Commits · d8aba75a768033c326613d85e8789703cb4565d2 · Roger Ferrer / llvm-epi · GitLab

May 07, 2021
- Internalize some cl::opt global variables or move them under namespace llvm · d8aba75a
  Fangrui Song authored May 07, 2021
  
  d8aba75a
- [libc++][ci] Run longer CI jobs first · 8002c5d6
  Louis Dionne authored May 07, 2021
  8002c5d6
- lit: revert 134b103f · d319005a
  Saleem Abdulrasool authored May 07, 2021
  d319005a
- [X86] AMD Zen 3: mark XMM/YMM (but not MMX!) reg moves as eliminatible in RegisterFile · b8701dc1
  Roman Lebedev authored May 07, 2021
  
  b8701dc1
- [X86] AMD Zen 3: MOVSX32rr32 is a zero-cycle move · 5b1610a2
  Roman Lebedev authored May 07, 2021
  5b1610a2
- [AArch64][ELF] Prefer to lower MC_GlobalAddress operands to .Lfoo$local · 6a2850f3
  Fangrui Song authored May 07, 2021
  6a2850f3
- [libFuzzer] Fix stack-overflow-with-asan.test. · f0941449
  Matt Morehouse authored May 07, 2021
  f0941449
- [LoopNest] Consider loop nest with inner loop guard using outer loop · 1006ac39
  Whitney Tsang authored May 07, 2021
  1006ac39
- [X86] combineXor - limit fold to non-opaque constants (PR50254) · f744723f
  Simon Pilgrim authored May 07, 2021
  f744723f
- [X86] AMD Zen 3: _REV variants of zero-cycles moves are also zero-cycles (PR50261) · 2819009b
  Roman Lebedev authored May 07, 2021
  2819009b
- [NFC][X86][MCA] AMD Zen3: add test for zero-cycle X87 move · a8e30e63
  Roman Lebedev authored May 07, 2021
  
  a8e30e63
- [libFuzzer] Fix stack overflow detection · 70cbc6db
  Sebastian Poeplau authored May 07, 2021
  70cbc6db
- [mlir][vector] add pattern to cast away leading unit dim for elementwise op · a970e69d
  thomasraoux authored May 06, 2021
  a970e69d
- [mlir][spirv] add support lowering of extract_slice to scalar type · 565ee6af
  thomasraoux authored May 06, 2021
  565ee6af
- BasicAA: Recognize inttoptr as isEscapeSource · bc302bfb
  Joseph Tremoulet authored May 07, 2021
  bc302bfb
- [AArch64] add test for missed vectorization; NFC · 0a6f11aa
  Sanjay Patel authored May 07, 2021
  0a6f11aa
- [libomptarget] Add support for target memory allocators to cuda RTL · a15f8589
  Joseph Huber authored May 06, 2021
  a15f8589
- [mlir][linalg] Remove redundant indexOp builder. · f31531a3
  Tobias Gysi authored May 07, 2021
  f31531a3
- [mlir] Rename BufferAliasAnalysis to BufferViewFlowAnalysis · faab8c14
  Tres Popp authored Apr 20, 2021
  faab8c14
- [PowerPC] Provide MMA builtins for compatibility · 25bbff63
  Ahsan Saghir authored May 04, 2021
  25bbff63
- [NFC][X86][MCA] AMD Zen3 Decrease iteration count in reg-move-elimination tests · 34de155f
  Roman Lebedev authored May 07, 2021
  34de155f
- [X86] AMD Zen 3: throughput for renameable XMM/YMM moves is 6 · 758c1733
  Roman Lebedev authored May 07, 2021
  758c1733
- [X86] AMD Zen 3: AVX YMM moves are zero-cycle · 715c0d0b
  Roman Lebedev authored May 07, 2021
  715c0d0b
- [X86] AMD Zen 3: AVX XMM moves are zero-cycle · ee020b93
  Roman Lebedev authored May 07, 2021
  ee020b93
- [X86] AMD Zen 3: SSE XMM moves are zero-cycle · 9db42038
  Roman Lebedev authored May 07, 2021
  9db42038
- [NFC][X86][MCA] AMD Zen 3: Add tests for renameable AVX YMM moves · 0d961fbd
  Roman Lebedev authored May 07, 2021
  
  0d961fbd
- [NFC][X86][MCA] AMD Zen 3: Add tests for renameable AVX XMM moves · bcbfc22f
  Roman Lebedev authored May 07, 2021
  
  bcbfc22f
- [NFC][X86][MCA] AMD Zen 3: Add tests for renameable SSE XMM moves · cbabe4f4
  Roman Lebedev authored May 07, 2021
  
  cbabe4f4
- [X86] AMD Zen 3: throughput for renameable GPR moves is 6 · d8c62025
  Roman Lebedev authored May 07, 2021
  d8c62025
- [NFC][X86] AMD Zen 3: move sched classes for renameables moves togeter · c3cd8ed0
  Roman Lebedev authored May 07, 2021
  
  c3cd8ed0
- [NFC][X86][MCA] Increase iteration count in reg move elimination tests · e6d688ec
  Roman Lebedev authored May 07, 2021
  e6d688ec
- [libc++] [test] Test that unordered_*::swap/move/assign does not invalidate iterators. · f42355e1
  Arthur O'Dwyer authored May 06, 2021
  f42355e1
- [libc++] [test] Simplify arithmetic in list.special/swap.pass.cpp. NFCI. · a1f75bf0
  Arthur O'Dwyer authored May 06, 2021
  a1f75bf0
- [libc++] [test] Test that list::swap/move/move-assign does not invalidate iterators. · 8935c844
  Arthur O'Dwyer authored May 06, 2021
  8935c844
- Reapply "[DebugInfo] Drop DBG_VALUE_LISTs with an excessive number of debug operands" · 7bc1dd11
  Stephen Tozer authored May 07, 2021
  7bc1dd11
- [CodeGen] Ensure UserValue::getDebugLoc() and UserLabel::getDebugLoc()... · c9d4b417
  Simon Pilgrim authored May 07, 2021
  c9d4b417
- [DAG] Ensure all SD classes consistently return a const reference with getDebugLoc(). NFCI. · dd21c6b8
  Simon Pilgrim authored May 07, 2021
  dd21c6b8
- Retire TargetRegisterInfo::getSpillAlignment · 6248d111
  Benjamin Kramer authored May 07, 2021
  6248d111
- [AMDGPU] Restrict immediate scratch offsets · 13c03162
  Sebastian Neubauer authored May 03, 2021
  13c03162
- AMDGPU: Correct const_index_stride for wave 32 for PAL ABI · 606d4e80
  David Stuttard authored Feb 24, 2020
  606d4e80