Gitiles
Code Review Sign In
gerrit-public.fairphone.software / fp2-dev / platform / external / llvm / 6520e20e4fb31f2e65e25c38b372b19d33a83df4 / lib / Target / X86
  1. 6520e20 Teach DAGCombine to fold constant offsets into GlobalAddress nodes, by Dan Gohman · 17 years ago
  2. 97b3827 This is done. by Dan Gohman · 17 years ago
  3. 109a562 Add implicit defs of XMM8 to XMM15 on 32-bit call instructions. While this is not technically true, it tells tblgen that these instructions "clobber" the entire XMM register file. by Evan Cheng · 17 years ago
  4. 1fa7198 add support for 128 bit inputs on both x86-64 and x86-32. by Chris Lattner · 17 years ago
  5. 3d0c4c4 Fix a bug where the x86 backend would reject 64-bit r constraints when by Chris Lattner · 17 years ago
  6. 4b299d4 Fix lfence and mfence encoding. These look like MRM5r and MRM6r instructions except they do not have any operands. The RegModRM byte is encoded with register number 0. by Evan Cheng · 17 years ago
  7. 244911b getX86RegNum has long been moved to X86RegisterInfo. by Evan Cheng · 17 years ago
  8. 63f9720 Fun x86 encoding tricks: when adding an immediate value of 128, by Dan Gohman · 17 years ago
  9. 74feef2 Define patterns for shld and shrd that match immediate by Dan Gohman · 17 years ago
  10. f522068 Trim #includes. by Dan Gohman · 17 years ago
  11. cc4ba78 fix typo noticed by sdt by Chris Lattner · 17 years ago
  12. a354825 add some notes by Chris Lattner · 17 years ago
  13. c74092b add some notes and a file to collect unimplemented features in the by Chris Lattner · 17 years ago
  14. 8e8b8a2 Const-ify several TargetInstrInfo methods. by Dan Gohman · 17 years ago
  15. fcab2bd Remove an unused variable. by Dan Gohman · 17 years ago
  16. a7250dd Fix the predicate for memop64 to be a regular load, not just an unindexed load. by Dan Gohman · 17 years ago
  17. 3358629 Now that predicates can be composed, simplify several of by Dan Gohman · 17 years ago
  18. 241ab47 add support for folding immediates into stores when they by Chris Lattner · 17 years ago
  19. 438949a fold immediates into stores in simple cases, this produces diffs like this: by Chris Lattner · 17 years ago
  20. d53886b fold compare of null pointer into compare with 0. by Chris Lattner · 17 years ago
  21. 160f6cc Some minor cleanups: by Chris Lattner · 17 years ago
  22. 45ac17f Use switch on VT instead of Type* comparisons. by Chris Lattner · 17 years ago
  23. 51ccb3d Use X86FastEmitCompare for FCMP_OEQ and FCMP_UNE: it doesn't by Chris Lattner · 17 years ago
  24. 9a08a61 refactor compare emission out into a new X86FastEmitCompare method, by Chris Lattner · 17 years ago
  25. 0e13c78 Fold immediates into compares when possible, producing "cmp $4, %eax" instead of by Chris Lattner · 17 years ago
  26. 709d829 more minor refactoring of X86SelectBranch, no functionality change. by Chris Lattner · 17 years ago
  27. 871d246 factor buildmi calls in X86SelectBranch by Chris Lattner · 17 years ago
  28. 8aeeeb9 factor some more BuildMI's in X86SelectCmp by Chris Lattner · 17 years ago
  29. 54aebde factor some BuildMI calls, no functionality change. by Chris Lattner · 17 years ago
  30. 7f04268 - Add target lowering hooks that specify which setcc conditions are illegal, by Evan Cheng · 17 years ago
  31. dd5b58a FastISel support for exception-handling constructs. by Dan Gohman · 17 years ago
  32. 2194d46 Accept -march=i586, because gcc does (a synonym by Dale Johannesen · 17 years ago
  33. 0329466 Rename LoadX to LoadExt. by Evan Cheng · 17 years ago
