LibCallSimplifier: optimize speed for short-lived instances
[oota-llvm.git] / lib / Transforms /
2013-03-12 Meador IngeLibCallSimplifier: optimize speed for short-lived instances
2013-03-11 Bill WendlingDon't remove a landing pad if the invoke requires a...
2013-03-10 Nick LewyckyUse LLVMBool instead of 'bool' in the C API. Based...
2013-03-10 Hal FinkelBBVectorize: Fixup debugging statements
2013-03-09 Benjamin KramerRemove a source of nondeterminism from the LoopVectorizer.
2013-03-09 Arnold SchwaighoferLoopVectorizer: Ignore all dbg intrinisic
2013-03-09 Arnold SchwaighoferLoopVectorizer: Ignore dbg.value instructions
2013-03-09 Jakub StaszakSimplify code. No functionality change.
2013-03-09 Nick LewyckyUse the correct index variable. This is the meat of...
2013-03-09 Nick LewyckyFix bug introduced in r176616 when making function...
2013-03-09 Nick LewyckyDon't emit the extra checksum into the .gcda file if...
2013-03-08 Benjamin KramerInsert the reduction start value into the first bypass...
2013-03-07 Jakub StaszakKeep coding stanard.
2013-03-07 Jakub StaszakDon't create IRBuilder if we can return from the method...
2013-03-07 Pekka JaaskelainenFixed a crash when cloning a function into a function...
2013-03-07 Nick LewyckySwitch from a version 4.2/4.4 switch to a four-byte...
2013-03-07 Andrew TrickSimplifyCFG fix for volatile load/store.
2013-03-06 Andrew TrickGeneralize my previous fix for -print-options.
2013-03-06 Andrew TrickGive -loop-vectorize an explicit default.
2013-03-06 Jim GrosbachInstCombine: Don't shrink allocas when combining with...
2013-03-04 Lang HamesCheck isDiscardableIfUnused, rather than hasLocalLinkag...
2013-03-04 Preston GurdBypass Slow Divides
2013-03-02 Nadav RotemPR14448 - prevent the loop vectorizer from vectorizing...
2013-03-02 Peter CollingbourneModify {Call,Invoke}Inst::addAttribute to take an AttrKind.
2013-03-01 Benjamin KramerLoopVectorize: Don't hang forever if a PHI only has...
2013-02-28 Quentin ColombetFix a bug in instcombine for fmul in fast math mode.
2013-02-28 Evgeniy Stepanov[msan] Implement sanitize_memory attribute.
2013-02-28 Evgeniy StepanovRemove unused leftover declarations.
2013-02-27 Benjamin KramerLoopVectorize: Vectorize math builtin calls.
2013-02-27 Nick LewyckyIn GCC 4.7, function names are now forbidden from ...
2013-02-27 Nick LewyckyDoh, fix behaviour change introduced in r176168 which...
2013-02-27 Nadav RotemFor each function that we optimize we initialize a...
2013-02-27 Nick LewyckyIRBuilder has grown all sorts of useful utility functio...
2013-02-26 Pedro ArtigasEnhance integer division emulation support to handle...
2013-02-26 Kostya SerebryanyUnify clang/llvm attributes for asan/tsan/msan (LLVM...
2013-02-24 Benjamin KramerCVP: If we have a PHI with an incoming select, try...
2013-02-23 Michael GottesmanFixed a careless mistake.
2013-02-22 Bill WendlingImplement the NoBuiltin attribute.
2013-02-21 Renato GolinAllow GlobalValues to vectorize with AliasAnalysis
2013-02-21 Chad RosierRemove dead code and whitespace.
2013-02-20 Chad RosierUpdate a comment that looks to have been accidentally...
2013-02-20 Kostya Serebryany[asan] instrument invoke insns with noreturn attribute...
2013-02-19 Jakub StaszakRemove unused variable.
2013-02-19 Jakub StaszakMinor cleanups. No functionality change.
2013-02-19 Jakub StaszakRemove unneeded #includes.
2013-02-19 Jakub StaszakFix typos.
2013-02-19 Kostya Serebryany[asan] instrument memory accesses with unusual sizes
2013-02-19 Bill WendlingTemporarily revert r175470 for more review.
2013-02-18 Bill WendlingCheck to see if the 'no-builtin' attribute is set befor...
2013-02-18 Kostya Serebryany[asan] revert r175266 as it breaks code with packed...
