Commits · df93c8a9195322a393e8b27063c083006a4ddb19 · Roger Ferrer / llvm-epi · GitLab

Nov 03, 2021
- [X86] `X86TTIImpl::getInterleavedMemoryOpCostAVX512()`: fallback to... · df93c8a9
  Roman Lebedev authored Nov 03, 2021
  df93c8a9
- Reland "[AArch64][SVE][InstCombine] Combine contiguous gather/scatter to load/store" · 7a34145f
  Peter Waller authored Nov 03, 2021
  7a34145f
- Revert "[AArch64][SVE][InstCombine] Combine contiguous gather/scatter to load/store" · 753eba64
  Peter Waller authored Nov 03, 2021
  753eba64
- [NVPTX] Mark special registers as reserved · 123ad720
  Andrew Savonichev authored Nov 03, 2021
  123ad720
- [NVPTX] Add MoveParam instruction for TargetExternalSymbol operand · 0e707855
  Andrew Savonichev authored Nov 03, 2021
  0e707855
- [ARM] Treat MVE gather add-like-or's like adds · 3bc586b9
  David Green authored Nov 03, 2021
  3bc586b9
- [AArch64][SVE][InstCombine] Combine contiguous gather/scatter to load/store · 1febf42f
  Peter Waller authored Oct 25, 2021
  1febf42f
- [ARM] Push gather/scatter shl index updates out of loops · d36dd1f8
  David Green authored Nov 03, 2021
  d36dd1f8
- [PowerPC] Implement longdouble pack/unpack builtins · 741aeda9
  Qiu Chaofan authored Nov 03, 2021
  741aeda9
- [NVPTX] Copy machine operand flags in TII::insertBranch · 30a3a17d
  Andrew Savonichev authored Nov 03, 2021
  30a3a17d
- [ARM][AsmParser] Don't emit "deprecated instruction in IT block" warning if requested · 803d4f8a
  Yi Kong authored Nov 03, 2021
  803d4f8a
- Revert "[AArch64] Optimize add/sub with immediate" · 59c3b48d
  Ben Shi authored Nov 03, 2021
  59c3b48d
- [PowerPC] handle more splat loads without stack operation · 5a8b1963
  Chen Zheng authored Nov 03, 2021
  5a8b1963
- [AMDGPU] Fix SGPR checks in S_MOV_B64_IMM_PSEUDO generation. · fbe61fb0
  Abinav Puthan Purayil authored Nov 02, 2021
  fbe61fb0
- [AArch64] Optimize add/sub with immediate · 3de3ca31
  Ben Shi authored Nov 01, 2021
  3de3ca31
- [X86][VARARG] Assign MMO earlier to avoid prolog insert point been sunk across... · 8f101971
  Phoebe Wang authored Nov 03, 2021
  8f101971
Nov 02, 2021
- [AArch64] Diagnose large adrp offset on Windows. · c964afb2
  Eli Friedman authored Nov 02, 2021
  c964afb2
- [X86][AVX] combineConcatVectorOps - use getBROADCAST_LOAD helper for splat of... · 53900a19
  Simon Pilgrim authored Nov 02, 2021
  53900a19
- [X86][AVX] combineConcatVectorOps - use getBROADCAST_LOAD helper. NFCI. · 82e0eb22
  Simon Pilgrim authored Nov 02, 2021
  82e0eb22
- [RISCV] Optimize vp.load with an all-ones mask · d065b038
  Fraser Cormack authored Nov 02, 2021
  d065b038
- [AMDGPU] Really preserve LiveVariables in SILowerControlFlow · be1a8f88
  Jay Foad authored Oct 27, 2021
  be1a8f88
- Revert "[AArch64][SVE] Combine predicated FMUL/FADD into FMA" · 895145aa
  Matt authored Nov 02, 2021
  895145aa
- [X86][AVX] SimplifyDemandedVectorEltsForTargetNode - use getBROADCAST_LOAD helper. NFCI. · e173631d
  Simon Pilgrim authored Nov 02, 2021
  e173631d
- [X86][AVX] lowerV2X128Shuffle - use getBROADCAST_LOAD helper. NFCI. · 8ca666a2
  Simon Pilgrim authored Nov 02, 2021
  
  8ca666a2
- Fix building with GCC 12: · c5029023
  Martin Liska authored Nov 02, 2021
  c5029023
- [RISCV] Fix invalid kill on callee save · 4ec1b8ee
  David Callahan authored Nov 02, 2021
  4ec1b8ee
Nov 01, 2021
- [WebAssembly] support "return" and unreachable code in asm type checker · ac653664
  Wouter van Oortmerssen authored Nov 01, 2021
  ac653664
- BPF: Workaround an InstCombine ICmp transformation with llvm.bpf.compare builtin · f63405f6
  Yonghong Song authored Oct 30, 2021
  f63405f6
- [SVE] Fix VLS FMA matching for CodeGenOpt::Aggressive. · 702fd3d3
  Cameron McInally authored Nov 01, 2021
  702fd3d3
- [X86] Remove X86ELFObjectWriter in X86AsmBackend.cpp (NFC) · d000431f
  Kazu Hirata authored Nov 01, 2021
  d000431f
- [AMDGPU] Use MachineInstrBuilder::addReg. NFC. · 7afef229
  Jay Foad authored Nov 01, 2021
  
  7afef229
- [AMDGPU] Shrink v_mac_legacy_f32 and v_fmac_legacy_f32 · 2b548b18
  Jay Foad authored Nov 01, 2021
  2b548b18
- [AArch64] Enablement of Cortex-X2 · 0b83a18a
  Mubashar Ahmad authored Nov 01, 2021
  0b83a18a
- [CostModel][X86] Remove old FIXME comments for AVX512F vector splitting · 6fc50e53
  Simon Pilgrim authored Nov 01, 2021
  6fc50e53
- [X86][AVX] Prefer VINSERTF128 over VPERM2F128 for 128->256 subvector concatenations · fd485d8c
  Simon Pilgrim authored Nov 01, 2021
  fd485d8c
- [AArch64] Remove unused declaration hasSwiftExtendedFrame (NFC) · 476e1ee3
  Kazu Hirata authored Oct 31, 2021
  
  476e1ee3
- [PowerPC] turn off chain commoning by default. · eeed1545
  Chen Zheng authored Nov 01, 2021
  
  eeed1545
- [CSKY] First patch to construct codegen infra and generate first add instruction · cf78715c
  Zi Xuan Wu authored Nov 01, 2021
  cf78715c
Oct 31, 2021
- [RISCV] Expand scalable vector bswap. Fix crash for bitreverse. · ada54585
  Craig Topper authored Oct 29, 2021
  ada54585
- [CodeGen, Target] Use MachineBasicBlock::terminators (NFC) · 72710af2
  Kazu Hirata authored Oct 31, 2021
  
  72710af2