aboutsummaryrefslogtreecommitdiff
path: root/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp
AgeCommit message (Expand)Author
2007-10-18Pointer arithmetic should be done with the index the same size as the pointer.Bill Wendling
2007-10-17Add support for ISD::SELECT in SplitVectorOp.Dan Gohman
2007-10-17Return Expand from getOperationAction for all extendedDuncan Sands
2007-10-16Fixes due to lack of type-safety for ValueType: (1) ValueTypeDuncan Sands
2007-10-15Move CreateStackTemporary out to SelectionDAGChris Lattner
2007-10-15One xform performed by LegalizeDAG is transformation of "store of fp" to "sto...Chris Lattner
2007-10-14remove misleading comment.Chris Lattner
2007-10-14If a target doesn't have HasMULHU or HasUMUL_LOHI, ExpandOp would returnChris Lattner
2007-10-13Add a simple optimization to simplify the input toChris Lattner
2007-10-12ppc long double. Implement fabs and fneg.Dale Johannesen
2007-10-12Implement i64->ppcf128 conversions.Dale Johannesen
2007-10-12Add support to SplitVectorOp for powi, where the second operandDan Gohman
2007-10-12PPC long double. Implement a couple more conversions.Dale Johannesen
2007-10-11Codegen support for vector intrinsics.Dan Gohman
2007-10-11Implement ppc long double->uint conversion.Dale Johannesen
2007-10-11Next PPC long double bits. First cut at constants.Dale Johannesen
2007-10-10Next PPC long double bits: ppcf128->i32 conversion.Dale Johannesen
2007-10-08Migrate X86 and ARM from using X86ISD::{,I}DIV and ARMISD::MULHILO{U,S} toDan Gohman
2007-10-07convertFromInteger, as originally written, expected sign-extendedNeil Booth
2007-10-06Next powerpc long double bits. Comparisons work,Dale Johannesen
2007-10-05First round of ppc long double. call/return andDale Johannesen
2007-10-05Legalize support for MUL_LOHI and DIVREM.Dan Gohman
2007-10-05Silence a warning.Evan Cheng
2007-09-30Constant fold int-to-long-double conversions;Dale Johannesen
2007-09-28Teach SplitVectorOp how to split INSERT_VECTOR_ELT.Dan Gohman
2007-09-28Fix long double -> uint64 conversion.Dale Johannesen
2007-09-28Add sqrt and powi intrinsics for long double.Dale Johannesen
2007-09-27Silence a compiler warning.Evan Cheng
2007-09-26Fix f80 UNDEF.Dale Johannesen
2007-09-24Use the correct result value type instead of using getValueType(0)Dan Gohman
2007-09-21Change APFloat::convertFromInteger to take the incomingDale Johannesen
2007-09-19More long double fixes. x86_64 should build now.Dale Johannesen
2007-09-19Fix longdouble -> uint conversion.Dale Johannesen
2007-09-16Adjust per revew comments.Dale Johannesen
2007-09-14Remove the assumption that FP's are either float orDale Johannesen
2007-09-12Revise previous patch per review comments.Dale Johannesen
2007-09-11Add APInt interfaces to APFloat (allows directlyDale Johannesen
2007-09-11Fold the adjust_trampoline intrinsic intoDuncan Sands
2007-09-08Implement misaligned FP loads and stores.Dale Johannesen
2007-08-30Change LegalFPImmediates to use APFloat.Dale Johannesen
2007-08-29Fix use of declaration inside case blockAnton Korobeynikov
2007-08-29Lower FRAME_TO_ADDR_OFFSET to zero by default (if not custom lowered)Anton Korobeynikov
2007-08-25Allow target constants to be illegal types. The target shouldChris Lattner
2007-08-17Avoid issue on 64-bit hosts.Evan Cheng
2007-08-16If dynamic_stackalloc alignment is > stack alignment, first issue an instruct...Evan Cheng
2007-08-15Implement FPOWI ExpandOp.Lauro Ramos Venancio
2007-08-09Update per review comments.Dale Johannesen
2007-08-09long double 9 of N. This finishes up the X86-32 bitsDale Johannesen
2007-08-08If a target really needs to custom lower constants, it should be allowedScott Michel
2007-08-02Style police: Expand the tabs to spaces!Scott Michel
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-12-29 13:04:01 +0000