Commits · 8171bd8e0f038fb72b39888fc34d39ce19804d40 · Raul Torres / llvm-target-spread · GitLab

Sep 21, 2018
- [X86][Sched] Add zero idiom sched data to the SNB model. · 8171bd8e
  Clement Courbet authored Sep 21, 2018
  8171bd8e
- [X86][BtVer2] Fix latency and resource cycles of AVX 256-bit zero-idioms. · 4cd5cf9f
  Andrea Di Biagio authored Sep 21, 2018
  4cd5cf9f
Sep 20, 2018
- [SelectionDAG] replace duplicated peekThroughBitcast helper functions; NFCI · 8a1227cc
  Sanjay Patel authored Sep 20, 2018
  8a1227cc
- [X86][SSE] Remove UNPCKL(SHUFFLE)->UNPCKH custom combine · 3e2de767
  Simon Pilgrim authored Sep 20, 2018
  3e2de767
- [X86][SSE] Remove PSHUFLW/PSHUFHW combineRedundantHalfShuffle combine · 46c1dcb1
  Simon Pilgrim authored Sep 20, 2018
  46c1dcb1
Sep 19, 2018
- [X86] Handle COPYs of physregs better (regalloc hints) · 2d0f20cc
  Simon Pilgrim authored Sep 19, 2018
  2d0f20cc
- [x86] change names of vector splitting helper functions; NFC · 1a1c0ee5
  Sanjay Patel authored Sep 19, 2018
  1a1c0ee5
- [X86] Add initial SimplifyDemandedVectorEltsForTargetNode support · 8191d63c
  Simon Pilgrim authored Sep 19, 2018
  8191d63c
- [TableGen][SubtargetEmitter] Add the ability for processor models to describe... · 8b6c314b
  Andrea Di Biagio authored Sep 19, 2018
  8b6c314b
- [DAGCombiner][x86] add transform/hook to decompose integer multiply into shift/add · 4fd2e2a4
  Sanjay Patel authored Sep 19, 2018
  4fd2e2a4
- [COFF] Emit @feat.00 on 64-bit and set the CFG bit when emitting guardcf tables · 4195eb10
  Hans Wennborg authored Sep 19, 2018
  4195eb10
Sep 18, 2018
- [X86][SSE] LowerShift - pull out repeated getTargetVShiftUniformOpcode calls. NFCI. · e9bf71e7
  Simon Pilgrim authored Sep 18, 2018
  e9bf71e7
Sep 17, 2018
- [X86ISel] Implement byval lowering for Win64 calling convention · c8ccaed3
  Keno Fischer authored Sep 17, 2018
  c8ccaed3
Sep 16, 2018
- [X86][SSE] Always enable ISD::SRL -> ISD::MULHU for v8i16 · cffa2064
  Simon Pilgrim authored Sep 16, 2018
  cffa2064
- [X86][AVX] Enable ISD::SRL -> ISD::MULHU for v16i16 · ea069ffd
  Simon Pilgrim authored Sep 16, 2018
  ea069ffd
- [x86] fix uses check in broadcast transform (PR38949) · bfee5a9b
  Sanjay Patel authored Sep 16, 2018
  bfee5a9b
Sep 15, 2018
- [X86] Remove an fp->int->fp domain crossing in LowerUINT_TO_FP_i64. · fe0b973f
  Craig Topper authored Sep 15, 2018
  fe0b973f
- [X86] Fold (movmsk (setne (and X, (1 << C)), 0)) -> (movmsk (X << C)) · 273f755d
  Craig Topper authored Sep 15, 2018
  273f755d
Sep 14, 2018
- [X86][SSE] Lower shuffles to permute(unpack(x,y)) (PR31151) · 32857c54
  Simon Pilgrim authored Sep 14, 2018
  32857c54
- [X86][BMI1] Fix BLSI/BLSMSK/BLSR BMI1 scheduling on btver2 · 1c1335a1
  Simon Pilgrim authored Sep 14, 2018
  1c1335a1
- [X86][BMI1] Add scheduler class for BLSI/BLSMSK/BLSR BMI1 instructions · 6a47cdbd
  Simon Pilgrim authored Sep 14, 2018
  6a47cdbd
Sep 13, 2018
- [X86] Fix register resizings for inline assembly register operands. · 59ad1c84
  Nirav Dave authored Sep 13, 2018
  59ad1c84
- [X86] Cleanup pair returns. NFCI. · 2060a16d
  Nirav Dave authored Sep 13, 2018
  2060a16d
- [X86] Type legalize v2i32 div/rem by scalarizing rather than promoting · f107123a
  Craig Topper authored Sep 13, 2018
  f107123a
Sep 12, 2018
- [X86] Remove isel patterns for ADCX instruction · 22626135
  Craig Topper authored Sep 12, 2018
  22626135
- [X86] Teach X86SelectionDAGInfo::EmitTargetCodeForMemcpy about GNUX32 · dc32e91b
  Craig Topper authored Sep 12, 2018
  dc32e91b
Sep 11, 2018
- [X86] Prefer unpckhpd over movhlps in isel for fake unary cases · 8238580a
  Craig Topper authored Sep 11, 2018
  8238580a
- [X86] Teach X86FastISel::X86SelectRet to use EAX for the sret pointer in GNUX32 · cc9efaff
  Craig Topper authored Sep 11, 2018
  cc9efaff
- [X86] Correct the one use check from r341915. · d7362a3e
  Craig Topper authored Sep 11, 2018
  d7362a3e
- [X86] In combineMOVMSK, look through int->fp bitcasts before callling SimplifyDemandedBits. · 844f035e
  Craig Topper authored Sep 11, 2018
  844f035e
Sep 10, 2018
- Move FeatureAES from SLM, WSM and SNB to GLM and SKL · 911ddd6d
  Erich Keane authored Sep 10, 2018
  911ddd6d
- [X86] Mark the ISD::SETLT/SETLE condition codes as illegal for v32i16/v64i8 to... · a5ae613c
  Craig Topper authored Sep 10, 2018
  a5ae613c
Sep 09, 2018
- [X86] Custom type legalize (v2i32 (fp_to_uint v2f64))) without avx512vl by... · 38235161
  Craig Topper authored Sep 09, 2018
  38235161
Sep 08, 2018
- [X86] Create paddus/psubus from narrower vectors with i8/i16 element types. · 7af5e333
  Craig Topper authored Sep 08, 2018
  7af5e333
- [X86] Mark the ADCX and ADOX instruction as commutable. · a2c9694b
  Craig Topper authored Sep 08, 2018
  a2c9694b
- [X86] Add commuted isel pattern for the load form of ADCX instructions. · c9630597
  Craig Topper authored Sep 08, 2018
  c9630597
Sep 07, 2018
- [X86] Don't create ZERO_EXTEND_INREG/SIGN_EXTEND_INREG for v1iX vectors. · 5cbce81c
  Craig Topper authored Sep 07, 2018
  5cbce81c
- [X86] Don't create X86ISD::AVG nodes from v1iX vectors. · 39f48fdc
  Craig Topper authored Sep 07, 2018
  39f48fdc
- [X86] Modify the the rdtscp intrinsic to return values instead of taking a pointer argument · 4863313b
  Craig Topper authored Sep 07, 2018
  4863313b
- [X86] Change the addcarry and subborrow intrinsics to return 2 results and... · 72964ae9
  Craig Topper authored Sep 07, 2018
  72964ae9