Commits · e27a6db5298f6ba3c1dbc8bab25c769cfa761b2a · Roger Ferrer / llvm-epi · GitLab

Sep 27, 2021
- Bad SLPVectorization shufflevector replacement, resulting in write to wrong memory location · e27a6db5
  Jameson Nash authored Jul 22, 2021
  e27a6db5
- [AMDGPU] Change ASAN init/fini kernels linkage to external. · e90b512c
  Praveen Velliengiri authored Sep 27, 2021
  e90b512c
- [PowerPC] Fix td pattern for P10 VSLDBI and VSRDBI · 682e15f3
  Quinn Pham authored Sep 16, 2021
  682e15f3
- [RISCV] Fold store of vmv.x.s to a vse with VL=1. · a2a07e8d
  Craig Topper authored Sep 27, 2021
  a2a07e8d
- [InstCombine] Fix an "unused variable" warning · 59540b29
  Kazu Hirata authored Sep 27, 2021
  
  59540b29
- [RISCV] Improve support for forming widening multiplies when one input is a scalar splat. · 933182e9
  Craig Topper authored Sep 27, 2021
  933182e9
- [NFC][AMDGPU] Update cost model tests: · 1f73f0c1
  Daniil Fukalov authored Sep 27, 2021
  1f73f0c1
- [InstCombine] move shl-only folds out from under commonShiftTransforms(); NFCI · 9075edc8
  Sanjay Patel authored Sep 27, 2021
  9075edc8
- [Lanai] Remove redundant declaration getTheLanaiTarget (NFC) · b68a62b3
  Kazu Hirata authored Sep 27, 2021
  b68a62b3
- [OpenMP] Introduce a new worksharing RTL function for distribute · b4a55436
  Joseph Huber authored Sep 24, 2021
  b4a55436
- [llvm/OptTable] Add named param comment for GroupedShortOption · 76645089
  Nico Weber authored Sep 27, 2021
  
  76645089
- Fix tests defaulting to incorrect triples on AIX · 56049b71
  Jake Egan authored Sep 27, 2021
  56049b71
- [llvm/OptTable] Drop "The" prefix on fields · 730bbc6f
  Nico Weber authored Sep 27, 2021
  
  730bbc6f
- [llvm] Convert OptTable::ParseOneArg() to std::unique_ptr<> · 6ffd8e39
  Nico Weber authored Sep 27, 2021
  
  6ffd8e39
- [llvm] Convert OptTable::parseOneArgGrouped() to std::unique_ptr<> · 7789a68e
  Nico Weber authored Sep 27, 2021
  
  7789a68e
- [llvm] ConvertOption::accept(), acceptInternal() to std::unique_ptr<> · 2f955424
  Nico Weber authored Sep 27, 2021
  2f955424
- [InstCombine] generalize fold for (trunc (X u>> C1)) u>> C · 21429cf4
  Sanjay Patel authored Sep 27, 2021
  21429cf4
- [InstCombine] match variable names and code comments; NFC · 025a805d
  Sanjay Patel authored Sep 26, 2021
  025a805d
- [AMDGPU] Ignore KILLs when forming clauses · bf980930
  Sebastian Neubauer authored Jul 16, 2021
  bf980930
- [FuncSpec] Don't specialise (or crash) on poison or constexpr values · eba76056
  Sjoerd Meijer authored Sep 27, 2021
  eba76056
- [AArch64] Fix neon-reverseshuffle test extension. NFC · ebee606e
  David Green authored Sep 27, 2021
  ebee606e
- [LoopFlatten] Precommit new test widen-iv2.ll for D110234. · a588ae48
  Sjoerd Meijer authored Sep 22, 2021
  
  a588ae48
- [llvm-dwarfdump][docs] Add missing options to the help output and the command guide · 05b1c7ae
  gbreynoo authored Sep 27, 2021
  05b1c7ae
- Revert "Recommit "Revert "[CVP] processSwitch: Remove default case when switch... · 3a998c06
  Jun Ma authored Sep 27, 2021
  3a998c06
- [gn build] Port 9da2fa27 · e2eb651c
  LLVM GN Syncbot authored Sep 27, 2021
  
  e2eb651c
- [X86][Costmodel] Load/store i16 Stride=2 VF=32 interleaving costs · 7424deb7
  Roman Lebedev authored Sep 27, 2021
  7424deb7
- [X86][Costmodel] Load/store i16 Stride=2 VF=16 interleaving costs · a5113e94
  Roman Lebedev authored Sep 27, 2021
  a5113e94
- [X86][Costmodel] Load/store i16 Stride=2 VF=8 interleaving costs · 70c90cc5
  Roman Lebedev authored Sep 27, 2021
  70c90cc5
- [X86][Costmodel] Load/store i16 Stride=2 VF=4 interleaving costs · 49e532aa
  Roman Lebedev authored Sep 27, 2021
  49e532aa
- [DAGCombiner][VP] Fold zero-length or false-masked VP ops · e2b46e33
  Fraser Cormack authored Sep 01, 2021
  e2b46e33
- [ARM] Improve detection of fallthough when aligning blocks · bb2d23dc
  David Green authored Sep 27, 2021
  bb2d23dc
- [Test] Regenerate test checks with autogen script · 4992220e
  Max Kazantsev authored Sep 27, 2021
  
  4992220e
- [X86] combineVectorHADDSUB - remove the broken HOP(x,x) merging code (PR51974) · 468ff703
  Simon Pilgrim authored Sep 27, 2021
  468ff703
- [LV] Add tests where rt checks may make vectorization unprofitable. · 4b581e87
  Florian Hahn authored Sep 07, 2021
  4b581e87
- [AArch64][test] Add more tests of add/sub with immediate · 683e5063
  Ben Shi authored Sep 25, 2021
  683e5063
- [llvm] Disable LLVM_ENABLE_PER_TARGET_RUNTIME_DIR by default on Arm Linux · 3c65d54e
  David Spickett authored Sep 27, 2021
  3c65d54e
- [RISCV] Create the correct mask type when lowering EXTRACT_VECTOR_ELT · d48f6df1
  Fraser Cormack authored Sep 24, 2021
  d48f6df1
- [llvm] [ADT] Add a range/iterator-based Split() · f4b71e34
  Michał Górny authored Sep 26, 2021
  f4b71e34
- [Test] Add test showing that SCEV cannot properly infer ranges of cycled phis · 0bd9162f
  Max Kazantsev authored Sep 27, 2021
  
  0bd9162f
- [X86][ISel] Lowering FROUND(f16) and FROUNDEVEN(f16) · 902ec614
  Freddy Ye authored Sep 27, 2021
  902ec614