aboutsummaryrefslogtreecommitdiff
path: root/lib/Analysis
AgeCommit message (Collapse)Author
2009-11-19Comparing a pointer with null is not a capture.Dan Gohman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89389 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-19Teach IVUsers to keep things simpler and track loop-invariant strides onlyJim Grosbach
for uses inside the loop. This works better with LSR. Disabled behind -simplify-iv-users while benchmarking. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89299 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-18Simplify ComputeMultiple so that it doesn't depend on TargetData.Dan Gohman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89175 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-17Remove dead code.Devang Patel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89156 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-16typo spotted by duncan.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@88884 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-15teach LVI to infer edge information from switch instructions.Chris Lattner
This allows JT to eliminate a ton of infeasible edges when handling code like the templates in PatternMatch.h git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@88869 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-15fix a logic error that would cause LVI-JT to miscompileChris Lattner
some conditionals git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@88868 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-15implement the first stab at caching queries. This isn't correctChris Lattner
(because the invalidation logic is missing) but LVI isn't enabled by default anyway. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@88867 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-15refactor a bunch of code forming the new LazyValueInfoCacheChris Lattner
and LVIQuery classes, no functionality change. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@88866 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-14Teach BasicAA that a constant expression can't alias memory provably notNick Lewycky
allocated until runtime (such as an alloca). Patch by Hans Wennborg! git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@88760 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-13Revert r87059 for now. It is failing clang tests.Devang Patel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@87070 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-13Remove unnecessary llvm.dbg.declare bitcastVictor Hernandez
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@87059 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-12"Attach debug info with llvm instructions" mode was enabled a month ago. Now ↵Devang Patel
make it permanent and remove old way of inserting intrinsics to encode debug info for line number and scopes. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@87014 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-12- Teach LSR to avoid changing cmp iv stride if it will create an immediate thatEvan Cheng
cannot be folded into target cmp instruction. - Avoid a phase ordering issue where early cmp optimization would prevent the later count-to-zero optimization. - Add missing checks which could cause LSR to reuse stride that does not have users. - Fix a bug in count-to-zero optimization code which failed to find the pre-inc iv's phi node. - Remove, tighten, loosen some incorrect checks disable valid transformations. - Quite a bit of code clean up. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86969 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-12various fixes to the lattice transfer functions.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86952 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-12Add a new getPredicateOnEdge method which returns more rich information forChris Lattner
constant constraints. Improve the LVI lattice to include inequality constraints. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86950 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-12expose edge information and switch j-t to use it.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86920 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-12move some stuff into DEBUG's and turn on lazy-value-info forChris Lattner
the basic.ll testcase. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86918 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-12Do not use StringRef in DebugInfo interface. Devang Patel
This allows StringRef to skip controversial if(str) check in constructor. Buildbots, wait for corresponding clang and llvm-gcc FE check-ins! git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86914 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-11make LazyValueInfo actually to some stuff. This isn't very tested but improvesChris Lattner
strswitch. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86889 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-11stub out some LazyValueInfo interfaces, and have JumpThreadingChris Lattner
start using them in a trivial way when -enable-jump-threading-lvi is passed. enable-jump-threading-lvi will be my playground for awhile. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86789 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-11Stub out a new lazy value info pass, which will eventuallyChris Lattner
vend value constraint information to the optimizer. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86767 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-11remove redundant foward declaration. This function is already in Chris Lattner
Analysis/Passes.h git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86765 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-10Implement support to debug inlined functions.Devang Patel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86748 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-10move some generally useful functions out of jump threadingChris Lattner
into libanalysis and transformutils. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86735 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-10Process InlinedAt location info.Devang Patel
Update InsertDeclare to return newly inserted llvm.dbg.declare intrinsic. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86727 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-10Update computeArraySize() to use ComputeMultiple() to determine the array ↵Victor Hernandez
size associated with a malloc; also extend PerformHeapAllocSRoA() to check if the optimized malloc's arg had its highest bit set, so that it is safe for ComputeMultiple() to look through sext instructions while determining the optimized malloc's array size git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86676 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-10Add ComputeMultiple() analysis function that recursively determines if a ↵Victor Hernandez
Value V is a multiple of unsigned Base git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86675 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-10I misread the parens, not so redundant after all.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86648 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-10remove some redundant parens.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86645 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-10add a new SimplifyInstruction API, which is like ConstantFoldInstruction, Chris Lattner
