1. 79ada10 LegalizeTypes support for what seems to be the by Duncan Sands · 16 years ago
  2. 29681a5 Remove unnecessary readme entry by Nate Begeman · 16 years ago
  3. f00e74f Turn LegalizeTypes back off again for the moment: by Duncan Sands · 16 years ago
  4. e99b255 Fix a typo in last commit by Nate Begeman · 16 years ago
  5. 30a0de9 SSE codegen for vsetcc nodes by Nate Begeman · 16 years ago
  6. bc1f989 Make GlobalOpt preserve address spaces when scalar replacing aggregate globals. by Matthijs Kooijman · 16 years ago
  7. d137ab4 Fix PR2553 by Chris Lattner · 16 years ago
  8. f595266 When lowering certain atomics, we need to copy the memoperand from the old by Mon P Wang · 16 years ago
  9. b2ff885 Factorize some code for determining which libcall to use. by Duncan Sands · 16 years ago
  10. 066fcf8 Inliner tweak. Function calls should cost more than one instruction! by Evan Cheng · 16 years ago
  11. b31b06d Factor MergeBlockIntoPredecessor out into BasicBlockUtils. by Owen Anderson · 16 years ago
  12. 4139070 Name string length is end position marker - begin position marker. by Devang Patel · 16 years ago
  13. c2aec57 Do not forget global definitions from inline asm code block. by Devang Patel · 16 years ago
  14. 0d885d1 Mark function used by asm block as used, otherwise optimizer may not see the use and may delete the function. by Devang Patel · 16 years ago
  15. 5d0af03 There's no need to iterate block merging and PRE. In fact, iterating the latter by Owen Anderson · 16 years ago
  16. 5af8f0e Somehow, custom lowering of i64 multiplications got dropped along the way. by Scott Michel · 16 years ago
  17. 2fbdf0e Fix the result type of X86's truncate to i8. by Dan Gohman · 16 years ago
  18. 7a9a5af Fix the result type of a VECTOR_SHUFFLE+BIT_CONVERT dagcombine. This by Dan Gohman · 16 years ago
  19. 9cf80fa Fix the name of BreakCriticalMachineEdge.h's include guard by Dan Gohman · 16 years ago
  20. 58ca8d5 Correct a top-level comment. by Dan Gohman · 16 years ago
  21. 95db39a Add support for promoting and expanding AssertZext by Duncan Sands · 16 years ago
  22. 61e804f Clarify the comments here, to make slightly more clear the by Dan Gohman · 16 years ago
  23. 2853367 Tidy a doxygen comment. by Dan Gohman · 16 years ago
  24. c0d6012 Fix a comment to say nonnegative instead of positive. by Dan Gohman · 16 years ago
  25. c5e1f98 Add an assert to check for empty flags for MachineMemOperand. by Dan Gohman · 16 years ago
  26. 836b3e6 Test passing of integer parameters for integers by Duncan Sands · 16 years ago
  27. 8b5e62f Test codegen of loads and stores of all integer by Duncan Sands · 16 years ago
  28. d8b4d60 Add a few cases to instcombine's extractvalue testcase. by Matthijs Kooijman · 16 years ago
  29. 5f0028e Un-XFAIL multdeadretval, since instcombine now properly handles the mess deadargelim leaves behind :-) by Matthijs Kooijman · 16 years ago
  30. 780ae5e Redo InstCombiner::visitExtractValueInst. Instead of using the (complicate) by Matthijs Kooijman · 16 years ago
  31. 05c397d Reorder methods alphabetically. No functionality change. by Duncan Sands · 16 years ago
  32. 7f1d8ae Turn on LegalizeTypes by default. by Duncan Sands · 16 years ago
  33. dddc827 Don't use ++idx_begin when I actually mean idx_begin + 1, especially since we by Matthijs Kooijman · 16 years ago
  34. 17991c3 The atomic.cmp.swap promotion logic is wrong: it by Duncan Sands · 16 years ago
  35. be8efa5 Fix PR2296. Do not transform x86_sse2_storel_dq into a full-width store. by Evan Cheng · 16 years ago