2013-02-17 Hal FinkelBBVectorize: Fix an invalid reference bug
2013-02-16 Bill WendlingThe transform is:
2013-02-16 Jakub StaszakReduce indents in LSRInstance::NarrowSearchSpaceByColla...
2013-02-15 Hal FinkelBBVectorize: Call a DAG and DAG instead of a tree
2013-02-15 Arnaud A. de Grand... Fix refactoring mistake in "Teach InstCombine to work...
2013-02-15 Arnaud A. de Grand... Teach InstCombine to work with smaller legal types...
2013-02-15 Kostya Serebryany[asan] support long double on 64-bit. See https://code...
2013-02-15 Benjamin KramerMake helpers static. Add missing include so LLVMInitial...
2013-02-15 Hal FinkelBBVectorize: Cap the number of candidate pairs in each...
2013-02-14 Hal FinkelBBVectorize: Remove the remaining instances of std...
2013-02-14 Hal FinkelBBVectorize: Don't store candidate pairs in a std:...
2013-02-13 Bill WendlingRetain the name of the new internal global that's been...
2013-02-13 Benjamin KramerLoopVectorize: Simplify code for clarity.
2013-02-13 Pekka JaaskelainenMetadata for annotating loops as parallel. The first...
2013-02-13 Kostya Serebryany[asan] fix confusing indentation
2013-02-13 Arnaud A. de Grand... Fix comment
2013-02-12 Michael GottesmanChanged isStoredObjCPointer => IsStoredObjCPointer...
2013-02-12 Dan GohmanActually delete this code, since it's really not clear...
2013-02-12 Dan GohmanRecord PRE predecessors with a SmallVector instead...
2013-02-12 Dan GohmanWhen disabling PRE for a value is directly redundant...
2013-02-12 Dan GohmanCheck that pointers are removed from maps before callin...
2013-02-12 Dan GohmanMinor code simplification.
2013-02-12 Alexander Potapenko[ASan] Do not use kDefaultShort64bitShadowOffset on...
2013-02-12 Kostya Serebryany[asan] change the default mapping offset on x86_64...
2013-02-11 Hal FinkelBBVectorize: Don't over-search when building the depend...
2013-02-11 Hal FinkelBBVectorize: Omit unnecessary entries in PairableInstUsers
2013-02-11 Michael IlsemanOptimization: bitcast (<1 x ...> insertelement ......
2013-02-11 Hal FinkelBBVectorize: Eliminate one more restricted linear search
2013-02-11 Kostya Serebryany[asan] added a flag -mllvm asan-short-64bit-mapping...
2013-02-11 Hal FinkelBBVectorize: Remove the linear searches from pair conne...
2013-02-11 Hal FinkelBBVectorize: Avoid linear searches within the load...
2013-02-11 Hal FinkelBBVectorize: isa/cast cleanup in getInstructionTypes
2013-02-11 Hal FinkelBBVectorize: Make the bookkeeping to support full cycle...
2013-02-09 Andrew TrickLSR IVChain improvement.
2013-02-09 Jakub StaszakRemove #includes from the commonly used LoopInfo.h.
2013-02-08 Bob WilsonRevert "Add LLVMContext::emitWarning methods and use...
2013-02-08 Hal FinkelBBVectorize: Use TTI->getAddressComputationCost
2013-02-08 Chad Rosier[SimplifyLibCalls] Library call simplification doen...
2013-02-08 Jakob Stoklund OlesenTypos.
2013-02-08 Arnold SchwaighoferARM cost model: Address computation in vector mem ops...
2013-02-08 Michael KupersteinTest Commit
2013-02-08 Andrew TrickRevert "Have InstCombine call SipmlifyCall when handlin...
2013-02-07 Michael IlsemanHave InstCombine call SipmlifyCall when handling calls...
2013-02-07 Nadav Rotemfix 80-col violation and fix the docs.
2013-02-07 Arnold SchwaighoferLoop Vectorizer: Refactor Memory Cost Computation
2013-02-07 Michael GottesmanMoved some comments due to the recent refactoring of...
2013-02-07 Michael IlsemanPreserve fast-math flags after reassociation and commut...
2013-02-05 Benjamin KramerInstCombine: Fix and simplify the inttoptr side too.
2013-02-05 Michael GottesmanRemoved explicit inline as per the LLVM style guide.
2013-02-05 Benjamin KramerInstCombine: Harden code to work with vectors of pointe...
next