  34. b8cab92 Fix command-line option printing to print two spaces where needed, by Dan Gohman · 17 years ago
  35. 552e3be Fix indentation. by Evan Cheng · 17 years ago
  36. 7b8e964 When doing the very-late shift-and address-mode optimization, by Dan Gohman · 17 years ago
  37. d21a630 Update size of inst correctly with segment override. by Anton Korobeynikov · 17 years ago
  38. e563bbc Change CALLSEQ_BEGIN and CALLSEQ_END to take TargetConstant's as by Chris Lattner · 17 years ago
  39. a4bb48a Fix comment typo. by Duncan Sands · 17 years ago
  40. ef93cec Add ability to override segment (mostly for code emitter purposes). by Anton Korobeynikov · 17 years ago
  41. e397acc Fix SSE4.1 roundss, roundsd. While the instructions have by Dale Johannesen · 17 years ago
  42. 23a9855 Add a "loses information" return value to APFloat::convert by Dale Johannesen · 17 years ago
  43. 8c3f8b6 Add MBB successors and physreg Uses in the same order that by Dan Gohman · 17 years ago
  44. 145b828 Instead of emitting an implicit use for the super-register of by Dan Gohman · 17 years ago
  45. 2cb48ea Model hardwired inputs & outputs of x86 8-bit divides correctly. by Dale Johannesen · 17 years ago
  46. ae436ce Certain patterns involving the "movss" instruction were marked as requiring SSE2, when in reality movss is an SSE1 instruction. by Anders Carlsson · 17 years ago
  47. 882cdfd Fix PR2850 and PR2863. Only generate movddup for 128-bit SSE vector shuffles. by Evan Cheng · 17 years ago
  48. e76225a It is possible that all functions in one module are not being by Devang Patel · 17 years ago
  49. e4d4b8c Remove unncessary isDeclaration() checks. by Devang Patel · 17 years ago
  50. 6381a13 Fix weird think-o and unbreak build on all gcc-3.4.x-based platforms (e.g. mingw) by Anton Korobeynikov · 17 years ago
  51. c4e8bec Revert r56675 - it breaks unwinding runtime everywhere. by Anton Korobeynikov · 17 years ago
  52. 880ae36 Make atomic Swap work, 64-bit on x86-32. Make it all work in non-pic mode. by Dale Johannesen · 17 years ago
  53. 1b54c7f Pass MemOperand through for 64-bit atomics on 32-bit, by Dale Johannesen · 17 years ago
  54. da8ac5f Avoid creating two TargetLowering objects for each target. by Dan Gohman · 17 years ago
  55. 38217fe Remove an unused field. by Dan Gohman · 17 years ago
  56. d735b80 Switch the MachineOperand accessors back to the short names like by Dan Gohman · 17 years ago
  57. 4e6ed5e Fix X86FastISel to handle dynamic allocas that have avoided by Dan Gohman · 17 years ago
  58. d98d620 Optimize conditional branches in X86FastISel. This replaces by Dan Gohman · 17 years ago
  59. 48c1bc2 Handle some 64-bit atomics on x86-32, some of the time. by Dale Johannesen · 17 years ago
  60. 5bbee4b Work around an interaction between fast-isel and regalloc=local. The by Dan Gohman · 17 years ago
  61. 5e249b4 "The original bug was a complaint that _mm_srli_si128 mis-compiled when passed by Bill Wendling · 17 years ago
  62. 4ae641f Remove OptimizeForSize global. Use function attribute optsize. by Devang Patel · 17 years ago
  63. 6d4b052 Split x86's ADJCALLSTACK instructions into 32-bit and 64-bit forms. by Dan Gohman · 17 years ago
  64. 2662d55 Mark CALL instructions as having a Use of ESP/RSP. by Dan Gohman · 17 years ago
  65. 6158d84 Implement the -fno-builtin option in the front-end, not in the back-end. by Bill Wendling · 17 years ago
  66. 6e08738 Just don't transform this memset into "bzero" if no-builtin is specified. by Bill Wendling · 17 years ago
