llvm-6502

mirror of https://github.com/c64scene-ar/llvm-6502.git synced 2025-01-04 21:30:49 +00:00

History

Nick Lewycky 614fef6d5a Make SCEV's brute force analysis stronger in two ways. Firstly, we should be able to constant fold load instructions where the argument is a constant. Second, we should be able to watch multiple PHI nodes through the loop; this patch only supports PHIs in loop headers, more can be done here. With this patch, we now constant evaluate: static const int arr[] = {1, 2, 3, 4, 5}; int test() { int sum = 0; for (int i = 0; i < 5; ++i) sum += arr[i]; return sum; } git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142731 91177308-0d34-0410-b5e6-96231b3b80d8		2011-10-22 19:58:20 +00:00
..
IPA	Simplify; no intended functional change.	2011-10-20 03:23:14 +00:00
AliasAnalysis.cpp	Enhance alias analysis for atomic instructions a bit. Upgrade a couple alias-analysis tests to the new atomic instructions.	2011-09-26 20:15:28 +00:00
AliasAnalysisCounter.cpp
AliasAnalysisEvaluator.cpp
AliasDebugger.cpp
AliasSetTracker.cpp
Analysis.cpp
BasicAliasAnalysis.cpp	Remove the old atomic instrinsics. autoupgrade functionality is included with this patch.	2011-10-06 23:20:49 +00:00
BlockFrequencyInfo.cpp	Add pass printing support to BlockFrequencyInfo pass. The implementation	2011-10-19 10:12:41 +00:00
BranchProbabilityInfo.cpp	Extend the floating point heuristic to consider NaN checks unlikely.	2011-10-21 21:13:47 +00:00
CaptureTracking.cpp
CFGPrinter.cpp
CMakeLists.txt
ConstantFolding.cpp
DbgInfoPrinter.cpp
DebugInfo.cpp	Update DebugInfoFinder to match recent debug info encoding changes.	2011-10-17 22:30:34 +00:00
DIBuilder.cpp	Add a new wrapper node for a DILexicalBlock that encapsulates it and a	2011-10-11 22:59:11 +00:00
DominanceFrontier.cpp
DomPrinter.cpp
InlineCost.cpp	A FIXME about block addresses and indirectbr.	2011-10-20 04:05:33 +00:00
InstCount.cpp
InstructionSimplify.cpp
Interval.cpp
IntervalPartition.cpp
IVUsers.cpp	Slightly more useful tracing.	2011-10-13 17:06:38 +00:00
LazyValueInfo.cpp
LibCallAliasAnalysis.cpp
LibCallSemantics.cpp
Lint.cpp
Loads.cpp
LoopDependenceAnalysis.cpp
LoopInfo.cpp
LoopPass.cpp
Makefile
MemDepPrinter.cpp	Enhance the memdep interface so that users can tell the difference between a dependency which cannot be calculated and a path reaching the entry point of the function. This patch introduces isNonFuncLocal, which replaces isUnknown in some cases.	2011-10-13 22:14:57 +00:00
MemoryBuiltins.cpp
MemoryDependenceAnalysis.cpp	Enhance the memdep interface so that users can tell the difference between a dependency which cannot be calculated and a path reaching the entry point of the function. This patch introduces isNonFuncLocal, which replaces isUnknown in some cases.	2011-10-13 22:14:57 +00:00
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
ScalarEvolution.cpp	Make SCEV's brute force analysis stronger in two ways. Firstly, we should be	2011-10-22 19:58:20 +00:00
ScalarEvolutionAliasAnalysis.cpp
ScalarEvolutionExpander.cpp	Fix SCEVExpander assert during LSR: "argument of incompatible type".	2011-10-15 06:19:55 +00:00
ScalarEvolutionNormalization.cpp	Fix memory corruption I introduced a few checkins ago.	2011-10-13 18:49:23 +00:00
SparsePropagation.cpp
Trace.cpp
TypeBasedAliasAnalysis.cpp
ValueTracking.cpp	"@string = constant i8 0" is a value i8* string of length zero. Analyze that	2011-10-20 00:34:35 +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))

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