projects / oota-llvm.git / history
? search:
summary | shortlog | log | commit | commitdiff | tree
first ⋅ prev ⋅ next
Added more insertps optimizations
[oota-llvm.git] / lib / Target / X86 / X86ISelLowering.cpp
2014-05-19 Filipe CabecinhasAdded more insertps optimizations blob | commitdiff | raw
2014-05-19 Benjamin KramerSDAG: Legalize vector BSWAP into a shuffle if the shuff... blob | commitdiff | raw | diff to current
2014-05-17 Saleem AbdulrasoolTarget: remove old constructors for CallLoweringInfo blob | commitdiff | raw | diff to current
2014-05-17 Chandler Carruth[x86] Fix a bad predicate I spotted by inspection ... blob | commitdiff | raw | diff to current
2014-05-16 Filipe CabecinhasImplemented special cases for PerformVSELECTCombine. blob | commitdiff | raw | diff to current
2014-05-16 Filipe CabecinhasLower vselects into X86ISD::BLENDI when appropriate. blob | commitdiff | raw | diff to current
2014-05-16 Filipe CabecinhasImplemented LowerVSELECT to custom lower some instructions. blob | commitdiff | raw | diff to current
2014-05-16 Rafael EspindolaDelete getAliasedGlobal. blob | commitdiff | raw | diff to current
2014-05-15 Andrea Di Biagio[X86] Teach the backend how to fold SSE4.1/AVX/AVX2... blob | commitdiff | raw | diff to current
2014-05-15 Alp TokerFix typos blob | commitdiff | raw | diff to current
2014-05-14 Jay FoadRename ComputeMaskedBits to computeKnownBits. "Masked... blob | commitdiff | raw | diff to current
2014-05-12 Reid KlecknerTry to fix an SDAG dependence issue with sret blob | commitdiff | raw | diff to current
2014-05-12 Aaron BallmanSilencing an MSVC warning about not all control paths... blob | commitdiff | raw | diff to current
2014-05-12 Benjamin KramerX86: Make sure that we have SSE4.1 before we generate... blob | commitdiff | raw | diff to current
2014-05-12 NAKAMURA TakumiX86ISelLowering.cpp:LowerINTRINSIC_W_CHAIN(): Prune... blob | commitdiff | raw | diff to current
2014-05-12 Elena DemikhovskyFixed compilation issue blob | commitdiff | raw | diff to current
2014-05-12 Elena DemikhovskyAVX-512: changes in intrinsics blob | commitdiff | raw | diff to current
2014-05-11 Hal FinkelPass the value type to TLI::getRegisterByName blob | commitdiff | raw | diff to current
2014-05-11 Filipe CabecinhasFixed a bug when lowering build_vector (PR19694) blob | commitdiff | raw | diff to current
2014-05-09 Reid KlecknerAllow sret on the second parameter as well as the first blob | commitdiff | raw | diff to current
2014-05-09 Andrea Di BiagioFix 80 col violation. blob | commitdiff | raw | diff to current
2014-05-08 Filipe CabecinhasOptimize shufflevector that copies an i64/f64 and zeros... blob | commitdiff | raw | diff to current
2014-05-08 Andrea Di Biagio[X86] Add target specific combine rules to fold SSE2... blob | commitdiff | raw | diff to current
2014-05-08 Filipe CabecinhasLower certain build_vectors to insertps instructions blob | commitdiff | raw | diff to current
2014-05-06 Andrea Di Biagio[X86] Improve the lowering of BITCAST dag nodes from... blob | commitdiff | raw | diff to current
2014-05-06 Renato GolinImplememting named register intrinsics blob | commitdiff | raw | diff to current
2014-05-06 Reid KlecknerFix i128 div/mod on mingw64 blob | commitdiff | raw | diff to current
2014-05-05 Filipe CabecinhasRevert "Optimize shufflevector that copies an i64/f64... blob | commitdiff | raw | diff to current
2014-05-05 Filipe CabecinhasOptimize shufflevector that copies an i64/f64 and zeros... blob | commitdiff | raw | diff to current
2014-04-30 Craig TopperUse makeArrayRef insted of calling ArrayRef<T> construc... blob | commitdiff | raw | diff to current
2014-04-29 Reid KlecknerImplement X86 code generation for musttail blob | commitdiff | raw | diff to current
2014-04-29 Elena DemikhovskyAVX-512: optimized a shuffle pattern to VINSERTI64x4. blob | commitdiff | raw | diff to current
2014-04-28 Quentin Colombet[X86] Add more details in the comments of X86TargetLowe... blob | commitdiff | raw | diff to current
