llvm-6502

mirror of https://github.com/c64scene-ar/llvm-6502.git synced 2024-10-05 22:59:16 +00:00

History

Duncan Sands c087e20331 Turn X-(X-Y) into Y. According to my auto-simplifier this is the most common simplification present in fully optimized code (I think instcombine fails to transform some of these when "X-Y" has more than one use). Fires here and there all over the test-suite, for example it eliminates 8 subtractions in the final IR for 445.gobmk, 2 subs in 447.dealII, 2 in paq8p etc. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123442 91177308-0d34-0410-b5e6-96231b3b80d8		2011-01-14 15:26:10 +00:00
..
IPA	Use the new addEscapingValue callback to update GlobalsModRef when GVN adds PHIs of GEPs. For the moment,	2011-01-03 23:51:43 +00:00
AliasAnalysis.cpp	Stub out a new updating interface to AliasAnalysis, allowing stateful analyses to be informed when	2011-01-03 21:38:41 +00:00
AliasAnalysisCounter.cpp
AliasAnalysisEvaluator.cpp
AliasDebugger.cpp
AliasSetTracker.cpp
Analysis.cpp
BasicAliasAnalysis.cpp	fix rdar://8813415 - a miscompilation of 164.gzip that loop-idiom	2011-01-03 21:03:33 +00:00
CaptureTracking.cpp
CFGPrinter.cpp
CMakeLists.txt
ConstantFolding.cpp	Teach constant folding to perform conversions from constant floating	2011-01-11 01:07:24 +00:00
DbgInfoPrinter.cpp
DebugInfo.cpp
DIBuilder.cpp
DomPrinter.cpp	split dom frontier handling stuff out to its own DominanceFrontier header,	2011-01-02 22:09:33 +00:00
InlineCost.cpp
InstCount.cpp
InstructionSimplify.cpp	Turn X-(X-Y) into Y. According to my auto-simplifier this is the most common	2011-01-14 15:26:10 +00:00
Interval.cpp
IntervalPartition.cpp
IVUsers.cpp	split dom frontier handling stuff out to its own DominanceFrontier header,	2011-01-02 22:09:33 +00:00
LazyValueInfo.cpp	Reorder, rename, and document some members to make this easier to follow.	2011-01-05 23:26:22 +00:00
LibCallAliasAnalysis.cpp
LibCallSemantics.cpp
Lint.cpp
LiveValues.cpp	split dom frontier handling stuff out to its own DominanceFrontier header,	2011-01-02 22:09:33 +00:00
Loads.cpp
LoopDependenceAnalysis.cpp	split dom frontier handling stuff out to its own DominanceFrontier header,	2011-01-02 22:09:33 +00:00
LoopInfo.cpp
LoopPass.cpp
Makefile
MemDepPrinter.cpp
MemoryBuiltins.cpp
MemoryDependenceAnalysis.cpp	Revert r123207: "Turn on memdep's verifyRemoved() in an attempt to smoke out the cause of our gcc bootstrap miscompare."	2011-01-11 04:05:39 +00:00
ModuleDebugInfoPrinter.cpp
NoAliasAnalysis.cpp	Use the new addEscapingValue callback to update GlobalsModRef when GVN adds PHIs of GEPs. For the moment,	2011-01-03 23:51:43 +00:00
PHITransAddr.cpp	split dom frontier handling stuff out to its own DominanceFrontier header,	2011-01-02 22:09:33 +00:00
PostDominators.cpp	split dom frontier handling stuff out to its own DominanceFrontier header,	2011-01-02 22:09:33 +00:00
ProfileEstimatorPass.cpp
ProfileInfo.cpp
ProfileInfoLoader.cpp
ProfileInfoLoaderPass.cpp
ProfileVerifierPass.cpp
README.txt
RegionInfo.cpp	Add single entry / single exit accessors.	2011-01-13 23:18:04 +00:00
RegionPass.cpp
RegionPrinter.cpp
ScalarEvolution.cpp	some comment improvements.	2011-01-11 17:11:59 +00:00
ScalarEvolutionAliasAnalysis.cpp
ScalarEvolutionExpander.cpp
ScalarEvolutionNormalization.cpp
SparsePropagation.cpp
Trace.cpp
TypeBasedAliasAnalysis.cpp
ValueTracking.cpp	fix an off-by-one bug that caused a crash analyzing	2011-01-04 18:19:15 +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))

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