cpp
ProfileDataLoaderPass.cpp
ProfileEstimatorPass.cpp
ProfileInfo.cpp
ProfileInfoLoader.cpp
ProfileInfoLoaderPass.cpp
ProfileVerifierPass.cpp
PtrUseVisitor.cpp
README.txt
RegionInfo.cpp Reorder headers according to lint. 2013-08-21 21:14:19 +00:00
RegionPass.cpp
RegionPrinter.cpp
ScalarEvolution.cpp Fix a severe compile time problem when forming large SCEV expressions. 2013-07-31 02:43:40 +00:00
ScalarEvolutionAliasAnalysis.cpp
ScalarEvolutionExpander.cpp Fix SCEVExpander creating distinct duplicate PHI entries 2013-08-18 00:16:23 +00:00
ScalarEvolutionNormalization.cpp
SparsePropagation.cpp
TargetTransformInfo.cpp Revert: r189565 - Add getUnrollingPreferences to TTI 2013-08-29 03:33:15 +00:00
Trace.cpp
TypeBasedAliasAnalysis.cpp
ValueTracking.cpp Fix assert with GEP ptr vector indexing structs 2013-08-19 21:43:16 +00:00

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))

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