Commits · e781620dee4415018bdf3769a97c2ec31a85ff68 · Roger Ferrer / llvm-epi · GitLab

Jan 18, 2022
- [LoopVectorize][AArch64] Use get.active.lane.mask intrinsic when SVE is enabled · e781620d
  David Sherwood authored Dec 06, 2021
  e781620d
- [VE] vp_select+vectorBinOp passthru isel and tests · 43994e9a
  Simon Moll authored Jan 18, 2022
  43994e9a
- [LVI] Handle implication from icmp of trunc (PR51867) · 3ec7f46e
  Nikita Popov authored Jan 18, 2022
  3ec7f46e
- [LegalizeTypes][VP] Add widening support for vp.reduce.* · b7bf96a2
  Victor Perez authored Jan 18, 2022
  b7bf96a2
- [CVP] Add tests for icmp of trunc implication (NFC) · 0861fbe1
  Nikita Popov authored Jan 18, 2022
  
  0861fbe1
- Revert "[DAG] Extend SearchForAndLoads with any_extend handling" · f4615fea
  Hans Wennborg authored Jan 18, 2022
  f4615fea
- [LVI] Handle commuted SPF min/max operands · 9e68557e
  Nikita Popov authored Jan 18, 2022
  9e68557e
- [LVI] Compute SPF range even if one operands is overdefined · d15823e3
  Nikita Popov authored Jan 17, 2022
  d15823e3
- [LegalizeTypes][VP] Add splitting support for vp.reduction.* · fd1dce35
  Victor Perez authored Jan 17, 2022
  fd1dce35
- [LVI] Consistently intersect assumes · 202d590a
  Nikita Popov authored Jan 18, 2022
  202d590a
- [CVP] Add test for select with assumes (NFC) · f9f865d8
  Nikita Popov authored Jan 18, 2022
  f9f865d8
- [DwarfDebug] Restore code that make comments stay aligned in DwarfDebug::emitDebugLocEntry · 65fbe38f
  Bjorn Pettersson authored Jan 16, 2022
  65fbe38f
- Revert "[CodeGen][AArch64] Ensure isSExtCheaperThanZExt returns true for negative constants" · f4515ab8
  David Sherwood authored Jan 18, 2022
  f4515ab8
- [SimplifyCFG] Fix assertion failure when reusing table switch comparison · 4e1c2077
  pvellien authored Jan 18, 2022
  4e1c2077
- [AMDGPU] Disable optimizeEndCf at -O0 · 56a5d788
  Christudasan Devadasan authored Jan 07, 2022
  56a5d788
- [AMDGPU] Autogenerate wqm.ll · ed4d8fda
  Carl Ritson authored Jan 18, 2022
  ed4d8fda
- [RISCV] Add patterns for vector widening floating-point multiply · 1090000b
  Jianjian GUAN authored Jan 18, 2022
  1090000b
- [RISCV] Provide VLOperand in td. · ec9cb3a7
  Han-Kuan Chen authored Jan 17, 2022
  ec9cb3a7
- [RISCV] Add patterns for vector widening floating-point add/subtract instructions · c29d6c41
  Jianjian GUAN authored Jan 17, 2022
  c29d6c41
- [SDAG] add demanded bits transform for bswap · ba6485e2
  Sanjay Patel authored Jan 17, 2022
  ba6485e2
- [GlobalOpt] Generalize malloc-to-global for any allocation function · 26049b8c
  Philip Reames authored Jan 17, 2022
  26049b8c
- [test] precommit new test for D117503 · 30715365
  Philip Reames authored Jan 17, 2022
  
  30715365
Jan 17, 2022
- [RISCV] When expanding CONCAT_VECTORS, don't create INSERT_SUBVECTORS for undef subvectors. · 116af698
  Craig Topper authored Jan 17, 2022
  116af698
- [RISCV] Legalize fixed length (insert_subvector undef, X, 0) to a scalable insert. · 9c410838
  Craig Topper authored Jan 17, 2022
  9c410838
- [PowerPC][AIX] Fallback to DAG-ISEL if global has toc-data attribute. · 10d3bf95
  Sean Fertile authored Jan 17, 2022
  10d3bf95
- [AArch64][x86] add tests for bswap demanded bits; NFC · 5fb39f09
  Sanjay Patel authored Jan 17, 2022
  
  5fb39f09
- [LoopDeletion] Revert 3af8a11b and add test coverage for breakage · 523573e9
  Philip Reames authored Jan 17, 2022
  523573e9
- [X86] Add test case for PR53247 · fea85d32
  Simon Pilgrim authored Jan 17, 2022
  fea85d32
- [Aarch64] Customer lowering of COPYSIGN to SIMD should check for NEON availability · 6f61fe7d
  Akshay Kumar authored Jan 17, 2022
  6f61fe7d
- [AMDGPU][GFX10][MC] Updated symbolic names of internal HW registers · c7ca4c63
  Dmitry Preobrazhensky authored Jan 17, 2022
  c7ca4c63
- [AMDGPU][MC] Corrected disassembly of s_waitcnt · b5fb7e48
  Dmitry Preobrazhensky authored Jan 17, 2022
  b5fb7e48
- [DebugInfo] ValueMapper impl for DIArgList respects IgnoreMissingLocals · 32417b32
  Stephen Tozer authored Jan 17, 2022
  32417b32
- [AArch64] Revive optimize add/sub with immediate through MIPeepholeOpt · e6698f09
  Micah Weston authored Jan 17, 2022
  e6698f09
- [test] Use opt -passes syntax in DeadArgElim lit tests. NFC · e76ecbb0
  Bjorn Pettersson authored Jan 16, 2022
  e76ecbb0
- [AIX] Support of Big archive (read) · 2164c543
  zhijian authored Jan 17, 2022
  2164c543
- Revert "[AIX] Support of Big archive (read)" · 76f1c396
  zhijian authored Jan 17, 2022
  76f1c396
- [MCA][X86] Add missing zero-idioms test file coverage · 8ea57920
  Simon Pilgrim authored Jan 17, 2022
  8ea57920
- [ConstantFold] Don't fold load from non-byte-sized vector · f104cc38
  Nikita Popov authored Jan 17, 2022
  f104cc38
- [InstCombine] FP with reassoc FMF: (X * C) + X --> X * (MulC + 1.0) · 4cdf30d9
  Sanjay Patel authored Jan 17, 2022
  4cdf30d9
- [InstCombine] add tests for fadd+fmul factorization fold; NFC · bcf5e687
  Sanjay Patel authored Jan 17, 2022
  bcf5e687