Have the SjLjEHPrepare pass do some more heavy lifting.
[oota-llvm.git] / lib / CodeGen /
2011-09-28 Bill WendlingHave the SjLjEHPrepare pass do some more heavy lifting.
2011-09-28 Duncan SandsA typeid of zero means a cleanup, not a catch. This...
2011-09-28 Bill WendlingStrip off pointer casts when looking at the eh.sjlj...
2011-09-28 Bill WendlingBitcast the alloca to an i8* to match the intrinsic...
2011-09-28 Bill WendlingCreate and use an llvm.eh.sjlj.functioncontext intrinsic.
2011-09-28 Bill WendlingIn the new EH model, setup the function context and...
2011-09-28 Bill WendlingDon't conditionalize execution of the SjLj EH prepare...
2011-09-28 Jakob Stoklund OlesenRename class and clean up source.
2011-09-28 Jakob Stoklund OlesenRename SSEDomainFix -> lib/CodeGen/ExecutionDepsFix.
2011-09-27 Bill WendlingThis is the start of the new SjLj EH preparation pass...
2011-09-27 Cameron ZwarichRemove an invalid assert that is really just asserting...
2011-09-27 Jim GrosbachRename AddSelectionDAGCSEId() to addSelectionDAGCSEId().
2011-09-27 Nadav RotemCleanup PromoteIntOp_EXTRACT_VECTOR_ELT and PromoteIntR...
2011-09-27 Nadav RotemRevert r140463; The patch assumes that <4 x i1> is...
2011-09-26 James MolloyFix emission of debug data for global variables. getCon...
2011-09-25 Jakob Stoklund OlesenAdd target hook for pseudo instruction expansion.
2011-09-25 Nadav Rotem[vector-select] Address one of the issues in pr10902...
2011-09-25 Jakob Stoklund OlesenClean up code after renaming LowerSubregs -> ExpandPost...
2011-09-25 Jakob Stoklund OlesenRename LowerSubregs to ExpandPostRAPseudos.
2011-09-24 Nadav RotemImplement Duncan's suggestion to use the result of...
2011-09-24 Nadav Rotem[Vector-Select] Address one of the problems in 10902.
2011-09-23 Jakob Stoklund OlesenVerify that terminators follow non-terminators.
2011-09-23 Eli FriedmanPR10998: It is not legal to sink an instruction past...
2011-09-23 Duncan SandsTweak the handling of MERGE_VALUES nodes: remove the...
2011-09-23 Nadav RotemVector-Select: Address one of the problems in pr10902...
2011-09-22 Dan GohmanFix SimplifySelectCC to add newly created nodes to...
2011-09-22 Jakob Stoklund OlesenConstrain register classes instead of emitting copies.
2011-09-22 Jakob Stoklund OlesenAdd a MinNumRegs argument to MRI::constrainRegClass().
2011-09-22 Bill WendlingUse the C personality function instead of the C++ perso...
2011-09-21 Devang PatelDo not unnecessarily use AT_specification DIE because...
2011-09-21 Bill WendlingAttempt to update the shadow stack GC pass to the new...
2011-09-21 Jim GrosbachTidy up. Whitepsace.
2011-09-21 Nadav Rotem[VECTOR-SELECT] Address one of the bugs in pr10902.
2011-09-21 Andrew TrickLower ARM adds/subs to add/sub after adding optional...
2011-09-20 Bruno Cardoso LopesAdd a DAGCombine for subvector extracts to remove usele...
2011-09-20 Andrew TrickRestore hasPostISelHook tblgen flag.
2011-09-20 Andrew TrickARM isel bug fix for adds/subs operands.
2011-09-20 Andrew Trickwhitespace
2011-09-18 Nadav Rotemwhite space cleanups
2011-09-16 Benjamin KramerNamespacify.
2011-09-16 Jakob Stoklund OlesenSpill mode: Hoist back-copies locally.
2011-09-16 Jakob Stoklund OlesenDisable local spill hoisting for non-killing copies.
2011-09-15 Eli FriedmanSome legalization fixes for atomic load and store.
2011-09-15 Jakob Stoklund OlesenAdd an option to disable spill hoisting.
2011-09-15 Jakob Stoklund OlesenVirtRegMap is counting spill slots, not register spills.
2011-09-15 Jakob Stoklund OlesenCount correctly when a COPY turns into a spill or reload.
2011-09-15 Jakob Stoklund OlesenCount inserted spills and reloads more accurately.
2011-09-15 Jakob Stoklund OlesenTrace through sibling PHIs in bulk.
2011-09-15 Jakob Stoklund OlesenSpeed up LiveIntervals::shrinkToUse with some caching.
