oota-llvm.git
2006-03-14 Chris LattnerFix the gcread/gcwrite intrinsic specifications, fixing
2006-03-14 Chris LattnerBugfix, unbreaking CodeGen/PowerPC/cttz.ll
2006-03-14 Chris LattnerFix an et-forest memory leak. Patch by Daniel Berlin.
2006-03-14 Chris Lattneradd a note
2006-03-14 Jim LaskeyReflect the fact that empty strings can be expressed...
2006-03-14 Jim Laskey1. Use null for serialized empty strings.
2006-03-14 Jim LaskeyBring debugging information up to date.
2006-03-14 Chris LattnerFix an off by one error that caused PPC LLC failures...
2006-03-14 Chris LattnerImplement a FIXME, recusively reassociating
2006-03-14 Chris Lattnernew testcase from a FIXME in the code
2006-03-14 Chris Lattnerextract some code into a method, no functionality change
2006-03-14 Chris Lattnertransformation implemented
2006-03-14 Chris LattnerPromote shifts by a constant to multiplies so that...
2006-03-14 Chris LattnerTest that we can reassociate (x<<1)+(y<<1) -> (X+Y...
2006-03-14 Reid SpencerAllow a sequence of digits at the end of the directory...
2006-03-14 Reid SpencerChange llvm.cs.uiuc.edu -> llvm.org
2006-03-14 Reid SpencerAdd a newline at the end to avoid gcc warnings.
2006-03-14 Reid SpencerConvert llvm.cs.uiuc.edu -> llvm.org
2006-03-14 Reid SpencerChange docs for llvm.cs.uiuc.edu -> llvm.org
2006-03-14 Reid SpencerChanges docs for llvm.cs.uiuc.edu -> llvm.org
2006-03-14 Jim Laskey1. Handle removal of all arguments for a morphed intrinsic.
2006-03-14 Jim LaskeyRemove the use of llvm.dbg.declare.
2006-03-13 Evan ChengPPC LSR pass should use target lowering hooks.
2006-03-13 Evan ChengAdded getTargetLowering() to TargetMachine. Refactored...
2006-03-13 Evan ChengUpdate
2006-03-13 Evan ChengAdd LSR hooks.
2006-03-13 Evan ChengAdded getTargetLowering() - returns DAG lowering info.
2006-03-13 Evan ChengAdded target lowering hooks which LSR consults to make...
2006-03-13 Chris LattnerHandle builtins that directly correspond to GCC builtins.
2006-03-13 Chris Lattneremit a mapping from LLVM intrinsic -> GCC builtins.
2006-03-13 Chris LattnerVerify that packed type operands have the right size...
2006-03-13 Chris LattnerAdd a first ppc altivec intrinsic. Add packed type...
2006-03-13 Chris LattnerFor functions that use vector registers, save VRSAVE...
2006-03-13 Evan ChengOptimizing to calloc() to malloc() + memset() can be...
2006-03-13 Chris Lattnermake sure dead token factor nodes are removed by the...
2006-03-13 Reid SpencerBring makefile back into compliance with standard by...
2006-03-13 Jim LaskeyHandle the removal of the debug chain.
2006-03-13 Chris Lattnerremove two implemented items
2006-03-13 Chris LattnerFold X+Y -> X|Y when safe. This implements:
2006-03-13 Chris Lattnernew testcase
2006-03-13 Chris LattnerI can't convince myself that this is safe, remove the...
2006-03-13 Chris Lattneradd a couple of missing folds
2006-03-13 Chris LattnerFor targets with FABS/FNEG support, lower copysign...
2006-03-13 Chris LattnerFix a couple of bugs that broke the alpha tester build
2006-03-13 Chris LattnerHandle cracked instructions in dispatch group formation.
2006-03-13 Chris LattnerMark instructions that are cracked by the PPC970 decode...
2006-03-12 Chris LattnerSeveral big changes:
2006-03-12 Chris LattnerDon't advance the hazard recognizer when there are...
