Commits · a6054328e8b31217bec6e6d5dcbb28ca304d48d1 · Roger Ferrer / llvm-epi-0.8 · GitLab

This project is archived. Its data is read-only.

Sep 18, 2017
- [X86] Teach the execution domain fixing tables to use movlhps inplace of... · a6054328
  Craig Topper authored Sep 18, 2017
  a6054328
- [X86] Teach execution domain fixing to convert between FP and int unpack instructions. · 87f7381e
  Craig Topper authored Sep 18, 2017
  87f7381e
- [X86] Teach execution domain fixing to convert between VPERMILPS and VPSHUFD. · d4341920
  Craig Topper authored Sep 18, 2017
  d4341920
- [X86] Remove the X86ISD::MOVLHPD. Lowering doesn't use it and it's not a real instruction. · 3b11fca7
  Craig Topper authored Sep 18, 2017
  3b11fca7
- [X86] Teach shuffle lowering to use MOVLHPS/MOVHLPS for lowering v4f32 unary... · ee6646d7
  Craig Topper authored Sep 17, 2017
  ee6646d7
Sep 17, 2017
- [X86] Synchronize a pattern between SSE1 and AVX/AVX512. · 0a197df6
  Craig Topper authored Sep 17, 2017
  0a197df6
- [X86] Colocate all of the X86VBroadcast patterns for v2i64 and v2f64. NFC · 9689fc6d
  Craig Topper authored Sep 17, 2017
  9689fc6d
- [X86] Remove patterns for X86Movddup with v4i64 type. Lowering doesn't emit these. · 9c0bf2c7
  Craig Topper authored Sep 17, 2017
  9c0bf2c7
- [X86] Remove isel patterns for X86Movhlps and X86Movlhps with integer types.... · 5831e2c8
  Craig Topper authored Sep 17, 2017
  5831e2c8
- [X86] Remove isel patterns for movlpd/movlps with integer types. Lowering doesn't emit these. · e305c5ab
  Craig Topper authored Sep 17, 2017
  e305c5ab
- [RISCV] Add support for disassembly · 8ab4a969
  Alex Bradbury authored Sep 17, 2017
  8ab4a969
- [RISCV] Add support for all RV32I instructions · 6758ecb9
  Alex Bradbury authored Sep 17, 2017
  6758ecb9
- [GlobalISel][X86] refactoring X86InstructionSelector.cpp .NFC. · 06335bbd
  Igor Breger authored Sep 17, 2017
  06335bbd
- [GlobalISel][X86] Legalize i1 G_ADD/G_SUB/G_MUL/G_XOR/G_OR/G_AND instructions. · f1d388a5
  Igor Breger authored Sep 17, 2017
  f1d388a5
- [GlobalISel][X86] G_FCONSTANT support. · 21200ed7
  Igor Breger authored Sep 17, 2017
  21200ed7
- [X86] Remove integer X86ISD::SHUFP patterns. Lowering doesn't emit these. · bef5d244
  Craig Topper authored Sep 17, 2017
  bef5d244
- [X86] Add patterns to make blends with immediate control commutable during isel for load folding. · 7c0de010
  Craig Topper authored Sep 17, 2017
  7c0de010
- [X86] Remove some unused defaults from some multiclass parameters. · e09907fc
  Craig Topper authored Sep 17, 2017
  e09907fc
- [X86] Make PLCMULQDQ instructions commutable during isel to fold loads. · ca05e9fd
  Craig Topper authored Sep 16, 2017
  ca05e9fd
- [X86] Add NoAVX predicates to the patterns for the legacy encoded PCLMUL and AES instructions. · ffca0ff9
  Craig Topper authored Sep 16, 2017
  ffca0ff9
Sep 16, 2017
- [X86] Remove some extra code that snuck into r313450. · b150deac
  Craig Topper authored Sep 16, 2017
  b150deac
- [x86] enable storeOfVectorConstantIsCheap() target hook · 65d67807
  Sanjay Patel authored Sep 16, 2017
  65d67807
- [X86] Add isel patterns to be able to fold loads into VPERM2F128 even when the... · 23f78c16
  Craig Topper authored Sep 16, 2017
  23f78c16
- [X86] Remove VPERM2X128 isel patterns with 32-bit elements. · 833788a0
  Craig Topper authored Sep 16, 2017
  833788a0
- [X86] Remove VPERM2F128/VPERM2I128 intrinsics and autoupgrade to native shuffles. · f264fcc7
  Craig Topper authored Sep 16, 2017
  f264fcc7
Sep 15, 2017
- Change encodeU/SLEB128 to pad to certain number of bytes · 66a99e41
  Sam Clegg authored Sep 15, 2017
  66a99e41
- [llvm] Fix some typos. NFC. · 1be19e6f
  Mandeep Singh Grang authored Sep 15, 2017
  1be19e6f
- Revert r313343 "[X86] PR32755 : Improvement in CodeGen instruction selection for LEAs." · 534bfbd3
  Hans Wennborg authored Sep 15, 2017
  534bfbd3
- [X86] Prefer VPERMQ over VPERM2F128 for any unary shuffle, not just the ones... · 7a183e27
  Craig Topper authored Sep 15, 2017
  7a183e27
- [X86] Use SDNode::ops() instead of makeArrayRef and op_begin(). NFCI · f1620b25
  Craig Topper authored Sep 15, 2017
  f1620b25
- [X86] Don't create i64 constants on 32-bit targets when lowering v64i1 constant build vectors · e0d724cf
  Craig Topper authored Sep 15, 2017
  e0d724cf
- [X86] Add isel pattern infrastructure to begin recognizing when we're... · 143797eb
  Craig Topper authored Sep 15, 2017
  143797eb
- [Hexagon] Switch to parameterized register classes for HVX · 55772976
  Krzysztof Parzyszek authored Sep 15, 2017
  55772976
- [AArch64] allow v8f16 types when FullFP16 is supported · 0c5ba21c
  Sjoerd Meijer authored Sep 15, 2017
  0c5ba21c
- [X86] PR32755 : Improvement in CodeGen instruction selection for LEAs. · 908c8b37
  Jatin Bhateja authored Sep 15, 2017
  908c8b37
- [X86] Remove an unnecessary SmallVector from LowerBUILD_VECTOR. · c2311f47
  Craig Topper authored Sep 14, 2017
  c2311f47
Sep 14, 2017
- Fix warnings in r313297. · 1f2f57a7
  Jan Sjodin authored Sep 14, 2017
  1f2f57a7
- AMDGPU: Fix violating constant bus restriction · c317287f
  Matt Arsenault authored Sep 14, 2017
  c317287f
- Add AddresSpace to PseudoSourceValue. · 312ccf76
  Jan Sjodin authored Sep 14, 2017
  312ccf76
- AMDGPU: Fix assert on alloca of array of struct · 37ab4cf8
  Matt Arsenault authored Sep 14, 2017
  37ab4cf8