Commits · ef96a8ecd6308cc31e43515c2ce0bae6ffbdbbc6 · Roger Ferrer / llvm-epi · GitLab

Jan 26, 2015
- [mips] Enable arithmetic and binary operations for the i128 data type. · ef96a8ec
  Vasileios Kalintiris authored Jan 26, 2015
  ef96a8ec
- [mips] Add tests for bitwise binary and integer arithmetic operators. · 2ed214f3
  Vasileios Kalintiris authored Jan 26, 2015
  2ed214f3
- Revert "[mips] Fix assertion on i128 addition/subtraction on MIPS64" · 30c5451f
  Vasileios Kalintiris authored Jan 26, 2015
  30c5451f
- [X86] Change comparision immediate type to i8 in test cases for AVX512... · d1e1d106
  Craig Topper authored Jan 25, 2015
  d1e1d106
- [X86] Use i8 immediate for comparison type on AVX512 packed integer... · 29f2e951
  Craig Topper authored Jan 25, 2015
  29f2e951
Jan 25, 2015
- [PowerPC] Revert ppc64le-aggregates.ll test changes from r227053 · 258ac3cc
  Bill Schmidt authored Jan 25, 2015
  258ac3cc
- [PowerPC] Reset the baseline for ppc64le to be equivalent to pwr8 · 279cabb4
  Bill Schmidt authored Jan 25, 2015
  279cabb4
- AVX-512: Changes in operations on masks registers for KNL and SKX · 1a603b3f
  Elena Demikhovsky authored Jan 25, 2015
  1a603b3f
Jan 24, 2015
- bpf: add missing lit.local.cfg · 8a3d0c15
  Alexei Starovoitov authored Jan 24, 2015
  8a3d0c15
- BPF backend · e4c8c807
  Alexei Starovoitov authored Jan 24, 2015
  e4c8c807
- [mips] Fix assertion on i128 addition/subtraction on MIPS64 · 5a9225b2
  Daniel Sanders authored Jan 24, 2015
  5a9225b2
- [DAG] Fix wrong canonicalization performed on shuffle nodes. · 8381475a
  Andrea Di Biagio authored Jan 24, 2015
  8381475a
- [AArch64][LoadStoreOptimizer] Form LDPSW when possible. · 29f55339
  Quentin Colombet authored Jan 24, 2015
  29f55339
- R600/SI: Emit .hsa.version section for amdhsa OS · edd188c4
  Tom Stellard authored Jan 23, 2015
  edd188c4
- Fix assertion when C++ EH filters are present in functions using SEH · 3d4638b3
  Reid Kleckner authored Jan 23, 2015
  3d4638b3
Jan 23, 2015
- [x86] Combine x86mmx/i64 to v2i64 conversion to use scalar_to_vector · 56567f91
  Bruno Cardoso Lopes authored Jan 23, 2015
  56567f91
- R600/SI: Move i64 -> v2i32 load promotion into AMDGPUDAGToDAGISel::Select() · 20f6c073
  Tom Stellard authored Jan 23, 2015
  20f6c073
- Classify functions by EH personality type rather than using the triple · 5cc1569c
  Reid Kleckner authored Jan 23, 2015
  5cc1569c
- This patch fixes issue with lowering below mentioned pattern :- · f1d7050a
  Jyoti Allur authored Jan 23, 2015
  f1d7050a
- [x86] Change u8imm operands to always print as unsigned. This makes shuffle... · 0271d10d
  Craig Topper authored Jan 23, 2015
  0271d10d
- R600: Try to use lower types for 64bit division if possible · 5f715d36
  Jan Vesely authored Jan 22, 2015
  5f715d36
Jan 22, 2015
- [X86][AVX] Added (V)MOVDDUP / (V)MOVSLDUP / (V)MOVSHDUP memory folding + tests. · 7e6d573e
  Simon Pilgrim authored Jan 22, 2015
  7e6d573e
- Line endings fixes. NFC. · c976e8ee
  Simon Pilgrim authored Jan 22, 2015
  c976e8ee
- [X86][SSE] Simplified PSUBUS tests · 507b37fc
  Simon Pilgrim authored Jan 22, 2015
  507b37fc
- Intrinsics: introduce llvm_any_ty aka ValueType Any · 75a4f35b
  Ramkumar Ramachandra authored Jan 22, 2015
  75a4f35b
- merge consecutive stores of extracted vector elements (PR21711) · 37c41c1d
  Sanjay Patel authored Jan 22, 2015
  37c41c1d
- [DAGCombine] Produce better code for constant splats · 25e34d11
  Michael Kuperstein authored Jan 22, 2015
  25e34d11
- Revert r226811, MSVC accepts code sane compilers don't. · ff740320
  Michael Kuperstein authored Jan 22, 2015
  ff740320
- [DAGCombine] Produce better code for constant splats · 84fad3e5
  Michael Kuperstein authored Jan 22, 2015
  84fad3e5
- Fixed a bug in type legalizer for masked load/store intrinsics. · 150d9f31
  Elena Demikhovsky authored Jan 22, 2015
  150d9f31
- Fixed a bug in narrowing store operation. · 9c26462a
  Elena Demikhovsky authored Jan 22, 2015
  9c26462a
- SEH: Finish writing the catch-all test case · 5eb6ade3
  Reid Kleckner authored Jan 22, 2015
  5eb6ade3
- Win64 SEH: Emit the constant 1 for catch-all into xdata · f690f505
  Reid Kleckner authored Jan 22, 2015
  f690f505
- [X86][SSE] Missing SSE/AVX1 memory folding integer instructions · 5fa0fb23
  Simon Pilgrim authored Jan 21, 2015
  5fa0fb23
- DAGCombine: fold (or (and X, M), (and X, N)) -> (and X, (or M, N)) · 3007ba0a
  Tim Northover authored Jan 21, 2015
  3007ba0a
Jan 21, 2015
- R600: Add checks for urem/srem by a constant · b45c78bc
  Matt Arsenault authored Jan 21, 2015
  b45c78bc
- [X86][SSE] Added support for SSE3 lane duplication shuffle instructions · b16b09b1
  Simon Pilgrim authored Jan 21, 2015
  b16b09b1
- R600: Add missing tests for i64 srem · d9987c7b
  Matt Arsenault authored Jan 21, 2015
  d9987c7b
- Fix load-store optimizer on thumbv4t · 229eb4ca
  Jon Roelofs authored Jan 21, 2015
  229eb4ca
- R600/SI: Custom lower fround · b0055488
  Matt Arsenault authored Jan 21, 2015
  b0055488