Gitiles
Code Review Sign In
gerrit-public.fairphone.software / fp2-dev / platform / external / llvm / 8af808a3d64dfdcc5d126e6ad762f57b1483671c / lib / CodeGen / SelectionDAG / ScheduleDAGRRList.cpp
  1. 8af808a Handle SUBREG_TO_REG instructions with the same heuristics by Dan Gohman · 15 years ago
  2. 8f4aa33 Minor compile-time optimization; don't bother checking by Dan Gohman · 15 years ago
  3. 002b44f Add a pre-pass to the burr-list scheduler which makes adjustments to by Dan Gohman · 15 years ago
  4. fa9afef When unfolding a load during scheduling, the new operator node has by Dan Gohman · 15 years ago
  5. a5c8ae2 Fix canClobberPhysRegDefs to check all SDNodes grouped together by Dan Gohman · 15 years ago
  6. 697cbbf For inline asm output operand that matches an input. Encode the input operand index in the high bits. by Evan Cheng · 15 years ago
  7. b398fca Fix a typo in a comment. by Dan Gohman · 15 years ago
  8. 16e8eda Fix ScheduleDAGRRList::CopyAndMoveSuccessors' handling of nodes by Dan Gohman · 15 years ago
  9. 599a6a8 Fix PR3701. 1. X86 target renamed eflags register to flags. This matches what llvm-gcc generates so codegen knows flags register is being clobbered by inline asm. 2. BURR scheduler should also check if inline asm nodes can clobber "live" physical registers. Previously it was only checking target nodes with implicit defs. by Evan Cheng · 15 years ago
  10. f2b1471 Oops. Last second clean up messed things up. by Evan Cheng · 15 years ago
  11. 8182347 Replace one of burr scheduling heuristic with something more sensible. Now calcMaxScratches simply compute the number of true data dependencies. This actually improve a couple of tests in dejagnu suite as many tests in llvm nightly test suite. by Evan Cheng · 15 years ago
  12. c7fa16f Fix a comment. by Dan Gohman · 15 years ago
  13. c8db34c Don't use special heuristics for nodes with no data predecessors by Dan Gohman · 15 years ago
  14. cc994ea Delete the heuristic for non-livein CopyFromReg nodes. Non-liveinness by Dan Gohman · 15 years ago
  15. 47ac0f0 When scheduling a block in parts, keep track of the overall by Dan Gohman · 15 years ago
  16. 9e64bbb Factor out more code for computing register live-range informationfor by Dan Gohman · 15 years ago
  17. f0e366a Fix PR3457: Ignore control successors when looking for closest scheduled successor. A control successor doesn't read result(s) produced by the scheduling unit being evaluated. by Evan Cheng · 15 years ago
  18. 84fbac5 Move ScheduleDAGSDNodes.h to be a private header. Front-ends by Dan Gohman · 16 years ago
  19. 1cc6b8e Make a few things const, fix some comments, and simplify some assertions. by Dan Gohman · 16 years ago
  20. 79ce276 Move a few containers out of ScheduleDAGInstrs::BuildSchedGraph by Dan Gohman · 16 years ago
  21. bc0b567 The list-td and list-tdrr schedulers don't yet support physreg by Dan Gohman · 16 years ago
  22. c29a56d Fix PR3241: Currently EmitCopyFromReg emits a copy from the physical register to a virtual register unless it requires an expensive cross class copy. That means we are only treating "expensive to copy" register dependency as physical register dependency. by Evan Cheng · 16 years ago
  23. 84036a7 Minor debug output tweak. by Evan Cheng · 16 years ago
  24. 25fd403 Remove redundant 'else's. No functionality change. by Dan Gohman · 16 years ago
  25. ce0d4b7 Update these argument lists for the isNormalMemory by Dan Gohman · 16 years ago
  26. fd2163b Use a latency value of 0 for the artificial edges inserted by by Dan Gohman · 16 years ago
  27. d68a076 Tidy up #includes, deleting a bunch of unnecessary #includes. by Dan Gohman · 16 years ago
  28. 1eb9268 CommuteNodesToReducePressure() is now removed. by Dan Gohman · 16 years ago
  29. 223942a Remove the code from the scheduler that commuted two-address by Dan Gohman · 16 years ago
