Commits · e3959de26867d65410d4d5e1fc8c9c8f5bb2951a · Roger Ferrer / llvm-epi · GitLab

Mar 25, 2019
- AMDGPU: Add support for cross address space synchronization scopes · 51809cbc
  Konstantin Zhuravlyov authored Mar 25, 2019
  51809cbc
- [llvm] Prevent duplicate files in debug line header in dwarf 5. · 312ab058
  Ali Tamur authored Mar 25, 2019
  312ab058
- [SLPVectorizer] Merge reorderAltShuffleOperands into reorderInputsAccordingToOpcode · 6f96795b
  Simon Pilgrim authored Mar 25, 2019
  6f96795b
- [SelectionDAG] Add icmp UNDEF handling to SelectionDAG::FoldSetCC · 167af1ba
  Simon Pilgrim authored Mar 25, 2019
  167af1ba
- [x86] add another vector zext test; NFC · f49e33e2
  Sanjay Patel authored Mar 25, 2019
  f49e33e2
- MISched: Don't schedule regions with 0 instructions · b27e4974
  Matt Arsenault authored Mar 25, 2019
  b27e4974
- [llvm-objcopy]Preserve data in segments not covered by sections · 1f448149
  James Henderson authored Mar 25, 2019
  1f448149
- [SLPVectorizer] Update file missed in rL356913 · 77749567
  Simon Pilgrim authored Mar 25, 2019
  77749567
- [x86] add tests for vector zext; NFC · 76c1ef3d
  Sanjay Patel authored Mar 25, 2019
  76c1ef3d
- [SLPVectorizer] reorderInputsAccordingToOpcode - remove non-Instruction canonicalization · ff3abef3
  Simon Pilgrim authored Mar 25, 2019
  ff3abef3
- [RegAlloc] Simplify MIR test · 0e75e21e
  Jonas Paulsson authored Mar 25, 2019
  0e75e21e
- [MIPS GlobalISel] Select copy for arguments from FPRBRegBank · a034a64f
  Petar Avramovic authored Mar 25, 2019
  a034a64f
- [MIPS GlobalISel] Add floating point register bank · 3dfa368d
  Petar Avramovic authored Mar 25, 2019
  3dfa368d
- [MIPS GlobalISel] Lower float and double arguments in registers · 5a457e08
  Petar Avramovic authored Mar 25, 2019
  5a457e08
- [llvm-readobj] Separate `Symbol Version` dumpers into `LLVM style` and `GNU style` · ea16be1c
  Xing GUO authored Mar 25, 2019
  ea16be1c
- [ARM GlobalISel] 64-bit memops should be aligned · 254b11a0
  Diana Picus authored Mar 25, 2019
  254b11a0
- Revert r356688 "[X86] Don't avoid folding multiple use sign extended 8-bit... · 7c2554dd
  Craig Topper authored Mar 25, 2019
  7c2554dd
Mar 24, 2019
- [X86][SSE41] Start shuffle combining from ZERO_EXTEND_VECTOR_INREG (PR40685) · 87d4ab8b
  Simon Pilgrim authored Mar 24, 2019
  87d4ab8b
- [X86] Remove icmp undef from reduced tests · 4465a765
  Simon Pilgrim authored Mar 24, 2019
  4465a765
- [X86][AVX] Start shuffle combining from ZERO_EXTEND_VECTOR_INREG (PR40685) · a71c0ed4
  Simon Pilgrim authored Mar 24, 2019
  a71c0ed4
- Recommit r356738 "[llvm-objcopy] - Implement replaceSectionReferences for GroupSection class." · 27257171
  George Rimar authored Mar 24, 2019
  27257171
- [x86] improve the default expansion of uaddsat/usubsat · 7d676dfd
  Sanjay Patel authored Mar 24, 2019
  7d676dfd
- [llvm-objcopy] - Report SHT_GROUP sections with invalid alignment. · 0a5d4b84
  George Rimar authored Mar 24, 2019
  0a5d4b84
Mar 23, 2019
- [X86][SLP] Show example of failure to uniformly commute splats for 'alt' shuffles. · 9eb0de85
  Simon Pilgrim authored Mar 23, 2019
  9eb0de85
Mar 22, 2019
- [pdb] Add -type-stats and sort stats by descending size · e6a81b9b
  Reid Kleckner authored Mar 22, 2019
  e6a81b9b
- Revert "[llvm-readobj] Separate `Symbol Version` dumpers into `LLVM style` and `GNU style`" · 8316ea42
  Douglas Yung authored Mar 22, 2019
  8316ea42
- [ARM] Don't form "ands" when it isn't scheduled correctly. · b906bba5
  Eli Friedman authored Mar 22, 2019
  b906bba5
- [X86] Use xmm registers to implement 64-bit popcnt on 32-bit targets if... · ce1ed55a
  Craig Topper authored Mar 22, 2019
  ce1ed55a
- [X86] Use movq for i64 atomic load on 32-bit targets when sse2 is enable · 1ffd8e81
  Craig Topper authored Mar 22, 2019
  1ffd8e81
- Fix non-determinism in Reassociate caused by address coincidences · ef8761fd
  Daniel Sanders authored Mar 22, 2019
  ef8761fd
- [AArch64, ARM] Add support for Exynos M5 · 4a7739b6
  Evandro Menezes authored Mar 22, 2019
  4a7739b6
- [SLP] fix variables names in test; NFC · a0aaa11a
  Sanjay Patel authored Mar 22, 2019
  a0aaa11a
- IR: Support parsing numeric block ids, and emit them in textual output. · c0e6b8ac
  James Y Knight authored Mar 22, 2019
  c0e6b8ac
- [X86] Regenerate powi tests to include i686 x87/sse targets · aea9db9d
  Simon Pilgrim authored Mar 22, 2019
  aea9db9d
- [X86] Add PR13897 test case (i128 mul on i686) · 08380afa
  Simon Pilgrim authored Mar 22, 2019
  08380afa
- [X86] lowerShuffleAsBitMask - ensure float bit masks are the correct width (PR41203) · 564392d7
  Simon Pilgrim authored Mar 22, 2019
  564392d7
- [Tests] Add masked.gather tests for non-constant masks + speculation possibilities · d627048c
  Philip Reames authored Mar 22, 2019
  d627048c
- [ConstantFolding] Fix GetConstantFoldFPValue to avoid cast overflow. · bdf0230c
  Bixia Zheng authored Mar 22, 2019
  bdf0230c
- [tests] Add a generic masked.gather test to show sometimes we can't transform · f032e85d
  Philip Reames authored Mar 22, 2019
  f032e85d
- [tests] Add tests for converting masked.load to load speculatively · e234fd61
  Philip Reames authored Mar 22, 2019
  e234fd61