llvm-6502

mirror of https://github.com/c64scene-ar/llvm-6502.git synced 2024-11-02 07:11:49 +00:00

History

Suyog Sarda adf8ae4c10 Rectify r213231. Use proper version of 'ComputeNumSignBits'. Earlier when the code was in InstCombine, we were calling the version of ComputeNumSignBits in InstCombine.h that automatically added the DataLayout* before calling into ValueTracking. When the code moved to InstSimplify, we are calling into ValueTracking directly without passing in the DataLayout*. This patch rectifies the same by passing DataLayout in ComputeNumSignBits. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@213295 91177308-0d34-0410-b5e6-96231b3b80d8		2014-07-17 19:07:00 +00:00
..
IPA	Add new lines to debugging information.	2014-07-03 02:11:49 +00:00
AliasAnalysis.cpp	Improve BasicAA CS-CS queries (redux)	2014-07-17 01:28:25 +00:00
AliasAnalysisCounter.cpp
AliasAnalysisEvaluator.cpp
AliasDebugger.cpp
AliasSetTracker.cpp
Analysis.cpp	Revert "Introduce a string_ostream string builder facilty"	2014-06-26 22:52:05 +00:00
BasicAliasAnalysis.cpp	Improve BasicAA CS-CS queries (redux)	2014-07-17 01:28:25 +00:00
BlockFrequencyInfo.cpp	Revert "Introduce a string_ostream string builder facilty"	2014-06-26 22:52:05 +00:00
BlockFrequencyInfoImpl.cpp	BFI: Add constructor for Weight	2014-07-12 00:26:00 +00:00
BranchProbabilityInfo.cpp
CaptureTracking.cpp
CFG.cpp
CFGPrinter.cpp
CGSCCPassManager.cpp
CMakeLists.txt	Add a new attribute called 'jumptable' that creates jump-instruction tables for functions marked with this attribute.	2014-06-05 19:29:43 +00:00
CodeMetrics.cpp
ConstantFolding.cpp	Look through addrspacecast in IsConstantOffsetFromGlobal	2014-07-14 22:39:26 +00:00
CostModel.cpp	[CostModel][x86] Improved cost model for alternate shuffles.	2014-07-03 22:24:18 +00:00
Delinearization.cpp	remove BasePointer before delinearizing	2014-05-27 22:41:51 +00:00
DependenceAnalysis.cpp	remove BasePointer before delinearizing	2014-05-27 22:41:51 +00:00
DominanceFrontier.cpp	Templatify DominanceFrontier.	2014-07-12 21:59:52 +00:00
DomPrinter.cpp
InstCount.cpp
InstructionSimplify.cpp	Rectify r213231. Use proper version of 'ComputeNumSignBits'.	2014-07-17 19:07:00 +00:00
Interval.cpp
IntervalPartition.cpp
IVUsers.cpp	Add back functionality removed in r210497.	2014-06-21 02:43:02 +00:00
JumpInstrTableInfo.cpp	Add a new attribute called 'jumptable' that creates jump-instruction tables for functions marked with this attribute.	2014-06-05 19:29:43 +00:00
LazyCallGraph.cpp
LazyValueInfo.cpp
LibCallAliasAnalysis.cpp
LibCallSemantics.cpp
Lint.cpp	Revert "Introduce a string_ostream string builder facilty"	2014-06-26 22:52:05 +00:00
LLVMBuild.txt
Loads.cpp
LoopInfo.cpp
LoopPass.cpp	Add back functionality removed in r210497.	2014-06-21 02:43:02 +00:00
Makefile
MemDepPrinter.cpp
MemoryBuiltins.cpp
MemoryDependenceAnalysis.cpp
ModuleDebugInfoPrinter.cpp
NoAliasAnalysis.cpp	Improve BasicAA CS-CS queries (redux)	2014-07-17 01:28:25 +00:00
PHITransAddr.cpp
PostDominators.cpp
PtrUseVisitor.cpp
README.txt
RegionInfo.cpp
RegionPass.cpp	Add back functionality removed in r210497.	2014-06-21 02:43:02 +00:00
RegionPrinter.cpp
ScalarEvolution.cpp	ScalarEvolution: Derive element size from the type of the loaded element	2014-06-08 19:21:20 +00:00
ScalarEvolutionAliasAnalysis.cpp
ScalarEvolutionExpander.cpp	SCEVExpander: Fold constant PHIs harder. The logic below only understands proper IVs.	2014-06-21 11:47:18 +00:00
ScalarEvolutionNormalization.cpp	test check-in: added missing parenthesis in comment	2014-05-28 19:03:33 +00:00
SparsePropagation.cpp
TargetTransformInfo.cpp
Trace.cpp
TypeBasedAliasAnalysis.cpp
ValueTracking.cpp	Teach computeKnownBits to look through addrspacecast.	2014-07-15 01:55:03 +00:00

README.txt

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))

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