  36. 817a6a9 x86-64 PIC JIT fixes: do not generate the extra load for external GV's. by Evan Cheng · 16 years ago
  37. 81fb5fe X86-64 PIC jump table values are different from x86-32 cases, they are dest - table base. by Evan Cheng · 16 years ago
  38. 9c578eb TargetAsmInfo::SectionForGlobal showed up in a profile. Simplify it a little. by Dan Gohman · 16 years ago
  39. 4093d83 SelectionDAG::AssignNodeIds is unused. by Dan Gohman · 16 years ago
  40. 90a7b8f Don't sort SDNodes by their addresses in SelectionDAG::dump. Instead, by Dan Gohman · 16 years ago
  41. 2b4f7bb Revert this, as it seems to still be broken. by Owen Anderson · 16 years ago
  42. b33f7cc Enable local PRE by default. by Owen Anderson · 16 years ago
  43. 961edc8 Have GVN do a pre-pass over the CFG that folds away unconditional branches where possible. This allows local PRE to be more aggressive. by Owen Anderson · 16 years ago
  44. b79d2a6 XFAIL the multdeadretval test for now, I will be fixing instcombine to make it work again tomorrow. by Matthijs Kooijman · 16 years ago
  45. 295a7cd LegalizeTypes support for fabs on ppc long double. by Duncan Sands · 16 years ago
  46. 790dd75 Remove a few tests which no longer hold for deadargelim (since it is now by Matthijs Kooijman · 16 years ago
  47. 2439911 Add a testcase for the canonicalizations now performed by deadargelim. by Matthijs Kooijman · 16 years ago
  48. 4f72368 Allow deadargelim to change return types even though now values were dead. This by Matthijs Kooijman · 16 years ago
  49. d1d1de7 Revert r53606. It turns out that explicitely tracking the liveness of the by Matthijs Kooijman · 16 years ago
  50. eb32b45 Make deadargelim a bit less smart, so it doesn't choke on nested structs as by Matthijs Kooijman · 16 years ago
  51. 9c9418d Don't use isa when we can reuse a previous dyn_cast. by Matthijs Kooijman · 16 years ago
  52. ddd1a79 Make DeadArgElim keep liveness of the return value as a whole in addition to by Matthijs Kooijman · 16 years ago
  53. ae15ddf Fix typo. by Matthijs Kooijman · 16 years ago
  54. c07e6e5 LegalizeTypes support for promotion of bswap. by Duncan Sands · 16 years ago
  55. 8d56a6f LegalizeTypes support for promotion of SIGN_EXTEND_INREG. by Duncan Sands · 16 years ago
  56. bf304c2 Reorder the integer promotion methods alphabetically. by Duncan Sands · 16 years ago
  57. 2bf5372 Let DAE keep a list of live functions, instead of simply marking all arguments by Matthijs Kooijman · 16 years ago
  58. 3015652 Split DAE::MarkLive into MarkLive and PropagateLiveness. by Matthijs Kooijman · 16 years ago
  59. 6cdd54b Pass around const RetOrArg references instead of copying values. Also, mark by Matthijs Kooijman · 16 years ago
  60. 03016ca Simplify debug code by using RetOrArg::getDescription(). by Matthijs Kooijman · 16 years ago
  61. 0d1730a Fix indentation (intentionally left out of the previous commit). by Matthijs Kooijman · 16 years ago
  62. a3ec5d6 Move the deadargelim code for intrinsically alive functions into its own by Matthijs Kooijman · 16 years ago
  63. 2920d2b Fixed potential bug if the source and target of a bit convert have different alignment by Mon P Wang · 16 years ago
  64. 8ae38e1 Correct this inversion! I swear that didn't show up in svn diff... by Nick Lewycky · 16 years ago
  65. 86dae65 Fix up comments. by Nick Lewycky · 16 years ago
  66. b27cb55 Fixed call stack alignment. Improved AsmPrinter alignment issues. by Bruno Cardoso Lopes · 16 years ago