2006-03-12 Chris LattnerChain operands aren't real uses: they don't require...
2006-03-12 Chris LattnerAs a pending queue data structure to keep track of...
2006-03-12 Chris LattnerMark llvm.stacksave as only reading memory, this fixes
2006-03-11 Chris Lattnerrename priorityqueue -> availablequeue. When a node...
2006-03-11 Chris LattnerMake CurrCycle a local var instead of an instance var
2006-03-11 Chris LattnerMove some methods around so that BU specific code is...
2006-03-11 Chris Lattnermerge preds/chainpreds -> preds set
2006-03-11 Chris Lattnerblr is a branch too
2006-03-11 Chris Lattneradd an example
2006-03-11 Chris Lattneradd a note
2006-03-11 Nate BegemanFix PR681 by using the standard Lengauer and Tarjan...
2006-03-11 Chris Lattnerfix pasto in generate assertion msg
2006-03-11 Evan ChengDoh!
2006-03-10 Evan ChengAdded a parameter to control whether Constant::getStrin...
2006-03-10 Chris Lattnerremove some dead code
2006-03-10 Chris Lattnerremove ShouldEmitDebugFunctions, a hack to support...
2006-03-10 Chris Lattnerscrape out bits of llvm-db
2006-03-10 Chris LattnerRemove UnixLocalInferiorProcess: debugging via the...
2006-03-10 Chris LattnerSimplify this testcase
2006-03-10 Chris Lattnerreorder these to make it work with static libraries
2006-03-10 Chris Lattnerteach the JIT to encode vector registers
2006-03-10 Chris LattnerFix another broken intrinsic.
2006-03-10 Chris Lattnerweak globals on darwin require an extra load, breaking...
2006-03-10 Chris LattnerFix incorrect definitions of these intrinsics, which...
2006-03-10 Chris LattnerMove simple-selector-specific types to the simple selector.
2006-03-10 Chris LattnerSimplify the interface to the schedulers, to not pass...
2006-03-10 Chris LattnerSimplify the interface to the schedulers, to not pass...
2006-03-10 Chris LattnerMove some simple-sched-specific instance vars to the...
2006-03-10 Chris Lattnerprune #includes
2006-03-10 Chris Lattnermove some simple scheduler methods into the simple...
2006-03-10 Chris LattnerMake EmitNode take a SDNode instead of a NodeInfo*
2006-03-10 Chris LattnerMove the VRBase field from NodeInfo to being a separate...
2006-03-10 Chris LattnerStore VRBase in a map, not in NodeInfo.
2006-03-10 Chris Lattnerno need to build groups anymore
2006-03-10 Chris LattnerCreate SUnits directly from the SelectionDAG.
2006-03-10 Chris LattnerPush PrepareNodeInfo/IdentifyGroups down the inheritanc...
2006-03-10 Chris Lattnermake some methods protected instead of private
2006-03-10 Chris LattnerTeach the latency scheduler some new tricks. In partic...
2006-03-10 Jeff CohenPut intrinsics.gen in its proper place.
2006-03-10 Chris Lattneradd an aggregate method for reinserting scheduled nodes...
2006-03-10 Chris LattnerFix an incorrect intrinsic description
2006-03-10 Jeff CohenFix VC++ build breakage.
2006-03-09 Chris LattnerAdding an intrinsic is simpler still.
2006-03-09 Chris Lattneruse autogenerated side-effect information
2006-03-09 Chris Lattnergenerate side-effect info
2006-03-09 Chris LattnerSimpler still
2006-03-09 Chris LattnerUse autogenerated mod/ref info for intrinsics.
2006-03-09 Chris LattnerParse mod/ref properties, autogen mod/ref information
2006-03-09 Chris LattnerIntrinsic adding is a little bit simpler now
2006-03-09 Chris LattnerUse the autogenerated intrinsic verifier
2006-03-09 Chris Lattnerparse intrinsic types
2006-03-09 Evan ChengAdd option -enable-x86-lsr to enable x86 loop strength...
next