2010-01-27 |
Bob Wilson | Avoid creating redundant PHIs in SSAUpdater::GetValueIn... |
tree | commitdiff |
2010-01-27 |
Jeffrey Yasskin | Kill ModuleProvider and ghost linkage by inverting... |
tree | commitdiff |
2010-01-27 |
Benjamin Kramer | Don't bother with sprintf, just pass the Twine through. |
tree | commitdiff |
2010-01-27 |
Benjamin Kramer | Use the less expensive getName function instead of... |
tree | commitdiff |
2010-01-27 |
Chris Lattner | some cleanups. |
tree | commitdiff |
2010-01-27 |
Chris Lattner | no need to check for null |
tree | commitdiff |
2010-01-27 |
Victor Hernandez | When converting dbg.declare to dbg.value, attach promot... |
tree | commitdiff |
2010-01-26 |
Victor Hernandez | Avoid extra calls to MD->getNumOperands() |
tree | commitdiff |
2010-01-26 |
Victor Hernandez | Switch AllocaDbgDeclares to SmallVector and don't leak... |
tree | commitdiff |
2010-01-26 |
Victor Hernandez | In mem2reg, for all alloca/stores that get promoted... |
tree | commitdiff |
2010-01-25 |
Bob Wilson | Remove check for an impossible condition: the condition... |
tree | commitdiff |
2010-01-25 |
Bob Wilson | Change Value::getUnderlyingObject to have the MaxLookup... |
tree | commitdiff |
2010-01-25 |
Victor Hernandez | Revert r94260 until findDbgDeclare() is made more efficient |
tree | commitdiff |
2010-01-24 |
Chris Lattner | make -fno-rtti the default unless a directory builds... |
tree | commitdiff |
2010-01-24 |
Chris Lattner | change the canonical form of "cond ? -1 : 0" to be |
tree | commitdiff |
2010-01-23 |
Chris Lattner | fix a potential overflow issue Eli pointed out. |
tree | commitdiff |
2010-01-23 |
Nick Lewycky | Speculatively revert r94322 to see if it fixes darwin... |
tree | commitdiff |
2010-01-23 |
Chris Lattner | third bug from PR6119: the xor dupe extension allows |
tree | commitdiff |
2010-01-23 |
Nick Lewycky | Teach DAE that even though it can't modify the function... |
tree | commitdiff |
2010-01-23 |
Chris Lattner | add an early out to ProcessBranchOnXOR to speed it up, |
tree | commitdiff |
2010-01-23 |
Chris Lattner | fix a crash in jump threading, PR6119 |
tree | commitdiff |
2010-01-23 |
Chris Lattner | implement a simple instcombine xform that has been... |
tree | commitdiff |
2010-01-23 |
Eric Christopher | Reapply 94059 while fixing the calling convention setup |
tree | commitdiff |
2010-01-23 |
Victor Hernandez | In mem2reg, for all alloca/stores that get promoted... |
tree | commitdiff |
2010-01-22 |
Benjamin Kramer | Another strncmp -> StringRef.startswith simplification. |
tree | commitdiff |
2010-01-22 |
Bob Wilson | Revert 94059. It is breaking the MultiSource/Benchmark... |
tree | commitdiff |
2010-01-22 |
Victor Hernandez | Keep ignoring pointer-to-pointer bitcasts |
tree | commitdiff |
2010-01-22 |
Chris Lattner | Stop building RTTI information for *most* llvm librarie... |
tree | commitdiff |
2010-01-22 |
Dan Gohman | Revert LoopStrengthReduce.cpp to pre-r94061 for now. |
tree | commitdiff |
2010-01-21 |
Victor Hernandez | No need to look through bitcasts for DbgInfoIntrinsic |
tree | commitdiff |
2010-01-21 |
Victor Hernandez | DbgInfoIntrinsic no longer appear in an instruction... |
tree | commitdiff |
2010-01-21 |
Victor Hernandez | No need to look through bitcasts for DbgInfoIntrinsic |
tree | commitdiff |
2010-01-21 |