except that the result may not be a constant. Switch jump threading to use it so that it gets things like (X & 0) -> 0, which occur when phi preds are deleted and the remaining phi pred was a zero. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86637 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-10Fix DenseMap iterator constness.Jeffrey Yasskin
This patch forbids implicit conversion of DenseMap::const_iterator to DenseMap::iterator which was possible because DenseMapIterator inherited (publicly) from DenseMapConstIterator. Conversion the other way around is now allowed as one may expect. The template DenseMapConstIterator is removed and the template parameter IsConst which specifies whether the iterator is constant is added to DenseMapIterator. Actually IsConst parameter is not necessary since the constness can be determined from KeyT but this is not relevant to the fix and can be addressed later. Patch by Victor Zverovich! git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86636 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-10factor simplification logic for AND and OR out to InstSimplify from instcombine.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86635 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-09pull a bunch of logic out of instcombine into instsimplify for compare Chris Lattner
simplification, this handles the foldable fcmp x,x cases among many others. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86627 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-09Pass the (optional) TargetData object to ConstantFoldInstOperandsDan Gohman
and ConstantFoldCompareInstOperands. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86626 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-09rename SimplifyCompare -> SimplifyCmpInst and split it into Chris Lattner
Simplify[IF]Cmp pieces. Add some predicates to CmpInst to determine whether a predicate is fp or int. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86624 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-09fix ConstantFoldCompareInstOperands to take the LHS/RHS asChris Lattner
individual operands instead of taking a temporary array git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86619 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-09stub out a new libanalysis "instruction simplify" interface thatChris Lattner
takes decimated instructions and applies identities to them. This is pretty minimal at this point, but I plan to pull some instcombine logic out into these and similar routines. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86613 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-09Default-addressspace null pointers don't alias anything. This allowsDan Gohman
GVN to be more aggressive. Patch by Hans Wennborg! (with a comment added by me) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86582 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-09Minor tidiness fixes.Dan Gohman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86565 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-07Re-commit r86077 now that r86290 fixes the 179.art and 175.vpr ARM regressions.Victor Hernandez
Here is the original commit message: This commit updates malloc optimizations to operate on malloc calls that have constant int size arguments. Update CreateMalloc so that its callers specify the size to allocate: MallocInst-autoupgrade users use non-TargetData-computed allocation sizes. Optimization uses use TargetData to compute the allocation size. Now that malloc calls can have constant sizes, update isArrayMallocHelper() to use TargetData to determine the size of the malloced type and the size of malloced arrays. Extend getMallocType() to support malloc calls that have non-bitcast uses. Update OptimizeGlobalAddressOfMalloc() to optimize malloc calls that have non-bitcast uses. The bitcast use of a malloc call has to be treated specially here because the uses of the bitcast need to be replaced and the bitcast needs to be erased (just like the malloc call) for OptimizeGlobalAddressOfMalloc() to work correctly. Update PerformHeapAllocSRoA() to optimize malloc calls that have non-bitcast uses. The bitcast use of the malloc is not handled specially here because ReplaceUsesOfMallocWithGlobal replaces through the bitcast use. Update OptimizeOnceStoredGlobal() to not care about the malloc calls' bitcast use. Update all globalopt malloc tests to not rely on autoupgraded-MallocInsts, but instead use explicit malloc calls with correct allocation sizes. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86311 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-06Tolerate invalid derived type.Devang Patel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86269 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-06Do not bother to emit debug info for nameless global variable.Devang Patel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86259 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-06remove a bunch of extraneous LLVMContext argumentsChris Lattner
from various APIs, addressing PR5325. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86231 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-06Revert r86077 because it caused crashes in 179.art and 175.vpr on ARMVictor Hernandez
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86213 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-05Fix IVUsers to avoid assuming that the loop has a unique backedge.Dan Gohman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86161 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-05Factor out the predicate code for loopsimplify form exit blocks intoDan Gohman
a separate helper function. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86159 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-05Update CreateMalloc so that its callers specify the size to allocate:Victor Hernandez
MallocInst-autoupgrade users use non-TargetData-computed allocation sizes. Optimization uses use TargetData to compute the allocation size. Now that malloc calls can have constant sizes, update isArrayMallocHelper() to use TargetData to determine the size of the malloced type and the size of malloced arrays. Extend getMallocType() to support malloc calls that have non-bitcast uses. Update OptimizeGlobalAddressOfMalloc() to optimize malloc calls that have non-bitcast uses. The bitcast use of a malloc call has to be treated specially here because the uses of the bitcast need to be replaced and the bitcast needs to be erased (just like the malloc call) for OptimizeGlobalAddressOfMalloc() to work correctly. Update PerformHeapAllocSRoA() to optimize malloc calls that have non-bitcast uses. The bitcast use of the malloc is not handled specially here because ReplaceUsesOfMallocWithGlobal replaces through the bitcast use. Update OptimizeOnceStoredGlobal() to not care about the malloc calls' bitcast use. Update all globalopt malloc tests to not rely on autoupgraded-MallocInsts, but instead use explicit malloc calls with correct allocation sizes. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86077 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-04While calculating original type size for a derived type, handle type ↵Devang Patel
variants encoded as DIDerivedType appropriately. This improves bitfield support. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86073 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-03Changes requested (avoid getFunction(), avoid Type creation via isVoidTy(), ↵Victor Hernandez
and avoid redundant isFreeCall cases) in feedback to r85176 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@85936 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-18-g5258 (git 2.32.0) at 2025-07-16 16:40:34 +0000