2014-04-28 Craig Topper[C++] Use 'nullptr'. blob | commitdiff | raw | diff to current
2014-04-27 Craig TopperConvert one last signature of getNode to take an ArrayR... blob | commitdiff | raw | diff to current
2014-04-27 Craig TopperConvert SelectionDAG::getMergeValues to use ArrayRef. blob | commitdiff | raw | diff to current
2014-04-27 Benjamin KramerX86: If SSE4.1 is missing lower SMUL_LOHI of v4i32... blob | commitdiff | raw | diff to current
2014-04-26 Craig TopperConvert getMemIntrinsicNode to take ArrayRef of SDValue... blob | commitdiff | raw | diff to current
2014-04-26 Craig TopperConvert SelectionDAG::getNode methods to use ArrayRef... blob | commitdiff | raw | diff to current
2014-04-26 Benjamin KramerPrint X86ISD::PMULDQ nodes properly in debug output. blob | commitdiff | raw | diff to current
2014-04-26 Benjamin KramerX86: Lower SMUL_LOHI of v4i32 to pmuldq when SSE4.1... blob | commitdiff | raw | diff to current
2014-04-26 Benjamin KramerX86: Add patterns for MULHU/MULHS of v8i16 and v16i16. blob | commitdiff | raw | diff to current
2014-04-26 Benjamin KramerRip out X86-specific vector SDIV lowering, make the... blob | commitdiff | raw | diff to current
2014-04-26 Benjamin KramerX86: Custom lower v4i32 UMUL_LOHI into 2 pmuludqs. blob | commitdiff | raw | diff to current
2014-04-26 Quentin Colombet[X86] Implement TargetLowering::getScalingFactorCost... blob | commitdiff | raw | diff to current
2014-04-25 Filipe CabecinhasOptimization for certain shufflevector by using insertps. blob | commitdiff | raw | diff to current
2014-04-25 Craig Topper[C++] Use 'nullptr'. Target edition. blob | commitdiff | raw | diff to current
2014-04-24 Benjamin KramerX86: Don't transform shifts into ands when the sign... blob | commitdiff | raw | diff to current
2014-04-24 Reid KlecknerAdd 'musttail' marker to call instructions blob | commitdiff | raw | diff to current
2014-04-24 Andrea Di Biagio[X86] Add support for Read Time Stamp Counter x86 built... blob | commitdiff | raw | diff to current
2014-04-24 Benjamin KramerX86: Emit test instead of constant shift + compare... blob | commitdiff | raw | diff to current
2014-04-22 Elena DemikhovskyAVX-512: store and truncstore for i1 values blob | commitdiff | raw | diff to current
2014-04-22 Lang Hames[X86] Use tablegen instead of DAG combines to match... blob | commitdiff | raw | diff to current
2014-04-22 Lang Hames[X86] Don't use BZHI for short masks (>=32 bits). Thank... blob | commitdiff | raw | diff to current
2014-04-22 Chandler Carruth[Modules] Fix potential ODR violations by sinking the... blob | commitdiff | raw | diff to current
2014-04-21 Lang Hames[X86] ISEL (and X, <constant mask>) to BZHI when BMI2... blob | commitdiff | raw | diff to current
2014-04-18 Adam Nemet[X86] Improve buildFromShuffleMostly for AVX blob | commitdiff | raw | diff to current
2014-04-15 Andrea Di Biagio[X86] Improve the lowering of packed shifts by constant... blob | commitdiff | raw | diff to current
2014-04-15 Nick LewyckyBreak PseudoSourceValue out of the Value hierarchy... blob | commitdiff | raw | diff to current
2014-04-14 David BlaikieChange argument order and add explanatory comment to... blob | commitdiff | raw | diff to current
2014-04-13 David BlaikieFix instruction debug info location during legalization blob | commitdiff | raw | diff to current
2014-04-10 Reid KlecknerMove the segmented stack switch to a function attribute blob | commitdiff | raw | diff to current
2014-04-09 Elena DemikhovskyAVX-512: insert element to mask vector; store i1 data blob | commitdiff | raw | diff to current
2014-04-08 Elena DemikhovskyAVX-512: Added fp_to_uint and uint_to_fp patterns. blob | commitdiff | raw | diff to current
2014-04-04 Matt ArsenaultAdd DAG parameter to ComputeNumSignBitsForTargetNode blob | commitdiff | raw | diff to current
2014-04-04 Craig TopperMake consistent use of MCPhysReg instead of uint16_t... blob | commitdiff | raw | diff to current
2014-04-02 Yaron KerenAdded isTargetWindowsMSVC(), renamed isTargetMingw... blob | commitdiff | raw | diff to current