Victor Hernandez | DbgInfoIntrinsics no longer appear in an instruction... |
tree | commitdiff |
2010-01-21 |
Dan Gohman | When inserting expressions for post-increment users... |
tree | commitdiff |
2010-01-21 |
Dan Gohman | Include IVUsers information in LSR's debug output. |
tree | commitdiff |
2010-01-21 |
Dan Gohman | Prune the search for candidate formulae if the number... |
tree | commitdiff |
2010-01-21 |
Dan Gohman | Add a comment. |
tree | commitdiff |
2010-01-21 |
Chris Lattner | It turns out that this #include is needed because otherwise |
tree | commitdiff |
2010-01-21 |
Chris Lattner | unbreak the build, apparently without this transformuti... |
tree | commitdiff |
2010-01-21 |
Chris Lattner | tidy up |
tree | commitdiff |
2010-01-21 |
Victor Hernandez | Don't need to include IntrinsicInst.h any more |
tree | commitdiff |
2010-01-21 |
Victor Hernandez | No need to map NULL operands of metadata |
tree | commitdiff |
2010-01-21 |
Dan Gohman | Re-implement the main strength-reduction portion of... |
tree | commitdiff |
2010-01-21 |
Eric Christopher | Add strcpy_chk -> strcpy support for "don't know" objec... |
tree | commitdiff |
2010-01-20 |
Chris Lattner | simplify this code. |
tree | commitdiff |
2010-01-20 |
Jakob Stoklund Olesen | Move per-function inline threshold calculation to a... |
tree | commitdiff |
2010-01-20 |
Victor Hernandez | Switch Elts from vector to SmallVector |
tree | commitdiff |
2010-01-20 |
Victor Hernandez | Map operands of all function-local metadata, not just... |
tree | commitdiff |
2010-01-19 |
Dan Gohman | When doing address-mode sinking, expand the base regist... |
tree | commitdiff |
2010-01-19 |
Chris Lattner | optimize ~(~X >>s Y) --> (X >>s Y), patch by Edmund... |
tree | commitdiff |
2010-01-19 |
Bob Wilson | Fix a crash in scalarrepl for memcpy/memmove where... |
tree | commitdiff |
2010-01-19 |
Eric Christopher | Fix comment. |
tree | commitdiff |
2010-01-18 |
Chris Lattner | my instcombine transformations to make extension elimin... |
tree | commitdiff |
2010-01-18 |
Devang Patel | While mapping llvm.dbg.declare intrinsic manually map... |
tree | commitdiff |
2010-01-17 |
Owen Anderson | Convert some of the dynamic opcode lookups into static... |
tree | commitdiff |
2010-01-17 |
Owen Anderson | Fix comment. |
tree | commitdiff |
2010-01-15 |
Bob Wilson | Fix a comment typo. |
tree | commitdiff |
2010-01-13 |
Bill Wendling | When the visitSub method was split into visitSub and... |
tree | commitdiff |
2010-01-12 |
Chris Lattner | 1) Use the new SimplifyInstructionsInBlock routine... |
tree | commitdiff |
2010-01-12 |
Chris Lattner | add a helper function. |
tree | commitdiff |
2010-01-12 |
Chris Lattner | tidy up |
tree | commitdiff |
2010-01-12 |
Chris Lattner | Teach jump threading to duplicate small blocks when... |
tree | commitdiff |
2010-01-11 |
Chris Lattner | some cleanup, and make it obvious that ProcessJumpOnPHI... |
tree | commitdiff |
2010-01-11 |
Chris Lattner | reenable the piece that turns trunc(zext(x)) -> x even... |
tree | commitdiff |
2010-01-11 |
Chris Lattner | Disable folding sext(trunc(x)) -> x (and other similar... |
tree | commitdiff |
2010-01-11 |
Chris Lattner | add one more bitfield optimization, allowing clang... |
tree | commitdiff |
2010-01-11 |
Chris Lattner | Extend CanEvaluateZExtd to handle and/or/xor more aggre... |
tree | commitdiff |
