Commits · eee52b5090de793274bbb589b5a9ac64808f01bc · Roger Ferrer / llvm-epi · GitLab

Jul 27, 2018
- AMDGPU: Fix code size for return_to_epilog pseudo · 0183c56c
  Matt Arsenault authored Jul 27, 2018
  0183c56c
- AMDGPU/GlobalISel: Fix crash in regbankselect on non-power-of-2 types · e9bdc5f1
  Tom Stellard authored Jul 27, 2018
  e9bdc5f1
- [X86] Remove an unnecessary 'if' that prevented treating INT64_MAX and... · 561e298e
  Craig Topper authored Jul 27, 2018
  561e298e
- [X86] Add matching for another pattern of PMADDWD. · e364baa8
  Craig Topper authored Jul 27, 2018
  e364baa8
- [X86] When removing sign extends from gather/scatter indices, make sure we... · f7bc5502
  Craig Topper authored Jul 27, 2018
  f7bc5502
- [SelectionDAGBuilder] Add masked loads to PendingLoads rather than calling DAG.setRoot. · 1a40a065
  Craig Topper authored Jul 26, 2018
  1a40a065
Jul 26, 2018
- [AMDGPU] Fix VGPR spills where offset doesn't fit in 12 bits · eb1f75d5
  Scott Linder authored Jul 26, 2018
  eb1f75d5
- [RISCV] Add support for _interrupt attribute · 2e4106b7
  Ana Pazos authored Jul 26, 2018
  2e4106b7
- MacroFusion: Fix macro fusion with ExitSU failing in top-down scheduling · 09810c92
  Matthias Braun authored Jul 26, 2018
  09810c92
- [DAGCombine] optimizeSetCCOfSignedTruncationCheck(): handle ule,ugt CondCodes. · 41ba5c14
  Roman Lebedev authored Jul 26, 2018
  41ba5c14
- [mips] Sign extend i32 return values on MIPS64 · 4a612d4b
  Stefan Maksimovic authored Jul 26, 2018
  4a612d4b
- Revert "[COFF] Use comdat shared constants for MinGW as well" · 9dafd6f6
  Martin Storsjö authored Jul 26, 2018
  9dafd6f6
- [X86] Don't use CombineTo to skip adding new nodes to the DAGCombiner worklist in combineMul. · 4e687d5b
  Craig Topper authored Jul 26, 2018
  4e687d5b
- [GlobalISel] Fall back to SDISel for swifterror/swiftself attributes. · fdd089aa
  Amara Emerson authored Jul 26, 2018
  fdd089aa
- bpf: new option -bpf-expand-memcpy-in-order to expand memcpy in order · 71d81e5c
  Yonghong Song authored Jul 25, 2018
  71d81e5c
Jul 25, 2018
- [SelectionDAG] try to convert funnel shift directly to rotate if legal · 215dcbf4
  Sanjay Patel authored Jul 25, 2018
  215dcbf4
- [AArch, PowerPC] add more tests for legal rotate ops; NFC · f94c4c84
  Sanjay Patel authored Jul 25, 2018
  f94c4c84
- [COFF] Use comdat shared constants for MinGW as well · ff33a95e
  Martin Storsjö authored Jul 25, 2018
  ff33a95e
- [COFF] Hoist constant pool handling from X86AsmPrinter into AsmPrinter · d2662c32
  Martin Storsjö authored Jul 25, 2018
  d2662c32
- [ARM] Prefer lsls+lsrs over lsls+ands or lsrs+ands in Thumb1. · 733f4ed1
  Eli Friedman authored Jul 25, 2018
  733f4ed1
- Fix corruption of result number in LegalizeVectorOps.cpp · 5f75371c
  Ulrich Weigand authored Jul 25, 2018
  5f75371c
- [AMDGPU] Use AssumptionCacheTracker in the divrem32 expansion · 7e7268ac
  Stanislav Mekhanoshin authored Jul 25, 2018
  7e7268ac
- [Hexagon] Properly scale bit index when extracting elements from vNi1 · 4e07509d
  Krzysztof Parzyszek authored Jul 25, 2018
  4e07509d
- [MIPS GlobalISel] Lower pointer arguments · 58c02100
  Petar Jovanovic authored Jul 25, 2018
  58c02100
- Fix PR34170: Crash on inline asm with 64bit output in 32bit GPR · 768d6ce4
  Thomas Preud'homme authored Jul 25, 2018
  768d6ce4
- [X86] Autogenerate complete checks and fix a failure introduced in r337875. · d9fa8147
  Craig Topper authored Jul 25, 2018
  d9fa8147
- [x86/SLH] Teach the x86 speculative load hardening pass to harden · 7024921c
  Chandler Carruth authored Jul 25, 2018
  7024921c
- [X86] Use a shift plus an lea for multiplying by a constant that is a power of 2 plus 2/4/8. · fc501a92
  Craig Topper authored Jul 25, 2018
  fc501a92
- [X86] Expand mul by pow2 + 2 using a shift and two adds similar to what we do for pow2 - 2. · 5be253d9
  Craig Topper authored Jul 25, 2018
  5be253d9
- [X86] Use a two lea sequence for multiply by 37, 41, and 73. · 56c104f1
  Craig Topper authored Jul 24, 2018
  56c104f1
- [X86] Add test cases for multiply by 37, 41, and 73. · b5342b59
  Craig Topper authored Jul 24, 2018
  b5342b59
- [X86] Change multiply by 26 to use two multiplies by 5 and an add instead of... · f8fcee70
  Craig Topper authored Jul 24, 2018
  f8fcee70
Jul 24, 2018
- [X86] When expanding a multiply by a negative of one less than a power of 2,... · 5ddc0a2b
  Craig Topper authored Jul 24, 2018
  5ddc0a2b
- [X86] Generalize the multiply by 30 lowering to generic multipy by power 2 minus 2. · 6d29891b
  Craig Topper authored Jul 24, 2018
  6d29891b
- [WebAssembly] Add tests for weaker memory consistency orderings · 8daef075
  Heejin Ahn authored Jul 24, 2018
  8daef075
- [X86] Change multiply by 19 to use (9 * X) * 2 + X instead of (5 * X) * 4 - 1. · 86d6320b
  Craig Topper authored Jul 24, 2018
  86d6320b
- [X86] Add test case to show failure to combine away negates that may be... · 1296c622
  Craig Topper authored Jul 24, 2018
  1296c622
- [mips] Fix local dynamic TLS with Sym64 · 28ded4ee
  Simon Atanasyan authored Jul 24, 2018
  28ded4ee
- [ARM] Disable ARMCodeGenPrepare by default · 8b93e82c
  Sam Parker authored Jul 24, 2018
  8b93e82c
- [x86] Clean up and convert test to use generated CHECK lines. · a25aca21
  Chandler Carruth authored Jul 24, 2018
  a25aca21