2014-04-01 Yaron KerenIf isKnownWindowsMSVCEnvironment then getOS == Triple... blob | commitdiff | raw | diff to current
2014-04-01 Yaron KerenisTargetWindows() renamed to isTargetKnownWindowsMSVC() blob | commitdiff | raw | diff to current
2014-04-01 Aaron BallmanAttempting to fix r205124, which had failed asserts... blob | commitdiff | raw | diff to current
2014-04-01 Alexey Volkov[x86] Do not convert to cmp32 for Atom arch by Sergey... blob | commitdiff | raw | diff to current
2014-03-27 Rafael EspindolaPrevent alias from pointing to weak aliases. blob | commitdiff | raw | diff to current
2014-03-26 Rafael EspindolaRevert "Prevent alias from pointing to weak aliases." blob | commitdiff | raw | diff to current
2014-03-26 Rafael EspindolaPrevent alias from pointing to weak aliases. blob | commitdiff | raw | diff to current
2014-03-25 Adam Nemet[X86] Generate VPSHUFB for in-place v16i16 shuffles blob | commitdiff | raw | diff to current
2014-03-25 Adam Nemet[X86] Factor out new helper getPSHUFB blob | commitdiff | raw | diff to current
2014-03-24 Adam Nemet[X86] Fix non-determinism in LowerVectorAllZeroTest blob | commitdiff | raw | diff to current
2014-03-19 Craig TopperPrune includes in X86 target. blob | commitdiff | raw | diff to current
2014-03-17 Adam Nemet[X86] Fix unused variable warning with NDEBUG from... blob | commitdiff | raw | diff to current
2014-03-17 Adam Nemet[VectorLegalizer/X86] Don't unvectorize fp_to_uint... blob | commitdiff | raw | diff to current
2014-03-15 Arnaud A. de Grand... Remove some dead assignements found by scan-build blob | commitdiff | raw | diff to current
2014-03-13 Owen AndersonPhase 2 of the great MachineRegisterInfo cleanup. ... blob | commitdiff | raw | diff to current
2014-03-11 Hans WennborgX86: Don't generate 64-bit movd after cmpneqsd in 32... blob | commitdiff | raw | diff to current
2014-03-11 Tim NorthoverIR: add a second ordering operand to cmpxhg for failure blob | commitdiff | raw | diff to current
2014-03-11 Jim GrosbachX86: Enable ISel of 16-bit MOVBE instructions. blob | commitdiff | raw | diff to current
2014-03-05 Cameron McInallyLower AVX v4i64->v4i32 truncate to one shuffle. blob | commitdiff | raw | diff to current
2014-03-04 Chandler Carruth[Modules] Move CallSite into the IR library where it... blob | commitdiff | raw | diff to current
2014-03-02 Benjamin Kramer[C++11] Replace llvm::next and llvm::prior with std... blob | commitdiff | raw | diff to current
2014-03-02 Elena DemikhovskyAVX-512: Fixed extract_vector_elt for v8i1 vector blob | commitdiff | raw | diff to current
2014-02-26 Quentin ColombetLower unsigned vsetcc to psubus in certain cases blob | commitdiff | raw | diff to current
2014-02-19 Tim NorthoverX86 CodeGenPrep: sink shufflevectors before shifts blob | commitdiff | raw | diff to current
2014-02-18 Tim NorthoverX86: use vpsllvd (& friends) for 16-bit shifts on Haswell blob | commitdiff | raw | diff to current
2014-02-16 Elena DemikhovskyAVX-512: simpyfied BUILD_VECTOR for masks; fixed cmp... blob | commitdiff | raw | diff to current
2014-02-12 Andrea Di Biagio[X86] Teach the backend how to lower vector shift left... blob | commitdiff | raw | diff to current
2014-02-11 Elena DemikhovskyAVX: fixed a bug in LowerVECTOR_SHUFFLE blob | commitdiff | raw | diff to current
2014-02-11 Elena DemikhovskyAVX-512: Optimized BUILD_VECTOR pattern; blob | commitdiff | raw | diff to current
2014-02-10 Elena DemikhovskyAVX-512: Fixed extract_vector_elt for v16i1 and v8i1... blob | commitdiff | raw | diff to current
2014-02-06 Tim NorthoverX86: deduplicate V[SZ]EXT_MOVL and V[SZ]EXT nodes blob | commitdiff | raw | diff to current
2014-02-05 Matt ArsenaultAdd address space argument to allowsUnalignedMemoryAccess. blob | commitdiff | raw | diff to current
2014-02-05 Elena DemikhovskyAVX-512: optimized icmp -> sext -> icmp pattern blob | commitdiff | raw | diff to current
next
C/C++ LLVM-based compilers that forbids OOTA behaviors