  30. c9a5b9e Rename BuildSchedUnits to BuildSchedGraph, and refactor the by Dan Gohman · 16 years ago
  31. 3f23744 Fix some register-alias-related bugs in the post-RA scheduler liveness by Dan Gohman · 16 years ago
  32. ffa3912 Make addPred and removePred return void, since the return value is not by Dan Gohman · 16 years ago
  33. 54e4c36 Rewrite the SDep class, and simplify some of the related code. by Dan Gohman · 16 years ago
  34. b29ffc8 Make debug output more informative. by Dan Gohman · 16 years ago
  35. 21d9003 Initial support for anti-dependence breaking. Currently this code does not by Dan Gohman · 16 years ago
  36. 942b8c8 Check in the rest of this change. The isAntiDep flag needs to be passed by Dan Gohman · 16 years ago
  37. cddd428 Add a flag to SDep for tracking which edges are anti-dependence edges. by Dan Gohman · 16 years ago
  38. 98adea1 Rename SDep's isSpecial to isArtificial, to make this field a little by Dan Gohman · 16 years ago
  39. f209c2c Remove the CycleBound computation code from the ScheduleDAGRRList by Dan Gohman · 16 years ago
  40. 47d1a21 Change these schedulers to not emit no-ops. It turns out that by Dan Gohman · 16 years ago
  41. 57bebbb Remove a remnant of list-burr's fast mode. by Dan Gohman · 16 years ago
  42. 117f3e9 Factor out the SethiUllman numbering logic from the list-burr and by Dan Gohman · 16 years ago
  43. 9e76fea Remove the "fast" form of the list-burr scheduler, and use the by Dan Gohman · 16 years ago
  44. 6be2ee4 Facter AddPseudoTwoAddrDeps and associated infrasructure out of by Dan Gohman · 16 years ago
  45. a1e6d36 Factor out the code for verifying the work of the scheduler, by Dan Gohman · 16 years ago
  46. 343f0c0 Experimental post-pass scheduling support. Post-pass scheduling by Dan Gohman · 16 years ago
  47. c2f9062 Rearrange code to reduce the nesting level. No functionality change. by Dan Gohman · 16 years ago
  48. 1256f5f Tidy up ScheduleNodeBottomUp methods, and make them more by Dan Gohman · 16 years ago
  49. 3cc6243 Change SUnit's dump method to take a ScheduleDAG* instead of a SelectionDAG*. by Dan Gohman · 16 years ago
  50. 2d093f3 Avoid using a loop in ReleasePred and ReleaseSucc methods to compute the by Dan Gohman · 16 years ago
  51. 8123419 Eliminate some trivial differences between the ScheduleNodeTopDown by Dan Gohman · 16 years ago
  52. d23e0f8 Remove the FlaggedNodes member from SUnit. Instead of requiring each SUnit by Dan Gohman · 16 years ago
  53. 550f5af Make the Node member of SUnit private, and add accessors. by Dan Gohman · 16 years ago
  54. a23b3b8 Change ScheduleDAG's DAG member from a reference to a pointer, to prepare by Dan Gohman · 16 years ago
  55. b13af2f In ScheduleDAGRRList::CopyAndMoveSuccessors, create the SUnit for the load by Dan Gohman · 16 years ago
  56. 9b75b37 Change the scheduler accessor methods to accept an explicit TargetMachine by Dan Gohman · 16 years ago
  57. 87e82f0 Remove some unused virtual function bodies. by Dan Gohman · 16 years ago
  58. b8cab92 Fix command-line option printing to print two spaces where needed, by Dan Gohman · 16 years ago
  59. 086ec99 Replace the LiveRegs SmallSet with a simple counter that keeps by Dan Gohman · 16 years ago
  60. 5993258 Don't worry about clobbering physical register defs that aren't used. by Dan Gohman · 16 years ago
  61. ba36cb5 erect abstraction boundaries for accessing SDValue members, rename Val -> Node to reflect semantics by Gabor Greif · 16 years ago
  62. e5a8dc5 Optimize ScheduleDAGRRList's topological sort to use one pass instead by Dan Gohman · 16 years ago
  63. 99a6cb9 disallow direct access to SDValue::ResNo, provide a getter instead by Gabor Greif · 16 years ago
  64. 04f4f4f Correct the filename in the top-of-file comment. by Dan Gohman · 16 years ago