2010-01-11 |
Chris Lattner | Remove the dead TD argument to CanEvaluateZExtd, and... |
tree | commitdiff |
2010-01-11 |
Chris Lattner | improve comments, remove dead TD argument to CanEvaluat... |
tree | commitdiff |
2010-01-10 |
Chris Lattner | teach sext optimization to handle truncs from types... |
tree | commitdiff |
2010-01-10 |
Chris Lattner | teach zext optimization how to deal with truncs that... |
tree | commitdiff |
2010-01-10 |
Chris Lattner | simplify CanEvaluateSExtd to return a bool now that... |
tree | commitdiff |
2010-01-10 |
Chris Lattner | the NumCastsRemoved argument to CanEvaluateSExtd is... |
tree | commitdiff |
2010-01-10 |
Chris Lattner | now that the cost model has changed, we can always... |
tree | commitdiff |
2010-01-10 |
Chris Lattner | change the preferred canonical form for a sign extensio... |
tree | commitdiff |
2010-01-10 |
Chris Lattner | fix indentation of switch statements, no functionality... |
tree | commitdiff |
2010-01-10 |
Chris Lattner | fix pasto that broke bootstrap. |
tree | commitdiff |
2010-01-10 |
Chris Lattner | simplify CanEvaluateZExtd now that we don't care about... |
tree | commitdiff |
2010-01-10 |
Chris Lattner | two changes: |
tree | commitdiff |
2010-01-10 |
Chris Lattner | enhance CanEvaluateZExtd to handle shift left and sext... |
tree | commitdiff |
2010-01-10 |
Chris Lattner | remove an xform subsumed by EvaluateInDifferentType. |
tree | commitdiff |
2010-01-10 |
Julien Lerouge | Fix nondeterministic behavior. |
tree | commitdiff |
2010-01-10 |
Chris Lattner | clean up this xform by using m_Trunc. |
tree | commitdiff |
2010-01-10 |
Chris Lattner | inline and remove the rest of commonIntCastTransforms. |
tree | commitdiff |
2010-01-10 |
Chris Lattner | Inline the expression type promotion/demotion stuff... |
tree | commitdiff |
2010-01-09 |
Chris Lattner | only factor from expressions whose uses are empty and... |
tree | commitdiff |
2010-01-09 |
Julien Lerouge | Fix nondeterministic behavior. |
tree | commitdiff |
2010-01-08 |
Eric Christopher | Remove unnecessary dyn_cast and add a comment. Part... |
tree | commitdiff |
2010-01-08 |
Chris Lattner | mplement a theoretical fixme. |
tree | commitdiff |
2010-01-08 |
Chris Lattner | rename CanEvaluateInDifferentType -> CanEvaluateTruncat... |
tree | commitdiff |
2010-01-08 |
Chris Lattner | teach instcombine to delete sign extending shift pairs... |
tree | commitdiff |
2010-01-08 |
Duncan Sands | Suppress an unused variable warning when assertions... |
tree | commitdiff |
2010-01-08 |
Chris Lattner | tidy up some stuff duncan pointed out. |
tree | commitdiff |
2010-01-07 |
Chris Lattner | teach ComputeNumSignBits to look through PHI nodes. |
tree | commitdiff |
2010-01-07 |
Chris Lattner | Enhance instcombine to reason more strongly about promo... |
tree | commitdiff |
2010-01-07 |
Benjamin Kramer | Use a do-while loop instead of while + boolean. |
tree | commitdiff |
2010-01-07 |
Duncan Sands | Be less stingy as to how many selects and phi nodes we |
tree | commitdiff |
2010-01-07 |
Chris Lattner | handle ConstantVector while I'm in here. |
tree | commitdiff |
2010-01-07 |
Chris Lattner | fix a globalopt crash on 'bullet' (handling evaluation... |
tree | commitdiff |
2010-01-06 |
Eric Christopher | Move the object size intrinsic optimization to inst... |
tree | commitdiff |
next |