aboutsummaryrefslogtreecommitdiff
path: root/llvm/lib/CodeGen/SelectionDAG/LegalizeVectorOps.cpp
AgeCommit message (Expand)AuthorFilesLines
2015-01-08[SelectionDAG] Allow targets to specify legality of extloads' resultAhmed Bougacha1-1/+2
2014-10-21Add minnum / maxnum codegenMatt Arsenault1-0/+2
2014-08-27Teach the AArch64 backend about v4f16 and v8f16Oliver Stannard1-6/+17
2014-07-31Make sure no loads resulting from load->switch DAGCombine are marked invariantLouis Gerbarg1-3/+3
2014-07-24[x86] Make vector legalization of extloads work more like the "normal"Chandler Carruth1-5/+22
2014-07-24AA metadata refactoring (introduce AAMDNodes)Hal Finkel1-5/+5
2014-07-10[x86,SDAG] Introduce any- and sign-extend-vector-inreg nodes analogousChandler Carruth1-0/+66
2014-07-10Make it possible for ints/floats to return different values from getBooleanCo...Daniel Sanders1-3/+3
2014-07-09[x86] Add a ZERO_EXTEND_VECTOR_INREG DAG node and use it when wideningChandler Carruth1-0/+42
2014-07-02[cleanup] Hoist an if-else chain on ISD opcodes (really designed forChandler Carruth1-17/+28
2014-07-02[cleanup] Remove dead 'break;' statements that I meant to nuke inChandler Carruth1-2/+0
2014-07-02[cleanup] Hoist the promotion dispatch logic into the promote functionChandler Carruth1-23/+22
2014-07-02[cleanup] Nuke the 'VectorOp' bit of the promote method names.Chandler Carruth1-9/+9
2014-07-02[x86] Clean up and modernize the doxygen and API comments for the vectorChandler Carruth1-24/+38
2014-05-19SDAG: Legalize vector BSWAP into a shuffle if the shuffle is legal but the bs...Benjamin Kramer1-0/+27
2014-04-28Convert more SelectionDAG functions to use ArrayRef.Craig Topper1-2/+1
2014-04-26Convert SelectionDAG::getNode methods to use ArrayRef<SDValue>.Craig Topper1-10/+7
2014-03-17[VectorLegalizer/X86] Don't unvectorize fp_to_uint for v8f32->v8i16Adam Nemet1-7/+41
2014-03-02[C++11] Replace llvm::next and llvm::prior with std::next and std::prev.Benjamin Kramer1-2/+2
2014-02-03Expand vector bswap in LegalizeVectorOpsHal Finkel1-0/+1
2013-10-28Keep TBAA info when rewriting SelectionDAG loads and storesRichard Sandiford1-4/+7
2013-09-12Remove pointless assertion after r190376Matt Arsenault1-2/+0
2013-09-10Don't use getSetCCResultType for creating a vselectMatt Arsenault1-2/+1
2013-08-26SelectionDAG: Remove unnecessary uses of TargetLowering::getPointerTy()Tom Stellard1-3/+3
2013-08-21SelectionDAG: Make sure stores are always added to the LegalizedNodes listTom Stellard1-1/+1
2013-08-19Add a llvm.copysign intrinsicHal Finkel1-0/+1
2013-08-07Add ISD::FROUND for libm round()Hal Finkel1-0/+1
2013-08-05TargetLowering: Add getVectorIdxTy() function v2Tom Stellard1-3/+3
2013-07-08Remove trailing whitespace from SelectionDAG/*.cppStephen Lin1-1/+1
2013-06-14Introduce getSelect usage and use more getSelectCCMatt Arsenault1-5/+5
2013-05-28Remove double semicolons.Benjamin Kramer1-9/+9
2013-05-25Track IR ordering of SelectionDAG nodes 2/4.Andrew Trick1-10/+10
2013-05-18Add LLVMContext argument to getSetCCResultTypeMatt Arsenault1-2/+3
2013-05-07Fix vselect when getSetCCResultType returns a different type from the operandsMatt Arsenault1-3/+8
2013-02-22SelectionDAG compile time improvement.Nadav Rotem1-0/+19
2013-02-20Fix PR15267Michael Liao1-14/+119
2013-01-25This patch aims to reduce compile time in LegalizeTypes by using SmallDenseMap,Preston Gurd1-1/+1
2013-01-12When lowering an inreg sext first shift left, then right arithmetically.Benjamin Kramer1-3/+3
2013-01-11PPC: Implement efficient lowering of sign_extend_inreg.Nadav Rotem1-1/+25
2012-12-19Change TargetLowering::getTypeToPromoteTo to take and return MVTs,Patrik Hagglund1-2/+2
2012-12-19Change TargetLowering::getTruncStoreAction to take MVTs, instead of EVTs.Patrik Hagglund1-2/+2
2012-12-11Revert EVT->MVT changes, r169836-169851, due to buildbot failures.Patrik Hagglund1-4/+4
2012-12-11Change TargetLowering::getTypeToPromoteTo to take and return MVTs,Patrik Hagglund1-2/+2
2012-12-11Change TargetLowering::getTruncStoreAction to take MVTs, instead of EVTs.Patrik Hagglund1-2/+2
2012-11-17Mark FP_EXTEND form v2f32 to v2f64 as "expand" for ARM NEON. Patch by Pete C...Eli Friedman1-0/+1
2012-11-15Mark FP_ROUND for converting NEON v2f64 to v2f32 as expand. Add a missingEli Friedman1-0/+1
2012-09-02Fix a typo.Nadav Rotem1-1/+1
2012-09-02Generate better select code by allowing the target to use scalar select, and ...Nadav Rotem1-4/+3
2012-09-01Only legalise a VSELECT in to bitwise operations if the vector mask bool is z...Pete Cooper1-1/+6
2012-08-30Currently targets that do not support selects with scalar conditions and vect...Nadav Rotem1-1/+65