Commits · a60c61dc1ae69f21d3ac66ce5099d11d4e75335b · Lorenzo Albano / LLVM bpEVL · GitLab

Jan 26, 2008
- If there are no machine instructions emitted for a function, then insert · 50794839
  Bill Wendling authored Jan 26, 2008
  50794839
- Create an explicit copy for byval parameters even · c4dc3dc3
  Duncan Sands authored Jan 26, 2008
  c4dc3dc3
- If we have a function like this: · 0862e342
  Bill Wendling authored Jan 26, 2008
  0862e342
- Fix some bugs in SimplifyNodeWithTwoResults where it would call deletenode to · 31e9edce
  Chris Lattner authored Jan 26, 2008
  31e9edce
- don't bother making x&-1 only to simplify it in dag combine. This commonly... · 720d8999
  Chris Lattner authored Jan 26, 2008
  720d8999
- reduce indentation · cb3cf546
  Chris Lattner authored Jan 25, 2008
  cb3cf546
Jan 25, 2008
- Do this more neatly. · f52faf9a
  Duncan Sands authored Jan 25, 2008
  f52faf9a
- fix long lines. · fc80996a
  Chris Lattner authored Jan 25, 2008
  fc80996a
- JITEmitter.cpp was trying to sync the icache for function stubs, but · 919ad97c
  Chris Lattner authored Jan 25, 2008
  919ad97c
- DeadStoreElimination can treat byval parameters as if there were alloca's for... · 6af19fd1
  Owen Anderson authored Jan 25, 2008
  6af19fd1
- Add skeletal code to increase the alignment of loads and stores when · 2d7a830f
  Chris Lattner authored Jan 25, 2008
  2d7a830f
- move MachineFrameInfo::CreateFixedObject out of line, give MachineFrameInfo · 6068832d
  Chris Lattner authored Jan 25, 2008
  6068832d
- include alignment and volatility information in -view-*-dags output · da52d9e0
  Chris Lattner authored Jan 25, 2008
  da52d9e0
- optimize fxor like for · f4523c35
  Chris Lattner authored Jan 25, 2008
  f4523c35
- Add target-specific dag combines for FAND(x,0) and FOR(x,0). This allows · 84ab724e
  Chris Lattner authored Jan 25, 2008
  84ab724e
- Provide correct DWARF register numbering for debug information emission on x86-32/Darwin. · fcde6168
  Anton Korobeynikov authored Jan 25, 2008
  fcde6168
Jan 24, 2008
- Don't dump the function! · 8d83271b
  Chris Lattner authored Jan 24, 2008
  8d83271b
- getUnderlyingObject can return null, handle this. · 23dd0551
  Chris Lattner authored Jan 24, 2008
  23dd0551
- Teach basicaa that 'byval' arguments define a new memory location that · 9104d712
  Chris Lattner authored Jan 24, 2008
  9104d712
- Add hasByValAttr() and hasNoAliasAttr() methods to the Argument class. · e30f09d0
  Chris Lattner authored Jan 24, 2008
  e30f09d0
- clarify a comment, thanks Duncan. · 34ed27c4
  Chris Lattner authored Jan 24, 2008
  34ed27c4
- Significantly simplify and improve handling of FP function results on x86-32. · a91f77ea
  Chris Lattner authored Jan 24, 2008
  a91f77ea
- Fix this buggy transformation. Two observations: · e97fa8cd
  Chris Lattner authored Jan 24, 2008
  e97fa8cd
- The dag combiner is missing revisiting nodes that it really should, and thus leaving · d66eac62
  Chris Lattner authored Jan 24, 2008
  d66eac62
- fold fp_round(fp_round(x)) -> fp_round(x). · 0feb1b0f
  Chris Lattner authored Jan 24, 2008
  0feb1b0f
- Fix potential buffer overflow · a4f27608
  Anton Korobeynikov authored Jan 24, 2008
  a4f27608
- Move some functionality for adding flags to MachineInstr's into methods on... · 2a8a4856
  Owen Anderson authored Jan 24, 2008
  2a8a4856
- Forgot these. · ec3da554
  Evan Cheng authored Jan 24, 2008
  ec3da554
- Let each target decide byval alignment. For X86, it's 4-byte unless the... · 35abd840
  Evan Cheng authored Jan 23, 2008
  35abd840
Jan 23, 2008
- The last pieces needed for loading arbitrary · 95d46ef8
  Duncan Sands authored Jan 23, 2008
  95d46ef8
- Honor explicit section information on Darwin. · 7f1ff5fe
  Dale Johannesen authored Jan 23, 2008
  7f1ff5fe
- Fix an iterator invalidation issue. · 7fe0bb2b
  Owen Anderson authored Jan 22, 2008
  7fe0bb2b
- SSE varargs arguments are passed in memory. · 1e0d4d2a
  Evan Cheng authored Jan 22, 2008
  1e0d4d2a
Jan 22, 2008
- Simplify SelectionDAG::getNode so that a big switch stmt is not #ifndef · 1671361c
  Chris Lattner authored Jan 22, 2008
  1671361c
- The final piece needed for storing arbitrary precision · 88de26cf
  Duncan Sands authored Jan 22, 2008
  88de26cf
- Enable the fix I just checked in, silly me. · f0692641
  Nick Lewycky authored Jan 22, 2008
  f0692641
- Multiply can be evaluated in a different type, so long as the target type has · 78712e5b
  Nick Lewycky authored Jan 22, 2008
  78712e5b
- Trivial patch to fix two warnings, please pull into llvm 2.2 · 1dea406e
  Chris Lattner authored Jan 22, 2008
  1dea406e
Jan 21, 2008
- Clarify a deviation from the original algorithm. · 7fb62417
  Owen Anderson authored Jan 21, 2008
  7fb62417
- Improve a few comments. · d990b4f6
  Owen Anderson authored Jan 21, 2008
  d990b4f6