aboutsummaryrefslogtreecommitdiff
path: root/lib/CodeGen/SelectionDAG
AgeCommit message (Expand)Author
2008-03-10Fix mul expansion to check the correct number of bits forDan Gohman
2008-03-10Somewhat better solution.Evan Cheng
2008-03-10Default ISD::PREFETCH to expand.Evan Cheng
2008-03-10Revert 48125, 48126, and 48130 for now to unbreak some x86-64 tests.Evan Cheng
2008-03-10Give TargetLowering::getSetCCResultType() a parameter so that ISD::SETCC'sScott Michel
2008-03-10DohEvan Cheng
2008-03-10Avoid creating BUILD_VECTOR of all zero elements of "non-normalized" type (e....Evan Cheng
2008-03-10Allow insert_subreg into implicit, target-specific values. Christopher Lamb
2008-03-10Increase ISD::ParamFlags to 64 bits. Increase the ByValSizeDale Johannesen
2008-03-09remove an extraneous (and ugly) default argument, thanks Duncan.Chris Lattner
2008-03-09fp_round's produced by getCopyFromParts should always be exact, becauseChris Lattner
2008-03-09teach X86InstrInfo::copyRegToReg how to copy into ST(0) from Chris Lattner
2008-03-09Add ScheduleDAG support for copytoreg where the src/dst register areChris Lattner
2008-03-09fix 80 col violationChris Lattner
2008-03-09extend fp values with FP_EXTEND not FP_ROUND.Chris Lattner
2008-03-09Fix two problems in SelectionDAGLegalize::ExpandBUILD_VECTOR's handlingChris Lattner
2008-03-08Teach SD some vector identities, allowing us to compile vec_set-9 into:Chris Lattner
2008-03-08Implement x86 support for @llvm.prefetch. It corresponds to prefetcht{0|1|2} ...Evan Cheng
2008-03-0680 col violation.Evan Cheng
2008-03-06Constant fold SIGN_EXTEND_INREG with ashr not lshr.Evan Cheng
2008-03-05Clarify that CALLSEQ_START..END may not be nested,Dale Johannesen
2008-03-05Generalize FP constant shrinking optimization to apply to any vtChris Lattner
2008-03-05Improve comment, pass in the original VT so that we can shrink a long double ...Chris Lattner
2008-03-05Codegen support for i128 UINT_TO_FP. This just fixes aDan Gohman
2008-03-05Add a target lowering hook to control whether it's worthwhile to compress fp ...Evan Cheng
2008-03-0564bit CAS on 32bit x86.Andrew Lenharth
2008-03-05Codegen support for i128 SINT_TO_FP.Dan Gohman
2008-03-04Some improvements related to the computation of heights, depths of SUnits.Roman Levenstein
2008-03-04Refactor ExpandConstantFP so it can optimize load from constpool of types lar...Evan Cheng
2008-03-04Rename isOperand() to isOperandOf() (and other similar methods). It always co...Evan Cheng
2008-03-03Misc. APInt-ification in the DAGCombiner.Dan Gohman
2008-03-03More APInt-ification.Dan Gohman
2008-03-03Yet more APInt-ification.Dan Gohman
2008-03-03More APInt-ification.Dan Gohman
2008-03-03More APInt-ification.Dan Gohman
2008-03-01all but CAS working on x86Andrew Lenharth
2008-03-01Add MVT::is128BitVector and is64BitVector. ShrinkDale Johannesen
2008-03-01Refactor / clean up code; remove td list scheduler special tie breaker (no re...Evan Cheng
2008-02-29More APInt-ification.Dan Gohman
2008-02-29Use the new convertFromAPInt instead of convertFromZeroExtendedInteger,Dan Gohman
2008-02-29Use the new APInt-enabled form of getConstant instead of convertingDan Gohman
2008-02-28Interface of getByValTypeAlignment differed betweenDale Johannesen
2008-02-28Fix an assertion message.Dale Johannesen
2008-02-28Keep track how many commutes are performed by the scheduler.Evan Cheng
2008-02-28implement expand for ISD::DECLARE by just deleting it.Chris Lattner
2008-02-28Add a quick and dirty "loop aligner pass". x86 uses it to align its loops to ...Evan Cheng
2008-02-27Handle load/store of misaligned vectors that are the Dale Johannesen
2008-02-27Remove the `else', at Evan's insistence.Dan Gohman
2008-02-27Add a FIXME about the VECTOR_SHUFFLE evil hack.Duncan Sands
2008-02-27LegalizeTypes support for EXTRACT_VECTOR_ELT. TheDuncan Sands