Gitiles
Code Review Sign In
gerrit-public.fairphone.software / toolchain / llvm-project / 64ddcb0da60d90dc48f47dc00098fed22fd1e9d8 / llvm / lib / Analysis / ValueTracking.cpp
  1. 91ce36c Revert 'Fix a typo 'iff' => 'if''. iff is an abreviation of if and only if. See: http://en.wikipedia.org/wiki/If_and_only_if Commit 164767 by Sylvestre Ledru · 13 years ago
  2. 721cffd Fix a typo 'iff' => 'if' by Sylvestre Ledru · 13 years ago
  3. 228e6d4 Fix integer undefined behavior due to signed left shift overflow in LLVM. by Richard Smith · 13 years ago
  4. 0d44a50 PHINode::hasConstantValue(): return undef if the PHI is fully recursive. by Nuno Lopes · 13 years ago
  5. ed7c24e2 Teach DeadStoreElimination to eliminate exit-block stores with phi addresses. by Dan Gohman · 14 years ago
  6. 34c4869 Just mark the sign bit as known zero, rather than any other irrelevant bits by Duncan Sands · 14 years ago
  7. 28192c9 Fix ValueTracking to conclude that debug intrinsics are safe to by Chandler Carruth · 14 years ago
  8. ba0a6ca Always compute all the bits in ComputeMaskedBits. by Rafael Espindola · 14 years ago
  9. 80c540e Teach CodeGen's version of computeMaskedBits to understand the range metadata. by Rafael Espindola · 14 years ago
  10. 5319053 Add computeMaskedBitsLoad back, as it was the change to instsimplify that by Rafael Espindola · 14 years ago
  11. e27081d Revert r153521 as it's causing large regressions on the nightly testers. by Chad Rosier · 14 years ago
  12. 8e6dbcc Reapply r153423; the original commit was fine. The failing test, distray, had by Chad Rosier · 14 years ago
  13. 08e57e5 Revert r153423 as this is causing failures on our internal nightly testers. by Chad Rosier · 14 years ago
  14. df9b4ad Use the new range metadata in computeMaskedBits and add a new optimization to by Rafael Espindola · 14 years ago
  15. fa30607 Factor out the multiply analysis code in ComputeMaskedBits and apply it to the by Nick Lewycky · 14 years ago
  16. fea3e00 Factor out the analysis of addition and subtraction in ComputeMaskedBits. Reuse by Nick Lewycky · 14 years ago
  17. 1d57ee3 No functionality change. Type::isSized() can be expensive, so avoid calling it by Nick Lewycky · 14 years ago
  18. af3c6fe A few more cases of missing masking in ComputeMaskedBits; found by inspection. by Eli Friedman · 14 years ago
  19. a8b75ac Make sure we don't return bits outside the mask in ComputeMaskedBits. PR12189. by Eli Friedman · 14 years ago
  20. 8213c8a Remove some dead code and tidy things up now that vectors use ConstantDataVector by Chris Lattner · 14 years ago
  21. 0aef16a [unwind removal] Remove all of the code for the dead 'unwind' instruction. There by Bill Wendling · 14 years ago
  22. cf9e8f6 reapply the patches reverted in r149470 that reenable ConstantDataArray, by Chris Lattner · 14 years ago
  23. 17c981a Revert Chris' commits up to r149348 that started causing VMCoreTests unit test to fail. by Argyrios Kyrtzidis · 14 years ago
  24. 997348e remove the last vestiges of llvm::GetConstantStringInfo, in CodeGen. by Chris Lattner · 14 years ago
  25. 108423a Change ConstantArray::get to form a ConstantDataArray when possible, by Chris Lattner · 14 years ago
  26. 61a1d6c progress making the world safe to ConstantDataVector. While by Chris Lattner · 14 years ago
  27. 6705883 use Constant::getAggregateElement to simplify a bunch of code. by Chris Lattner · 14 years ago
  28. 9be5959 Use the right method to get the # elements in a CDS. by Chris Lattner · 14 years ago
  29. f7eb543 teach valuetracking about ConstantDataSequential by Chris Lattner · 14 years ago
  30. 7ac046a Generalize isSafeToSpeculativelyExecute to work on arbitrary by Dan Gohman · 14 years ago
  31. 9442cd0 PatternMatch: Introduce a matcher for instructions with the "exact" bit. Use it to simplify a few matchers. by Benjamin Kramer · 14 years ago
