llvm-6502/lib/Analysis
Andrew Trick b12a754cce Added isValidRewrite() to check the result of ScalarEvolutionExpander.
SCEV may generate expressions composed of multiple pointers, which can
lead to invalid GEP expansion. Until we can teach SCEV to follow strict
pointer rules, make sure no bad GEPs creep into IR.
Fixes rdar://problem/9038671.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127839 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-17 23:51:11 +00:00
..
IPA
AliasAnalysis.cpp
AliasAnalysisCounter.cpp
AliasAnalysisEvaluator.cpp
AliasDebugger.cpp
AliasSetTracker.cpp
Analysis.cpp Delete the LiveValues pass. I won't get get back to the project it 2011-02-28 19:37:59 +00:00
BasicAliasAnalysis.cpp
CaptureTracking.cpp
CFGPrinter.cpp
CMakeLists.txt Unbreak CMake build. 2011-03-01 00:02:51 +00:00
ConstantFolding.cpp convert ConstantVector::get to use ArrayRef. 2011-02-15 00:14:00 +00:00
DbgInfoPrinter.cpp Move DbgInfoPrinter specific utlities inside DbgInfoPrinter.cpp 2011-02-15 17:36:11 +00:00
DebugInfo.cpp Remove DIFactory. Patch by Devang. 2011-03-02 20:30:37 +00:00
DIBuilder.cpp Today, the language front ends produces llvm.dbg.* intrinsics, used to encode arguments' debug info, in order any way, most of the times. However, if a front end mix-n-matches llvm.dbg.declare and llvm.dbg.value intrinsics to encode debug info for arguments then code generator needs a way to find argument order. 2011-03-01 22:58:13 +00:00
DominanceFrontier.cpp
DomPrinter.cpp
InlineCost.cpp
InstCount.cpp
InstructionSimplify.cpp Fix two cases I forgot to update when doing a mental "getSwappedPredicate". 2011-03-09 08:20:06 +00:00
Interval.cpp
IntervalPartition.cpp
IVUsers.cpp
LazyValueInfo.cpp
LibCallAliasAnalysis.cpp
LibCallSemantics.cpp
Lint.cpp
Loads.cpp
LoopDependenceAnalysis.cpp
LoopInfo.cpp
LoopPass.cpp Introduce DebugInfoProbe. This is used to monitor how llvm optimizer is treating debugging information. 2011-03-10 00:21:25 +00:00
Makefile
MemDepPrinter.cpp
MemoryBuiltins.cpp Add comments for the demanglings. Correct mangled form of operator delete! 2011-03-17 05:20:12 +00:00
MemoryDependenceAnalysis.cpp
ModuleDebugInfoPrinter.cpp
NoAliasAnalysis.cpp
PathNumbering.cpp
PathProfileInfo.cpp
PathProfileVerifier.cpp
PHITransAddr.cpp
PostDominators.cpp
ProfileEstimatorPass.cpp
ProfileInfo.cpp
ProfileInfoLoader.cpp
ProfileInfoLoaderPass.cpp
ProfileVerifierPass.cpp
README.txt
RegionInfo.cpp
RegionPass.cpp
RegionPrinter.cpp RegionPrinter: Ignore back edges when layouting the graph 2011-02-27 04:11:07 +00:00
ScalarEvolution.cpp Added isValidRewrite() to check the result of ScalarEvolutionExpander. 2011-03-17 23:51:11 +00:00
ScalarEvolutionAliasAnalysis.cpp
ScalarEvolutionExpander.cpp Added SCEV::NoWrapFlags to manage unsigned, signed, and self wrap 2011-03-14 16:50:06 +00:00
ScalarEvolutionNormalization.cpp Added SCEV::NoWrapFlags to manage unsigned, signed, and self wrap 2011-03-14 16:50:06 +00:00
SparsePropagation.cpp
Trace.cpp
TypeBasedAliasAnalysis.cpp
ValueTracking.cpp Teach ComputeMaskedBits about sub nsw. 2011-03-12 17:18:11 +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))

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