Now that we have int/fp lattice values, implement the SDTCisOpSmallerThanOp
[oota-llvm.git] / utils / TableGen / DAGISelEmitter.cpp
2005-10-14 Chris LattnerNow that we have int/fp lattice values, implement the...
2005-10-14 Chris LattnerFairly serious rework of the typing code to add new...
2005-10-14 Chris Lattnersimplify the code a bit
2005-10-14 Chris LattnerAdd basic support for recognizing a new SDTCisOpSmaller...
2005-10-14 Chris LattnerImplement a couple of new (important) features.
2005-09-29 Chris LattnerTeach tablegen to reassociate operators when possible...
2005-09-29 Chris LattnerTeach tblgen to build permutations of instructions...
2005-09-28 Chris Lattneradd support for an associative marker
2005-09-28 Chris LattnerEmit an error if instructions or patterns are defined...
2005-09-28 Chris Lattnercollect commutativity information
2005-09-28 Chris LattnerPrefer cheaper patterns to more expensive ones. Print...
2005-09-28 Chris LattnerSelect Constant nodes to TargetConstant nodes
2005-09-26 Chris Lattnermemoize the assert results
2005-09-26 Chris LattnerEmit the switch stmt cases in alphabetical order instea...
2005-09-26 Chris Lattnerimplement a fixme: only select values once, even if...
2005-09-25 Jeff CohenFix VC++ build errors.
2005-09-24 Chris Lattnermemoize translations
2005-09-24 Chris LattnerTeach the DAG isel generator to emit code that creates...
2005-09-23 Chris LattnerEmit better code (no more copies for var references...
2005-09-23 Chris LattnerFix a fixme by passing around SDOperand's instead of...
2005-09-23 Chris LattnerEmit code that matches the incoming DAG pattern and...
2005-09-23 Chris Lattneremit information about the order patterns are to be...
2005-09-23 Chris Lattnerstart filling in the switch stmt
2005-09-16 Chris LattnerFix a minor bug, add comments
2005-09-15 Chris Lattnerteach the type inference code how to infer types for...
2005-09-15 Chris Lattnerput instructions into a map instead of a vector for...
2005-09-15 Chris Lattnerwhen parsing instructions remember information about...
2005-09-15 Chris LattnerStart parsing "Pattern" nodes
2005-09-15 Chris Lattnerrename a couple of methods, add structure for pattern...
2005-09-14 Chris LattnerVerify that xform functions only occur in logical places
2005-09-14 Chris LattnerPromote xform fns to be explicit nodes in result patter...
2005-09-14 Chris Lattnerstart building the instruction dest pattern correctly...
2005-09-14 Chris Lattnercatch unnamed inputs
2005-09-14 Chris Lattnercheck that there are no unexpected operands
2005-09-14 Chris Lattnerforce all instruction operands to be named.
2005-09-14 Chris Lattnerfix a broke range check
2005-09-14 Chris LattnerParse significantly more of the instruction pattern...
2005-09-14 Chris LattnerVerify that set destinations occur first in the instruc...
2005-09-14 Chris Lattnerremove some code that isn't ready for prime time
2005-09-14 Chris LattnerSwitch to a slightly more structured representation...
2005-09-14 Chris LattnerAdd some more checking/verification code
2005-09-14 Chris Lattnerstart parsing instructions into patterns, start doing...
2005-09-13 Chris Lattnerdon't emit the namespace inside the class!
2005-09-13 Chris LattnerEmit code suitable for emission into the ISel class...
2005-09-13 Chris Lattnercontinue xform function parsing
2005-09-13 Chris LattnerStart parsing node transformation information
2005-09-13 Chris Lattnercompletely eliminate TreePattern::PatternType
2005-09-09 Chris Lattneradd an accessor to provide more checking
2005-09-09 Chris Lattneruse new accessors to simplify code. Add checking to...
2005-09-08 Chris LattnerFix incorrect comment
2005-09-08 Chris LattnerImplement a complete type inference system for dag...
2005-09-08 Chris LattnerParse information about type constraints on SDNodes
2005-09-08 Chris Lattneruse node info in the one place we currently use it
2005-09-08 Chris Lattnerstart parsing SDNode info records
2005-09-08 Chris LattnerKeep names even when inlining. This allows us to reali...
2005-09-07 Chris LattnerInitial cut of the dag isel generator. This is still...