Commits · db63088ea7cc8890972e875c87550609e6bfaf4a · Raul Torres / llvm-target-spread · GitLab

Aug 16, 2018
- [SelectionDAG] Improve the legalisation lowering of UMULO. · 73e8a784
  Eli Friedman authored Aug 16, 2018
  73e8a784
- [RegisterCoalescer] Shrink to uses if needed after removeCopyByCommutingDef · 17143f61
  Krzysztof Parzyszek authored Aug 16, 2018
  17143f61
- Fix memory leak in demangling of string literals. · af738f72
  Zachary Turner authored Aug 16, 2018
  af738f72
- [TargetLowering] Add support for non-uniform vectors to BuildSDIV · 87d0039a
  Simon Pilgrim authored Aug 16, 2018
  87d0039a
- [codeview] Use push_macro to avoid conflicts instead of a prefix · bd5d7122
  Reid Kleckner authored Aug 16, 2018
  bd5d7122
- [MC] Cleanup noop default case spelling. NFC. · eb189a0e
  Nirav Dave authored Aug 16, 2018
  eb189a0e
- AMDGPU: Custom lower fexp · 7121bed2
  Matt Arsenault authored Aug 16, 2018
  7121bed2
- [TargetLowering] Refactor BuildSDIV in preparation for D50765. NFCI. · ede49053
  Simon Pilgrim authored Aug 16, 2018
  ede49053
- [MC] Remove unused variable · 0ce64c81
  Benjamin Kramer authored Aug 16, 2018
  0ce64c81
- [MC][X86] Enhance X86 Register expression handling to more closely match GCC. · 7fd992a7
  Nirav Dave authored Aug 16, 2018
  7fd992a7
- Fix -Wmicrosoft-goto warnings. · d78fe2f4
  Zachary Turner authored Aug 16, 2018
  d78fe2f4
- Add support for AVX-512 CodeView registers. · 2838b591
  Zachary Turner authored Aug 16, 2018
  2838b591
- [MS Demangler] Demangle string literals. · 970fdc32
  Zachary Turner authored Aug 16, 2018
  970fdc32
- [MS Demangler] Don't fail on MD5-mangled names. · 83313f8f
  Zachary Turner authored Aug 16, 2018
  83313f8f
- [InstCombine] Expand the simplification of pow(x, 0.5) to sqrt(x) · c05c7e11
  Evandro Menezes authored Aug 16, 2018
  c05c7e11
- [InstCombine] move vector compare before same-shuffled ops · 039f556f
  Sanjay Patel authored Aug 16, 2018
  039f556f
- [ARM] Ignore GEPs in ARMCodeGenPrepare · 0d511970
  Sam Parker authored Aug 16, 2018
  0d511970
- [ARM] Allow zext in ARMCodeGenPrepare · 0e2f0bd4
  Sam Parker authored Aug 16, 2018
  0e2f0bd4
- [RISCV][MC] Don't fold symbol differences if requiresDiffExpressionRelocations is true · fdc4647c
  Alex Bradbury authored Aug 16, 2018
  fdc4647c
- [ADT] Replace APInt::WORD_MAX with APInt::WORDTYPE_MAX · 8f46505b
  Simon Pilgrim authored Aug 16, 2018
  8f46505b
- [ARM] Allow signed icmps in ARMCodeGenPrepare · 13567dbb
  Sam Parker authored Aug 16, 2018
  13567dbb
- [mips] Remove dead code from MipsPassConfig · a8ac4308
  Simon Atanasyan authored Aug 16, 2018
  a8ac4308
- [NFC] Remove const modifier to allow further development in LICM · 72d7d649
  Max Kazantsev authored Aug 16, 2018
  72d7d649
- [NFC] Add missing const modifier · a7415874
  Max Kazantsev authored Aug 16, 2018
  a7415874
- [X86] Remove masking from the 512-bit padds and psubs intrinsics. Use select in IR instead. · 9c1d9fde
  Craig Topper authored Aug 16, 2018
  9c1d9fde
- [X86] Remove the unused masked 128 and 256-bit masked padds/psubs intrinsics. · 9d6983c9
  Craig Topper authored Aug 16, 2018
  9d6983c9
- [x86] Actually initialize the SLH pass with the x86 backend and use · 00c35c77
  Chandler Carruth authored Aug 16, 2018
  00c35c77
- [BFI] Use rounding while computing profile counts. · aca738b7
  Easwaran Raman authored Aug 16, 2018
  aca738b7
- [Metadata] Replace a SmallVector with an array; NFC · 3083105b
  George Burgess IV authored Aug 15, 2018
  3083105b
- · 8c17f9a7
  Guozhi Wei authored Aug 15, 2018
  8c17f9a7
Aug 15, 2018
- AMDGPU: Fold fneg into fmed3 · f533e6b0
  Matt Arsenault authored Aug 15, 2018
  f533e6b0
- AMDGPU: Improve extract_vector_elt reduction combine · a8160737
  Matt Arsenault authored Aug 15, 2018
  a8160737
- AMDGPU: Implement llvm.amdgcn.icmp/fcmp for i16/f16 · b3a80e53
  Matt Arsenault authored Aug 15, 2018
  b3a80e53
- [X86] Improve AVX1 shuffle lowering for v8f32 shuffles where the low half... · 08e08261
  Craig Topper authored Aug 15, 2018
  08e08261
- AMDGPU: Stop producing icmp/fcmp intrinsics with invalid types · 9a389fbd
  Matt Arsenault authored Aug 15, 2018
  9a389fbd
- AMDGPU: Address todo for handling 1/(2 pi) · 6c7ba829
  Matt Arsenault authored Aug 15, 2018
  6c7ba829
- DAG: Use getObjectOffset helper · 0f2c1cf4
  Matt Arsenault authored Aug 15, 2018
  0f2c1cf4
- DAG: Try to custom lower when promoting float operands · 22f01268
  Matt Arsenault authored Aug 15, 2018
  22f01268
- [MCJIT] Fix a case of Error::success() being passed to report_fatal_error. · 942cb7b3
  Lang Hames authored Aug 15, 2018
  942cb7b3
- Revert "[ARM] Allow signed icmps in ARMCodeGenPrepare" · ed4239f4
  Vitaly Buka authored Aug 15, 2018
  ed4239f4