  32. 4ee5747 ComputeMaskedBits: Make knownzero computation more aggressive for ctlz with undef zero. by Benjamin Kramer · 14 years ago
  33. b4039f6 Make some intrinsics safe to speculatively execute. by Nick Lewycky · 14 years ago
  34. 75d7d5e Move Instruction::isSafeToSpeculativelyExecute out of VMCore and by Dan Gohman · 14 years ago
  35. 8abf65a Probably not a good idea to convert a single vector load into a memcpy. We by Chad Rosier · 14 years ago
  36. 3924cb0 by Nadav Rotem · 14 years ago
  37. ca6f8dd Fix a theoretical problem (not seen in the wild): if different instances of a by Duncan Sands · 14 years ago
  38. e7ab1a2 Make SelectionDAG::InferPtrAlignment use llvm::ComputeMaskedBits instead of duplicating the logic for globals. Make llvm::ComputeMaskedBits handle GlobalVariables slightly more aggressively, to match what InferPtrAlignment knew how to do. by Eli Friedman · 14 years ago
  39. 81a2af1 Fix a crash in which a multiplication was being reported as being both negative by Duncan Sands · 14 years ago
  40. 985ba63 A shift of a power of two is a power of two or zero. by Duncan Sands · 14 years ago
  41. 7cb61e5 Reapply commit 143028 with a fix: the problem was casting a ConstantExpr Mul by Duncan Sands · 14 years ago
  42. 1455ce2 Revert Duncan's r143028 expression folding which appears to be the culprit by Bob Wilson · 14 years ago
  43. ba286d7 The maximum power of 2 dividing a power of 2 is itself. This occurs by Duncan Sands · 14 years ago
  44. 1d2bb98 My super-optimizer noticed that we weren't folding this expression to by Duncan Sands · 14 years ago
  45. 4620988 "@string = constant i8 0" is a value i8* string of length zero. Analyze that by Nick Lewycky · 14 years ago
  46. 717d7ed Migrate LLVM and Clang to use the new makeArrayRef(...) functions where previously explicit non-default constructors were used. by Frits van Bommel · 14 years ago
  47. 229907c land David Blaikie's patch to de-constify Type, with a few tweaks. by Chris Lattner · 14 years ago
  48. 57aa636 Convert InsertValueInst and ExtractValueInst APIs to use ArrayRef. by Jay Foad · 14 years ago
  49. 3e334a4 Move onlyUsedByLifetimeMarkers to ValueTracking so that it can be used by other by Nick Lewycky · 14 years ago
  50. b362884 Renamed llvm.x86.sse42.crc32 intrinsics; crc64 doesn't exist. by Chad Rosier · 15 years ago
  51. 83791ce Teach valuetracking that byval arguments with a specified alignment are aligned. by Chris Lattner · 15 years ago
  52. 2a746bf Teach ValueTracking about x86 crc32 intrinsics. by Evan Cheng · 15 years ago
  53. 0ab5e2c Fix a ton of comment typos found by codespell. Patch by by Chris Lattner · 15 years ago
  54. 8baa2c7 Don't assume something which might be a constant expression is an instruction. by Eli Friedman · 15 years ago
  55. f0469af Fix INT_MIN gotcha pointed out by Eli Friedman. by Nick Lewycky · 15 years ago
  56. 5acc751 Teach ComputeMaskedBits about sub nsw. by Benjamin Kramer · 15 years ago
  57. 391a946 ComputeMaskedBits: sub falls through to add, and sub doesn't have the same overflow semantics as add. by Benjamin Kramer · 15 years ago
  58. cc79973 Teach ComputeMaskedBits about nsw on add. I don't think there's anything we can by Nick Lewycky · 15 years ago
  59. 7dc3d47 Fix PR9331. Simplified version of a patch by Jakub Staszak. by Duncan Sands · 15 years ago
  60. e467979 Add more analysis of the sign bit of an srem instruction. If the LHS is negative by Nick Lewycky · 15 years ago
  61. afe4a30 Fix comment. by Nick Lewycky · 15 years ago
  62. 66f4f22 srem doesn't actually have the same resulting sign as its numerator, you could by Nick Lewycky · 15 years ago
  63. c9aab85 Teach value tracking to make use of flags in more situations. by Nick Lewycky · 15 years ago
  64. 29dbbd1 Teach ValueTracking to look at the dividend when determining the sign bit of an by Nick Lewycky · 15 years ago