  67. f0d286b LinkOnce definitions have default scope, like weak definitions. Otherwise, the linker may not be able to match LinkOnce definition from one module with an exteranl reference from other module. by Devang Patel · 16 years ago
  68. 5f2f218 Goodbye tail duplication (for good this time). by Evan Cheng · 16 years ago
  69. d28516e This replaces all $(SolutionDir) macros with $(ProjectDir)..\ by Chris Lattner · 16 years ago
  70. 5e84368 Reapply 53476 and 53480, with a fix so that it properly updates by Dan Gohman · 16 years ago
  71. 3fea643 Fix uninitialized use of the Changed variable. by Dan Gohman · 16 years ago
  72. 91d49f5 Improve debug output for MemOperandSDNode. PseudoSourceValue nodes by Dan Gohman · 16 years ago
  73. 2b06299 Fix edito in the PseudoSourceValue name list. by Dan Gohman · 16 years ago
  74. cebf59b Reformat this message to fit in 80 cols. by Dan Gohman · 16 years ago
  75. 189a2b3 I don't think BUILD_PAIR can have a vector result. Remove support for this. by Duncan Sands · 16 years ago
  76. a489be5 Tighten up some checks. Fix FPOWI splitting for non-power-of-two vectors. by Duncan Sands · 16 years ago
  77. 46868c0 Reapply r53540, now with the matching header! by Chris Lattner · 16 years ago
  78. 262e04b An INSERT_VECTOR_ELT can insert a larger value by Duncan Sands · 16 years ago
  79. 8cc364c According to the docs, it is possible to have an by Duncan Sands · 16 years ago
  80. 28c05ac There should be no extending loads or truncating by Duncan Sands · 16 years ago
  81. d164ea2 Ignore TargetConstant with an illegal type. These by Duncan Sands · 16 years ago
  82. 43d526d Added Subtarget support into RegisterInfo by Bruno Cardoso Lopes · 16 years ago
  83. 91e1c32 Revert r53540 - it does not compile. by Duncan Sands · 16 years ago
  84. 6157e38 Reimplement LinkFunctionProtos in terms of GetLinkageResult. This fixes by Chris Lattner · 16 years ago
  85. ae1132d don't do any linkage, not even type resolution, of symbols that have by Chris Lattner · 16 years ago
  86. d1ec48c implement linking of globals to functions, in one direction by Chris Lattner · 16 years ago
  87. 0bb8757 wrap long lines, remove some code from a non-assert build. by Chris Lattner · 16 years ago
  88. bbedb0e Fix a bunch of bugs handling vector compare constant expressions, fixing PR2317. by Chris Lattner · 16 years ago
  89. 071aade Document and fix Constant::getVectorElements to return an empty vector by Chris Lattner · 16 years ago
  90. ac77bfd If a function calls setjmp, never inline it into other functions. This is by Chris Lattner · 16 years ago
  91. 4238453 simplify some code, shuffle and insertelt always return a vector. by Chris Lattner · 16 years ago
  92. de29d92 whitespace fix. by Chris Lattner · 16 years ago
  93. 6838d87 doxygenate comments and wrap to 80 cols. by Chris Lattner · 16 years ago
  94. 10c5d36 Add a note. by Chris Lattner · 16 years ago
  95. fdb19e5 Fix PR2506 by being a bit more careful about reverse fact propagation when by Chris Lattner · 16 years ago
  96. 093a438 Fix mishandling of the infinite loop case when merging two blocks. This by Chris Lattner · 16 years ago
  97. b824512 more refactoring. Use early exits instead of really complex logic. by Chris Lattner · 16 years ago
  98. eb388af improve comments. by Chris Lattner · 16 years ago
  99. 867661a factor another large hunk of code out into its own function. by Chris Lattner · 16 years ago
  100. 4f60e4d add a description. by Chris Lattner · 16 years ago