Commits · a9fb321a67943b9fffac6ff2d56ad5acb458b4f4 · Lorenzo Albano / LLVM bpEVL · GitLab

May 14, 2021
- [AMDGPU] Fix codegen of image intrinsics for g16 and a16 · 31b62aa1
  David Stuttard authored May 07, 2021
  31b62aa1
- [AMDGPU][AsmParser/Disassembler] Correct A16 and G16 handling · 72d570ca
  David Stuttard authored Apr 30, 2021
  72d570ca
- [llvm][AsmPrinter] Restore source location to register clobber warning · 2db090a2
  David Spickett authored May 11, 2021
  2db090a2
- New tag for ittapi - fix an error related to cross-compiling ITTAPI in LLVM with mingw · 444f02d7
  Alexey Bader authored May 14, 2021
  444f02d7
- [GVN] Clobber partially aliased loads. · fdae3fc8
  dfukalov authored Apr 09, 2021
  fdae3fc8
- [DSE] Move isOverwrite into DSEState. NFC · f7cb6547
  David Green authored May 14, 2021
  f7cb6547
- [ORC] Add support for adding LinkGraphs directly to ObjectLinkingLayer. · 0fda4c47
  Lang Hames authored May 13, 2021
  0fda4c47
- [AMDGPU] Do not clause NSA instructions · 9cf6ff7a
  Carl Ritson authored May 14, 2021
  9cf6ff7a
- [Debug-Info] change Tag type to dwarf::Tag for createAndAddDIE; NFC · 61484762
  Chen Zheng authored May 10, 2021
  61484762
- [Debug-Info] make DIE attributes generation under strict DWARF control · 75f3beee
  Chen Zheng authored May 09, 2021
  75f3beee
- [AArch64][GlobalISel] Fix a crash during unsuccessful G_CTPOP <2 x s64> legalization. · af6eb1c7
  Amara Emerson authored May 13, 2021
  af6eb1c7
- AMDGPU/GlobalISel: Don't hardcode stack alignment in assert message · 85394d9e
  Matt Arsenault authored May 13, 2021
  
  85394d9e
- AMDGPU/GlobalISel: Implement tail calls · 6a70874d
  Matt Arsenault authored Jan 12, 2021
  6a70874d
- [IR] Introduce the opaque pointer type · 2155dc51
  Arthur Eubanks authored May 01, 2021
  2155dc51
May 13, 2021
- [CaptureTracking] Use isIdentifiedFunctionLocal() (NFC) · 425781bc
  Nikita Popov authored May 13, 2021
  425781bc
- [AA] Use isIdentifiedFunctionLocal() (NFC) · dce158c5
  Nikita Popov authored May 13, 2021
  dce158c5
- Revert "[X86][CostModel] X86TTIImpl::getMemoryOpCost(): rewrite vector handling again" · 5fddc331
  Roman Lebedev authored May 13, 2021
  5fddc331
- Revert "[X86] X86TTIImpl::getInterleavedMemoryOpCostAVX2(): use getMemoryOpCost()" · 6b95fd19
  Roman Lebedev authored May 13, 2021
  6b95fd19
- [X86] AMD Zen 3: same-reg SSE XMM XORPS is a 1-cycle(!) dep-breaking one-idiom · aa0dcb3b
  Roman Lebedev authored May 13, 2021
  aa0dcb3b
- [OpenMP] Prevent Attributor from deleting functions in OpenMPOptCGSCC pass · 8b57ed09
  Joseph Huber authored May 13, 2021
  8b57ed09
- [AMDGPU] Add gfx1034 target · 464e4dc5
  Aakanksha Patil authored May 13, 2021
  464e4dc5
- Support unwinding from inline assembly · 8ec9fd48
  cynecx authored May 13, 2021
  8ec9fd48
- [PowerPC] Add ROP Protection to prologue and epilogue · 54310fc1
  Stefan Pintilie authored May 13, 2021
  54310fc1
- [ARM] Constrain CMPZ shift combine to a single use · 1011d4ed
  David Green authored May 13, 2021
  1011d4ed
- [InstSimplify] Remove redundant {insert,extract}_vector intrinsic chains · 2ed7db0d
  Joe Ellis authored May 13, 2021
  2ed7db0d
- [AMDGPU] Only allow global fp atomics with unsafe option · 8f98356b
  Stanislav Mekhanoshin authored May 12, 2021
  8f98356b
- [AArch64][SVE] Fix missed immediate selection due to mishandling of signedness · b1a07495
  Bradley Smith authored May 04, 2021
  b1a07495
- [TableGen] Make the NUL character invalid in .td files · fe9101c3
  Paul C. Anagnostopoulos authored May 11, 2021
  fe9101c3
- Reapply [ConstantFold] Fold more operations to poison · 395607af
  Juneyoung Lee authored Nov 29, 2020
  395607af
- [PowerPC] Handle inline assembly clobber of link regsiter · 15051f0b
  Stefan Pintilie authored May 13, 2021
  15051f0b
- [VPlan] Adjust assert in splitBlock to allow splitting at end. · bdada754
  Florian Hahn authored May 13, 2021
  bdada754
- [X86] X86ExpandPseudo.cpp - try to pass DebugLoc by const-ref to avoid costly... · ba0ec1be
  Simon Pilgrim authored May 13, 2021
  ba0ec1be
- [X86] X86InstrInfo.cpp - try to pass DebugLoc by const-ref to avoid costly... · 49566556
  Simon Pilgrim authored May 13, 2021
  49566556
- [X86] VZeroUpperInserter::insertVZeroUpper - avoid DebugLoc creation by... · 9dfc4ac4
  Simon Pilgrim authored May 13, 2021
  9dfc4ac4
- [Passes] Run GlobalsAA before LICM during LTO in new PM. · 860b3752
  Florian Hahn authored May 13, 2021
  860b3752
- [Passes] Use MemorySSA for LICM during LTO. · 3eaf2358
  Florian Hahn authored May 13, 2021
  3eaf2358
- [PowerPC] Provide doubleword vector predicate form comparisons on Power7 · 39e4676c
  Nemanja Ivanovic authored May 12, 2021
  39e4676c
- [SCEV] Apply guards to max with non-unitary steps. · e2759f11
  Florian Hahn authored May 13, 2021
  e2759f11
- Revert "[SimpleLoopUnswitch] Port partially invariant unswitch from... · 107d19eb
  Jingu Kang authored May 13, 2021
  107d19eb
- [FPEnv][X86] Implement lowering of llvm.set.rounding · 12537ab7
  Serge Pavlov authored May 13, 2021
  12537ab7