llvm-6502/lib/Analysis
Hal Finkel e14fb07357 Consider (x == -1) unlikely in BranchProbabilityInfo
This adds another heuristic to BPI, similar to the existing heuristic that
considers (x == 0) unlikely to be true. As suggested in the PACT'98 paper by
Deitrich, Cheng, and Hwu, -1 is often used to indicate an invalid index, and
equality comparisons with -1 are also unlikely to succeed. Local
experimentation supports this hypothesis: This yields a 1-2% speedup in the
test-suite sqlite benchmark on the PPC A2 core, with no significant
regressions.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@193855 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-01 10:58:22 +00:00
..
IPA Merge CallGraph and BasicCallGraph. 2013-10-31 03:03:55 +00:00
AliasAnalysis.cpp
AliasAnalysisCounter.cpp
AliasAnalysisEvaluator.cpp
AliasDebugger.cpp
AliasSetTracker.cpp
Analysis.cpp Remove the very substantial, largely unmaintained legacy PGO 2013-10-02 15:42:23 +00:00
BasicAliasAnalysis.cpp Revert r193251 : Use address-taken to disambiguate global variable and indirect memops. 2013-10-27 03:08:44 +00:00
BlockFrequencyInfo.cpp
BranchProbabilityInfo.cpp Consider (x == -1) unlikely in BranchProbabilityInfo 2013-11-01 10:58:22 +00:00
CaptureTracking.cpp CaptureTracking: Plug a loophole in the "too many uses" heuristic. 2013-10-03 13:24:02 +00:00
CFG.cpp
CFGPrinter.cpp
CMakeLists.txt Remove the very substantial, largely unmaintained legacy PGO 2013-10-02 15:42:23 +00:00
CodeMetrics.cpp
ConstantFolding.cpp
CostModel.cpp
DependenceAnalysis.cpp
DominanceFrontier.cpp
DomPrinter.cpp
InstCount.cpp
InstructionSimplify.cpp
Interval.cpp
IntervalPartition.cpp
IVUsers.cpp
LazyValueInfo.cpp
LibCallAliasAnalysis.cpp
LibCallSemantics.cpp
Lint.cpp
LLVMBuild.txt
Loads.cpp
LoopInfo.cpp Quick look-up for block in loop. 2013-10-26 03:08:02 +00:00
LoopPass.cpp
Makefile
MemDepPrinter.cpp
MemoryBuiltins.cpp fix PR17635: false positive with packed structures 2013-10-24 09:17:24 +00:00
MemoryDependenceAnalysis.cpp
ModuleDebugInfoPrinter.cpp
NoAliasAnalysis.cpp
PHITransAddr.cpp
PostDominators.cpp
PtrUseVisitor.cpp
README.txt
RegionInfo.cpp
RegionPass.cpp
RegionPrinter.cpp
ScalarEvolution.cpp SCEV: Make the final add of an inbounds GEP nuw if we know that the index is positive. 2013-10-28 07:30:06 +00:00
ScalarEvolutionAliasAnalysis.cpp
ScalarEvolutionExpander.cpp Fix SCEVExpander: don't try to expand quadratic recurrences outside a loop. 2013-10-25 21:35:56 +00:00
ScalarEvolutionNormalization.cpp Fix LSR: don't normalize quadratic recurrences. 2013-10-25 21:35:52 +00:00
SparsePropagation.cpp
TargetTransformInfo.cpp
Trace.cpp
TypeBasedAliasAnalysis.cpp TBAA: fix PR17620. 2013-10-22 01:40:25 +00:00
ValueTracking.cpp Remove x86_sse42_crc32_64_8 intrinsic. It has no functional difference from x86_sse42_crc32_32_8 and was not mapped to a clang builtin. I'm not even sure why this form of the instruction is even called out explicitly in the docs. Also add AutoUpgrade support to convert it into the other intrinsic with appropriate trunc and zext. 2013-10-15 05:20:47 +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))

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