2011-09-15 Jakob Stoklund OlesenRevert r139782, "RemoveCopyByCommutingDef doesn't need...
2011-09-15 Jakob Stoklund OlesenStop verifying hasPHIKill() flags.
2011-09-15 Jakob Stoklund OlesenRemoveCopyByCommutingDef doesn't need hasPHIKill().
2011-09-15 Jakob Stoklund OlesenIt is safe to remat a value killed by phis.
2011-09-15 Jakob Stoklund OlesenLeave hasPHIKill flags alone in LiveInterval::RenumberV...
2011-09-15 Andrew Trick[regcoalescing] bug fix for RegistersDefinedFromSameValue.
2011-09-14 Devang PatelAdd support to emit debug info for C++0x nullptr type.
2011-09-14 Jakob Stoklund OlesenIgnore the cloning of unknown registers.
2011-09-14 Jakob Stoklund OlesenHoist back-copies to the least busy dominator.
2011-09-14 Nadav RotemAdd integer promotion support for vselect
2011-09-13 Jakob Stoklund OlesenDistinguish complex mapped values from forced recomputa...
2011-09-13 Jakob Stoklund OlesenImplement -split-spill-mode=size.
2011-09-13 Eli FriedmanFix check for unaligned load/store so it doesn't catch...
2011-09-13 Eli FriedmanError out on CodeGen of unaligned load/store. Fix...
2011-09-13 Nadav RotemFix the assertion which checks the size of the input...
2011-09-13 Nadav RotemAdd vselect target support for targets that do not...
2011-09-13 Devang PatelUse a cache to maintain list of machine basic blocks...
2011-09-13 Jakob Stoklund OlesenAdd SplitEditor::markOverlappedComplement().
2011-09-13 Jakob Stoklund OlesenEliminate the extendRange() wrapper.
2011-09-13 Jakob Stoklund OlesenSwitch extendInBlock() to take a kill slot instead...
2011-09-13 Jakob Stoklund OlesenUse a separate LiveRangeCalc for the complement in...
2011-09-13 NAKAMURA TakumiUnbreak msvc.
2011-09-13 Jakob Stoklund OlesenExtract live range calculations from SplitKit.
2011-09-12 Bill WendlingIntroduce a bit of a hack.
2011-09-12 Jakob Stoklund OlesenRemove the -compact-regions flag.
2011-09-12 Jakob Stoklund OlesenAdd an interface for SplitKit complement spill modes.
2011-09-12 Jakob Stoklund OlesenUpdate comments to reflect some (not so) recent changes.
2011-09-10 Richard TrieuFix asserts in CodeGen from:
2011-09-09 Chris Lattnertidy up a bit
2011-09-09 Eli FriedmanMake the SelectionDAG verify that all the operands...
2011-09-09 Jakob Stoklund OlesenReapply r139247: Cache intermediate results during...
2011-09-08 Devang PatelDirectly point debug info to the stack slot of the...
2011-09-07 Jakob Stoklund OlesenRevert r139247 "Cache intermediate results during trace...
2011-09-07 Jakob Stoklund OlesenCache intermediate results during traceSiblingValue.
2011-09-07 James MolloyRefactor instprinter and mcdisassembler to take a Subta...
2011-09-07 Eli FriedmanRelax the MemOperands on atomics a bit. Fixes -verify...
2011-09-07 Devang PatelWhile sinking machine instructions, sink matching DBG_V...
2011-09-06 Duncan SandsAdd codegen support for vector select (in the IR this...
2011-09-06 Duncan SandsSplit the init.trampoline intrinsic, which currently...
2011-09-03 Owen AndersonFix a truly heinous bug in DAGCombine related to Assert...
2011-09-02 Jakob Stoklund OlesenSimplify by using isFullCopy().
2011-09-02 Duncan SandsDarwin wants ctors/dtors to be ordered the other way...
2011-09-01 Dan GohmanRevert r131152, r129796, r129761. This code is currentl...
2011-09-01 Benjamin KramerDon't drop alignment info on local common symbols.
2011-09-01 Jakob Stoklund OlesenPermit remat of partial register defs when it is safe.
2011-09-01 Jakob Stoklund OlesenRevert r138794, "Do not try to rematerialize a value...
2011-09-01 Jakob Stoklund OlesenPrevent remat of partial register redefinitions.
2011-09-01 Evan ChengTeach MachineLICM reg pressure tracking code to deal...
2011-09-01 Andrew TrickPreRA scheduler should avoid cloning compares.
2011-08-31 David GreeneFix Size Typing
2011-08-31 Eli FriedmanMisc cleanup; addresses Duncan's comments on r138877.
next