  65. 430b8a2 Fix several const-correctness issues, resolving some -Wcast-qual warnings. by Dan Gohman · 16 years ago
  66. 475871a Rename SDOperand to SDValue. by Dan Gohman · 16 years ago
  67. e8be6c6 Add a new function, ReplaceAllUsesOfValuesWith, which handles bulk by Dan Gohman · 16 years ago
  68. 5e84368 Reapply 53476 and 53480, with a fix so that it properly updates by Dan Gohman · 16 years ago
  69. b5eec33 Back out 53476 and 53480 for now. Somehow they cause llc to miscompile 179.art. by Evan Cheng · 16 years ago
  70. 0f4012f Factor out debugging code into the common base class. by Dan Gohman · 16 years ago
  71. dbe2a5a Add support for putting NamedRegionTimers in TimerGroups, and by Dan Gohman · 16 years ago
  72. c6be777 - Use a faster priority comparison function if -fast. - Code clean up. by Evan Cheng · 16 years ago
  73. 4576f6d Do not use computationally expensive scheduling heuristics with -fast. by Evan Cheng · 16 years ago
  74. 8cb8245 Move a DenseMap's declaration outside of a loop, and just call by Dan Gohman · 16 years ago
  75. 2f1d310 canClobberPhysRegDefs shouldn't called without checking hasPhysRegDefs; by Dan Gohman · 16 years ago
  76. 94d7a5f Remove ScheduleDAG's SUnitMap altogether. Instead, use SDNode's NodeId by Dan Gohman · 16 years ago
  77. 3627e34 Add a priority queue class, which is a wrapper around std::priority_queue by Dan Gohman · 16 years ago
  78. 4c8c830 Change ScheduleDAG's SUnitMap from DenseMap<SDNode*, vector<SUnit*> > by Dan Gohman · 16 years ago
  79. 16c6859 Simplify some template parameterization. by Dan Gohman · 16 years ago
  80. 83ec4b6 Wrap MVT::ValueType in a struct to get type safety by Duncan Sands · 16 years ago
  81. 419cafb Silence the compiler warning differently. The by Duncan Sands · 16 years ago
  82. 5087124 Silence some compiler warnings. by Evan Cheng · 16 years ago
  83. a0201d5 Use std::set instead of std::priority_queue for the RegReductionPriorityQueue. by Roman Levenstein · 16 years ago
  84. 5a2d335 Fix the new scheduler assertion checks to work when by Dan Gohman · 16 years ago
  85. 80792f3 Treat EntryToken nodes as "passive" so that they aren't added to the by Dan Gohman · 16 years ago
  86. fd5da6c Cosmetic changes. by Evan Cheng · 16 years ago
  87. 2a88a5c ifdef out a dead function. Should this be removed? by Chris Lattner · 16 years ago
  88. d7d3ea0 Fix spelling. Thanks, Duncan! :-) by Roman Levenstein · 16 years ago
  89. 95d4184 Speed-up the SumOfUnscheduledPredsOfSuccs by introducing a new function by Roman Levenstein · 16 years ago
  90. 8dba9af Fixed some spelling errors. Thanks, Duncan! by Roman Levenstein · 16 years ago
  91. e513ba4 Some improvements related to the computation of isReachable. by Roman Levenstein · 16 years ago
  92. cffbd25 Fix typos. by Dan Gohman · 16 years ago
  93. 676dd7c When the register allocator runs out of registers, spill a physical register around the def's and use's of the interval being allocated to make it possible for the interval to target a register and spill it right away and restore a register for uses. This likely generates terrible code but is before than aborting. by Evan Cheng · 16 years ago
  94. 917be68 Rename isOperand() to isOperandOf() (and other similar methods). It always confuses me. by Evan Cheng · 16 years ago
  95. 84d4a2b Refactor / clean up code; remove td list scheduler special tie breaker (no real benefit). by Evan Cheng · 16 years ago
  96. 4c71dfe Update gcc 4.3 warnings fix patch with recent head changes by Anton Korobeynikov · 16 years ago
  97. 94ebde1 Revert 47177, which was incorrect. by Dan Gohman · 16 years ago
  98. 53c607f Skip over the defs and start at the uses when looking for operands by Dan Gohman · 16 years ago
  99. 3b66555 Use the TargetInstrDescr to determine the number of operands by Dan Gohman · 16 years ago
  100. 6f0d024 Rename MRegisterInfo to TargetRegisterInfo. by Dan Gohman · 16 years ago