Skip to content
  1. Oct 30, 2009
    • Dan Gohman's avatar
      Fix MachineLICM to use the correct virtual register class when · 49fa51d9
      Dan Gohman authored
      unfolding loads for hoisting.  getOpcodeAfterMemoryUnfold returns the
      opcode of the original operation without the load, not the load
      itself, MachineLICM needs to know the operand index in order to get
      the correct register class. Extend getOpcodeAfterMemoryUnfold to
      return this information.
      
      llvm-svn: 85622
      49fa51d9
  2. Oct 29, 2009
  3. Oct 28, 2009
  4. Oct 25, 2009
  5. Oct 10, 2009
    • Dan Gohman's avatar
      Factor out LiveIntervalAnalysis' code to determine whether an instruction · 87b02d5b
      Dan Gohman authored
      is trivially rematerializable and integrate it into
      TargetInstrInfo::isTriviallyReMaterializable. This way, all places that
      need to know whether an instruction is rematerializable will get the
      same answer.
      
      This enables the useful parts of the aggressive-remat option by
      default -- using AliasAnalysis to determine whether a memory location
      is invariant, and removes the questionable parts -- rematting operations
      with virtual register inputs that may not be live everywhere.
      
      llvm-svn: 83687
      87b02d5b
  6. Oct 09, 2009
  7. Oct 07, 2009
  8. Sep 26, 2009
  9. Aug 22, 2009
  10. Aug 11, 2009
  11. Jul 25, 2009
    • Daniel Dunbar's avatar
      More migration to raw_ostream, the water has dried up around the iostream hole. · 0dd5e1ed
      Daniel Dunbar authored
       - Some clients which used DOUT have moved to DEBUG. We are deprecating the
         "magic" DOUT behavior which avoided calling printing functions when the
         statement was disabled. In addition to being unnecessary magic, it had the
         downside of leaving code in -Asserts builds, and of hiding potentially
         unnecessary computations.
      
      llvm-svn: 77019
      0dd5e1ed
  12. Feb 27, 2009
  13. Feb 05, 2009
  14. Feb 04, 2009
  15. Jan 15, 2009
  16. Jan 05, 2009
  17. Dec 23, 2008
  18. Dec 18, 2008
  19. Oct 03, 2008
  20. Sep 04, 2008
  21. Aug 31, 2008
  22. Aug 20, 2008
  23. Aug 18, 2008
  24. Jul 07, 2008
  25. May 13, 2008
  26. May 12, 2008
  27. Mar 10, 2008
  28. Feb 26, 2008
  29. Feb 10, 2008
  30. Jan 11, 2008
    • Chris Lattner's avatar
      Simplify the side effect stuff a bit more and make licm/sinking · c8226f32
      Chris Lattner authored
      both work right according to the new flags.
      
      This removes the TII::isReallySideEffectFree predicate, and adds
      TII::isInvariantLoad. 
      
      It removes NeverHasSideEffects+MayHaveSideEffects and adds
      UnmodeledSideEffects as machine instr flags.  Now the clients
      can decide everything they need.
      
      I think isRematerializable can be implemented in terms of the
      flags we have now, though I will let others tackle that.
      
      llvm-svn: 45843
      c8226f32
Loading