Commits · fedc9549d50d80f74169ecce4d0d0648a62249f0 · Roger Ferrer / llvm-epi · GitLab

Sep 21, 2020
- [SCEV] Recognize @llvm.usub.sat as `%x - (umin %x, %y)` · fedc9549
  Roman Lebedev authored Sep 21, 2020
  fedc9549
- [NFC][SCEV] Add tests for @llvm.*.sat intrinsics · 0592de55
  Roman Lebedev authored Sep 21, 2020
  
  0592de55
- [SCEV] Recognize @llvm.abs as smax(x, -x) · 1bb7ab8c
  Roman Lebedev authored Sep 21, 2020
  1bb7ab8c
- [NFC][SCEV] Add tests for @llvm.abs intrinsic · 83c2d10d
  Roman Lebedev authored Sep 21, 2020
  
  83c2d10d
- [SCEV] Add additional max BTC tests with loop guards. · 3cbdfe42
  Florian Hahn authored Sep 21, 2020
  
  3cbdfe42
- [ObjCARC][NewPM] Port objc-arc-contract to NPM · 024979b7
  Arthur Eubanks authored Sep 02, 2020
  024979b7
- [ARM][CMSE] Issue an error if passing arguments through memory across · 742250bf
  Momchil Velikov authored Sep 21, 2020
  742250bf
- [PowerPC] Add vector pair load/store instructions and vector pair register class · 1372e23c
  Baptiste Saleil authored Sep 21, 2020
  1372e23c
- [LoopSimplifyCFG][NewPM] Rename simplify-cfg -> loop-simplifycfg · 5249e6f2
  Arthur Eubanks authored Sep 18, 2020
  5249e6f2
- [CostModel][X86] Add some select shuffle costs tests for D87884 · 18a3ebcd
  Simon Pilgrim authored Sep 21, 2020
  
  18a3ebcd
- [SLP] Allow reordering of vectorization trees with reused instructions. · 3ff07fcd
  Alexey Bataev authored Sep 18, 2020
  3ff07fcd
- [Statepoints][ISEL] gc.relocate uniquification should be based on SDValue, not IR Value. · ee86688b
  Denis Antrushin authored Sep 15, 2020
  ee86688b
- [SVE] Change definition of reduction ISD nodes to have an SVE vector result type. · f3fa954b
  Paul Walker authored Sep 17, 2020
  f3fa954b
- [SVE] Use NEON for extract_vector_elt when the index is in range. · 64574552
  Paul Walker authored Sep 17, 2020
  64574552
- Recommit "[SCEV] Look through single value PHIs." · 11dccf8d
  Florian Hahn authored Sep 21, 2020
  11dccf8d
- [ARM] Select f32 constants with vmov.f16 · f4c5cadb
  David Green authored Sep 21, 2020
  f4c5cadb
- [llvm-readelf/obj] - Stop printing invalid names for unnamed section symbols. · 095f6fbb
  Georgii Rymar authored Sep 15, 2020
  095f6fbb
- [NFC][ARM] More tail predication tests. · 13c73632
  Sam Parker authored Sep 21, 2020
  13c73632
- [Test] Test auto-update · 98aed8aa
  Max Kazantsev authored Sep 21, 2020
  
  98aed8aa
- [CodeGen] Fixing inconsistent ABI mangling of vlaues in SelectionDAGBuilder · 53d238a9
  Lucas Prates authored Sep 17, 2020
  53d238a9
- [LSR] Preserve MSSA when using SplitCriticalEdge. · 57ae9bb9
  Florian Hahn authored Sep 20, 2020
  57ae9bb9
- [PowerPC] Pass nofpexcept flag to custom lowered constrained ops · 1d782c29
  Qiu Chaofan authored Sep 21, 2020
  1d782c29
- [XRay] Change mips to use version 2 sled (PC-relative address) · d0648568
  Fangrui Song authored Sep 19, 2020
  d0648568
- [llvm-profdata]Fix llvm-profdata crash on compact binary profile · a8b8a937
  wlei authored Sep 20, 2020
  a8b8a937
Sep 20, 2020
- [X86] Make reduceMaskedLoadToScalarLoad/reduceMaskedStoreToScalarStore work... · a74b1fab
  Craig Topper authored Sep 20, 2020
  a74b1fab
- [X86] Pre-commit test cases for D87863. NFC · c89b3af0
  Craig Topper authored Sep 20, 2020
  
  c89b3af0
- [X86] Stop reduceMaskedLoadToScalarLoad/reduceMaskedStoreToScalarStore from... · 4e8c0281
  Craig Topper authored Sep 20, 2020
  4e8c0281
- [X86] Add 32-bit command lines to masked_store.ll and masked_load.ll · 9b1c98c0
  Craig Topper authored Sep 20, 2020
  
  9b1c98c0
- [ARM] Constant fold VMOVrh · 29bd8ea1
  David Green authored Sep 20, 2020
  29bd8ea1
- [CVP] Additional tests for comparison with offset (NFC) · 1a272380
  Nikita Popov authored Sep 20, 2020
  1a272380
- [LVI] Get value range from mask comparison · 445db89b
  Nikita Popov authored Sep 20, 2020
  445db89b
- [CVP] Add tests for mask comparisons (NFC) · 91af6a78
  Nikita Popov authored Sep 20, 2020
  
  91af6a78
- [X86][SSE] Fold EXTEND_VECTOR_INREG(EXTRACT_SUBVECTOR(EXTEND(X),0)) -> EXTEND_VECTOR_INREG(X) · 0bfeede6
  Simon Pilgrim authored Sep 20, 2020
  
  0bfeede6
- [X86][SSE] Fold SIGN_EXTEND(SIGN_EXTEND_VECTOR_INREG(X)) -> SIGN_EXTEND_VECTOR_INREG(X) · bb0078e5
  Simon Pilgrim authored Sep 20, 2020
  bb0078e5
- [InstCombine] factorize left shifts of add/sub · 7903ae47
  Sanjay Patel authored Sep 20, 2020
  7903ae47
- [InstCombine] replace zombie unreachable values with 'undef' before erasing · cf75e832
  Sanjay Patel authored Sep 20, 2020
  cf75e832
- [X86][SSE] Fold EXTEND_VECTOR_INREG(EXTEND_VECTOR_INREG(X)) -> EXTEND_VECTOR_INREG(X) · 15c83060
  Simon Pilgrim authored Sep 20, 2020
  15c83060
- [X86][SSE] Enable ZERO_EXTEND_VECTOR_INREG shuffle combining on SSE41 targets. · a0c8793c
  Simon Pilgrim authored Sep 20, 2020
  a0c8793c
- [MemLoc] Support lllvm.memcpy.inline in MemoryLocation::getForArgument · 2990518b
  Dávid Bolvanský authored Sep 20, 2020
  2990518b
Sep 19, 2020
- [InstCombine] Regenerate test checks (NFC) · a2f9098f
  Nikita Popov authored Sep 19, 2020
  
  a2f9098f