  65. acf6b07 Stores of null pointers should turn into memset, we weren't recognizing by Chris Lattner · 15 years ago
  66. ac0b62c Tolerate degenerate phi nodes that can occur in the middle of optimization by Nick Lewycky · 15 years ago
  67. 4b397fc Add a m_SignBit pattern for convenience. by Duncan Sands · 15 years ago
  68. 2e9e4f1 Fix typo: should have been testing that X was odd, not V. by Duncan Sands · 15 years ago
  69. 8a33733 APInt has a method for determining whether a number is a power of 2 by Duncan Sands · 15 years ago
  70. 9e9d5b2 In which I discover that zero+zero is zero, d'oh! by Duncan Sands · 15 years ago
  71. fced762 See if this fixes llvm-gcc bootstrap. by Duncan Sands · 15 years ago
  72. d395108 According to my auto-simplifier the most common missed simplifications in by Duncan Sands · 15 years ago
  73. 0f124e1 Give GetUnderlyingObject a TargetData, to keep it in sync by Dan Gohman · 15 years ago
  74. c86e67e fix an off-by-one bug that caused a crash analyzing by Chris Lattner · 15 years ago
  75. 9cb1035 move isBytewiseValue out to ValueTracking.h/cpp by Chris Lattner · 15 years ago
  76. 7aa18bf Add vector versions of some existing scalar transforms to aid codegen in matching psign & pblend operations to the IR produced by clang/gcc for their C idioms. by Nate Begeman · 15 years ago
  77. 05b18f1 Reapply r121886, and also update DecomposeGEPExpression to keep it in sync. by Dan Gohman · 15 years ago
  78. d02b659 Revert r121886. DecomposeGEPExpression needs to be kept in sync. by Dan Gohman · 15 years ago
  79. 949ab78 Strengthen GetUnderlyingObject using InstructionSimplify. by Dan Gohman · 15 years ago
  80. a4fcd24 Move Value::getUnderlyingObject to be a standalone by Dan Gohman · 15 years ago
  81. 583abbc PR5207: Change APInt methods trunc(), sext(), zext(), sextOrTrunc() and by Jay Foad · 15 years ago
  82. 25a5e4c PR5207: Rename overloaded APInt methods set(), clear(), flip() to by Jay Foad · 15 years ago
  83. e28618d move GetPointerBaseWithConstantOffset out of GVN into ValueTracking.h by Chris Lattner · 15 years ago
  84. 15084f0 PR5207: Make APInt::set(), APInt::clear() and APInt::flip() return void. by Jay Foad · 15 years ago
  85. 72d283c fix PR8063, a crash in globalopt in the malloc analysis code. by Chris Lattner · 15 years ago
  86. 9f7500f move gep decomposition out of ValueTracking into BasicAA. The form of by Chris Lattner · 15 years ago
  87. fbeeb13 Revert r111375, "move gep decomposition out of ValueTracking into BasicAA. The by Daniel Dunbar · 15 years ago
  88. 54fe883 move gep decomposition out of ValueTracking into BasicAA. The form of by Chris Lattner · 15 years ago
  89. a33edcb fix PR7589: In brief: gep P, (zext x) != gep P, (sext x) by Chris Lattner · 15 years ago
  90. 1abbde3 use ArgOperand accessors by Gabor Greif · 15 years ago
  91. 7258dcd Revert 101465, it broke internal OpenGL testing. by Eric Christopher · 16 years ago
  92. f375520 reapply r101434 with a fix for self-hosting by Gabor Greif · 16 years ago
  93. 403e969 back out r101423 and r101397, they break llvm-gcc self-host on darwin10 by Gabor Greif · 16 years ago
  94. 33ae80b reapply r101364, which has been backed out in r101368 with a fix by Gabor Greif · 16 years ago
  95. 9fd00c7 back out r101364, as it trips the linux nightlybot on some clang C++ tests by Gabor Greif · 16 years ago
  96. aafd209 rotate CallInst operands, i.e. move callee to the back of the operand array by Gabor Greif · 16 years ago
  97. 0b4df04 Constify GetConstantStringInfo. by Dan Gohman · 16 years ago
  98. 2a65429 Fix a typo in ValueTracking that's causing instcombine to delete needed shift instructions. by Evan Cheng · 16 years ago
  99. 4899cbc Move GetStringLength and helper from SimplifyLibCalls to ValueTracking. by Eric Christopher · 16 years ago
  100. 19d0b47 There are two ways of checking for a given type, for example isa<PointerType>(T) by Duncan Sands · 16 years ago