  67. 6f287b2 Add the new `-no-builtin' flag. This flag is meant to mimic the GCC by Bill Wendling · 17 years ago
  68. f06c835 Optimize SelectionDAG's AssignTopologicalOrder even further. by Dan Gohman · 17 years ago
  69. 5396c99 Fix X86FastISel's output for x86-32 PIC constant pool addresses. by Dan Gohman · 17 years ago
  70. 57c3dac Move the GlobalBaseReg field out of X86ISelDAGToDAG.cpp by Dan Gohman · 17 years ago
  71. 9b66d73 Disable all x87 usage, including f32 and f64 when the subtarget by Dan Gohman · 17 years ago
  72. 71d1bf5 Remove misuse of ReplaceNodeResults for atomics with by Dale Johannesen · 17 years ago
  73. 7962e85 Fix an over-pessimization about GlobalVariable addresses in X86FastISel. by Dan Gohman · 17 years ago
  74. d47e0b6 Fix PR2835. Do not change the width of a volatile load. by Evan Cheng · 17 years ago
  75. 5df3186 Rename isWeakForLinker to mayBeOverridden. Use it by Duncan Sands · 17 years ago
  76. 184793f Re-apply 56683 with fixes. by Evan Cheng · 17 years ago
  77. 02a4da2 Unbreak build. by Daniel Dunbar · 17 years ago
  78. 2c9c3e7 Implement function notes as function attributes. by Devang Patel · 17 years ago
  79. b7a75a5 Implement "punpckldq %xmm0, $xmm0" as "pshufd $0x50, %xmm0, %xmm" unless optimizing for code size. by Evan Cheng · 17 years ago
  80. d5d8191 Temporarily reverting r56683. This is causing a failure during the build of llvm-gcc: by Bill Wendling · 17 years ago
  81. 7810bfe Rename ConstantSDNode's getSignExtended to getSExtValue, for by Dan Gohman · 17 years ago
  82. 83ccd14 Mark lea fi# as being really rematerializable. by Dan Gohman · 17 years ago
  83. c739489 unpckhps requires sse1, punpckhdq requires sse2. by Evan Cheng · 17 years ago
  84. 09aae46 Fix X86FastISel's address folding to check displacement values for overflow. by Dan Gohman · 17 years ago
  85. 36a5502 Fix @llvm.frameaddress codegen. FP elimination optimization should be disabled when frame address is desired. Also add support for depth > 0. by Evan Cheng · 17 years ago
  86. 86098bd Add "inreg" field to CallSDNode (doesn't increase by Dale Johannesen · 17 years ago
  87. 97135e1 Factor out the code for determining when symblic addresses by Dan Gohman · 17 years ago
  88. a67f32a Avoid spilling EBP / RBP twice in the prologue. by Evan Cheng · 17 years ago
  89. 4367ff7 X86 address displacement field must be interpreted as a 32-bit value. by Evan Cheng · 17 years ago
  90. e1ad087 CMake: Builds all targets. by Oscar Fuentes · 17 years ago
  91. 5af29c2 Disable support for x86_f80 in X86FastISel. Supporting it would by Dan Gohman · 17 years ago
  92. d751c64 If we have a function with an unreachable statement such that the ending debug by Bill Wendling · 17 years ago
  93. 4dcc8a3 Prefer movlhps over punpcklqdq, etc. in more cases. by Evan Cheng · 17 years ago
  94. 789ce77 Fix a bug in which address displacements were being added to the by Dan Gohman · 17 years ago
  95. 0598866 Large mechanical patch. by Devang Patel · 17 years ago
  96. 32b952a Reapply 56585:56589 with proper fix for some gcc versions by Anton Korobeynikov · 17 years ago
  97. 0b457f0 With sse3 and when the source is a load or has multiple uses, favors movddup over shuffp*, pshufd, etc. Without sse3 or when the source is from a register, make use of movlhps by Evan Cheng · 17 years ago
  98. c9c6da6 Accept 'inreg' attribute on x86 functions as by Dale Johannesen · 17 years ago
  99. 2cc3aa4 PIC support in X86FastISel. by Dan Gohman · 17 years ago
  100. 688535e Temporarily backing out 56585:56589 to unbreak the build. by Evan Cheng · 17 years ago