Commits · eea21da12a16d6f81a9cf213c988ab058c6b2722 · Roger Ferrer / llvm-epi · GitLab

Nov 20, 2018
- [InstructionSimplify] Add support for saturating add/sub · eea21da1
  Sanjay Patel authored Nov 20, 2018
  eea21da1
- [ConstantFolding] Add support for saturating add/sub · efc3d1df
  Sanjay Patel authored Nov 20, 2018
  efc3d1df
- [AMDGPU] Regenerate weird stores tests. · bac49ac4
  Simon Pilgrim authored Nov 20, 2018
  bac49ac4
- [LoopSink] Add preheader to alias set · c21fba1b
  Guozhi Wei authored Nov 20, 2018
  c21fba1b
- [PatternMatch] Handle undef vectors consistently · f5ead29b
  Sanjay Patel authored Nov 20, 2018
  f5ead29b
- [AArch64, x86] add tests for shift-not (PR39657); NFC · 8aeffd8c
  Sanjay Patel authored Nov 20, 2018
  8aeffd8c
- [DAGCombine] Add calls to SimplifyDemandedVectorElts from visitINSERT_SUBVECTOR (PR37989) · 37351059
  Simon Pilgrim authored Nov 20, 2018
  37351059
- [PowerPC] Add Itineraries for STWU/STWUX etc · 9a0ed200
  Jinsong Ji authored Nov 20, 2018
  9a0ed200
- [PowerPC][NFC]Add testcase for STWU scheduling check · 42c13c22
  Jinsong Ji authored Nov 20, 2018
  42c13c22
- [X86][SSE] Add computeKnownBits/ComputeNumSignBits support for PACKSS/PACKUS instructions. · ee8b96f2
  Simon Pilgrim authored Nov 20, 2018
  ee8b96f2
- [TargetLowering] Improve SimplifyDemandedVectorElts/SimplifyDemandedBits support · b356d046
  Simon Pilgrim authored Nov 20, 2018
  b356d046
- [X86][SSE] Lower immediately to PACKUS instead of VECTOR_SHUFFLE. · a6fb85ff
  Simon Pilgrim authored Nov 20, 2018
  a6fb85ff
- [X86][SSE] Add SimplifyDemandedVectorElts support for PACKSS/PACKUS instructions. · 7198506b
  Simon Pilgrim authored Nov 20, 2018
  7198506b
- [X86] Preserve undef information when creating a punpckl/hbw from a v16i8... · 17fa42a6
  Craig Topper authored Nov 20, 2018
  17fa42a6
- [X86] Replace more calls to getZeroVector with regular getConstant. · c733c7bf
  Craig Topper authored Nov 20, 2018
  c733c7bf
- Recommit "[LoopSimplifyCFG] Teach LoopSimplifyCFG to constant-fold branches and switches" · c04b5307
  Max Kazantsev authored Nov 20, 2018
  c04b5307
- [PowerPC] Don't combine to bswap store on 1-byte truncating store · 9b393909
  Nemanja Ivanovic authored Nov 20, 2018
  9b393909
- [SelectionDAG] Compute known bits and num sign bits for live out vector... · 4954c664
  Craig Topper authored Nov 20, 2018
  4954c664
- [ExecutionEngine][Interpreter] Fix out-of-bounds array access. · a4d5b34f
  Lang Hames authored Nov 20, 2018
  a4d5b34f
- [CodeView] Don't print PointerAttributes when dumping. · b35e1d7d
  Zachary Turner authored Nov 20, 2018
  b35e1d7d
- Implement computeKnownBits for scalar_to_vector · 54ebfe8a
  Stanislav Mekhanoshin authored Nov 19, 2018
  54ebfe8a
Nov 19, 2018
- [X86] Add test case to show missed opportunity to use a single pmuludq to... · dbe34736
  Craig Topper authored Nov 19, 2018
  dbe34736
- AMDGPU: Fix V_FMA_F16 selection on GFX9 · 700b1ef5
  Konstantin Zhuravlyov authored Nov 19, 2018
  700b1ef5
- Revert "[LoopSimplifyCFG] Teach LoopSimplifyCFG to constant-fold branches and switches" · fdd9b4fc
  Benjamin Kramer authored Nov 19, 2018
  fdd9b4fc
- [AMDGPU] Restored selection of scalar_to_vector (v2x16) · 8bafbae8
  Stanislav Mekhanoshin authored Nov 19, 2018
  8bafbae8
- [InstCombine] Set debug loc on `mergeStoreIntoSuccessor` phi · 238533ec
  Vedant Kumar authored Nov 19, 2018
  238533ec
- [TargetLowering] expandFP_TO_UINT - improve fp16 support · de3605f5
  Simon Pilgrim authored Nov 19, 2018
  de3605f5
- [X86][SSE] Remove unnecessary bit-and in pshufb vector ctlz (PR39703) · c4861ab1
  Simon Pilgrim authored Nov 19, 2018
  c4861ab1
- [X86] Attempt to improve v32i8/v64i8 multiply lowering by applying the v16i8... · 311bbcd5
  Craig Topper authored Nov 19, 2018
  311bbcd5
- [ARM] Attempt to fix arm selfhost bots after rL347191 · 1c803f59
  Sam Parker authored Nov 19, 2018
  1c803f59
- [AMDGPU] Convert insert_vector_elt into set of selects · 054f8101
  Stanislav Mekhanoshin authored Nov 19, 2018
  054f8101
- [WebAssembly] replaced .param/.result by .functype · 49482f82
  Wouter van Oortmerssen authored Nov 19, 2018
  49482f82
- [SelectionDAG] simplify vector select with undef operand(s) · b25adf5e
  Sanjay Patel authored Nov 19, 2018
  b25adf5e
- Revert "[LICM] Make LICM able to hoist phis" · 2cad359c
  Benjamin Kramer authored Nov 19, 2018
  2cad359c
- [LV] Avoid vectorizing unsafe dependencies in uniform address · 5e9215f0
  Anna Thomas authored Nov 19, 2018
  5e9215f0
- [Hexagon] make test immune to improvements in undef simplification · 08c0a0ac
  Sanjay Patel authored Nov 19, 2018
  08c0a0ac
- [x86] add/make tests immune to improvements in undef simplification · 60abc29b
  Sanjay Patel authored Nov 19, 2018
  60abc29b
- [SelectionDAG] simplify select FP with undef condition · a1dca355
  Sanjay Patel authored Nov 19, 2018
  a1dca355
- [x86] add test for select FP with undef condition; NFC · 7a51bdcf
  Sanjay Patel authored Nov 19, 2018
  7a51bdcf
- Subject: [PATCH] [CodeGen] Add pass to combine interleaved loads. · fef3036d
  Martin Elshuber authored Nov 19, 2018
  fef3036d