Commits · f674d7dab1b17c7e8b1f37ce06345760f94d2ede · Roger Ferrer / llvm-epi · GitLab

Sep 25, 2019
- [AArch64][GlobalISel] Tweak legalization rule for G_BSWAP to handle widening s16. · f674d7da
  Amara Emerson authored Sep 25, 2019
  f674d7da
- [BPF] Generate array dimension size properly for zero-size elements · 1487bf6c
  Yonghong Song authored Sep 24, 2019
  1487bf6c
Sep 24, 2019
- Extends the expansion of the LWZtoc pseduo op for AIX. · b3a9320c
  Sean Fertile authored Sep 24, 2019
  b3a9320c
- [X86] Add MMX MOVD/MOVQ stores to folding tables to support stack folding · a7f27f35
  Simon Pilgrim authored Sep 24, 2019
  a7f27f35
- [X86] Add tests showing failure to stack fold MMX MOVD/MOVQ stores · 682d41a5
  Simon Pilgrim authored Sep 24, 2019
  682d41a5
- Revert r372333: [DAG][X86] Convert isNegatibleForFree/GetNegatedExpression to... · 60e5e0b6
  Ilya Biryukov authored Sep 24, 2019
  60e5e0b6
- [ARM] Split large widening MVE loads · 2fb41fc7
  David Green authored Sep 24, 2019
  2fb41fc7
- [ARM] MVE sext and widen/narrow tests from larger types. NFC · 2462d421
  David Green authored Sep 24, 2019
  2462d421
- [ARM] Split large truncating MVE stores · 49d851f4
  David Green authored Sep 24, 2019
  49d851f4
- [GlobalISel][IRTranslator] Fix switch table lowering to use signed LE not unsigned. · adec1209
  Amara Emerson authored Sep 24, 2019
  adec1209
Sep 23, 2019
- [X86] Reduce the number of unique check prefixes in memset-nonzero.ll. NFC · 8a6916e6
  Craig Topper authored Sep 23, 2019
  8a6916e6
- [WebAssembly] vNxM.load_splat instructions · 99d3dd28
  Thomas Lively authored Sep 23, 2019
  99d3dd28
- [BreakFalseDeps] ignore function with minsize attribute · 74141519
  Sanjay Patel authored Sep 23, 2019
  74141519
- [Hexagon] Bitcast v4i16 to v8i8, unify no-op casts between scalar and HVX · f97fdf57
  Krzysztof Parzyszek authored Sep 23, 2019
  f97fdf57
- [x86] fix assert with horizontal math + broadcast of vector (PR43402) · 31b9dfe2
  Sanjay Patel authored Sep 23, 2019
  31b9dfe2
- [ARM][MVE] Remove old tail predicates · 9feb429a
  Sam Parker authored Sep 23, 2019
  9feb429a
- [AArch64] support neon_sshl and neon_ushl in performIntrinsicCombine. · 3e2fdbee
  Florian Hahn authored Sep 23, 2019
  3e2fdbee
- [ARM][LowOverheadLoops] Use subs during revert. · 4ba6d0de
  Sam Parker authored Sep 23, 2019
  4ba6d0de
- [ARM][LowOverheadLoops] Use tBcc when reverting · 566127e3
  Sam Parker authored Sep 23, 2019
  566127e3
- [MIPS GlobalISel] VarArg argument lowering, select G_VASTART and vacopy · c063b0b0
  Petar Avramovic authored Sep 23, 2019
  c063b0b0
- [X86] Canonicalize all zeroes vector to RHS in X86DAGToDAGISel::tryVPTESTM. · 03b5a13e
  Craig Topper authored Sep 23, 2019
  03b5a13e
- [X86] Remove SETEQ/SETNE canonicalization code from LowerIntVSETCC_AVX512 to... · 5e26064c
  Craig Topper authored Sep 23, 2019
  5e26064c
- [X86] Add 32-bit command line to avx512f-vec-test-testn.ll · 1f058538
  Craig Topper authored Sep 23, 2019
  1f058538
- Prefer AVX512 memcpy when applicable · a7a515cb
  David Zarzycki authored Sep 23, 2019
  a7a515cb
- [X86][SelectionDAGBuilder] Move the hack for handling MMX shift by i32... · a533e877
  Craig Topper authored Sep 23, 2019
  a533e877
- [X86] X86DAGToDAGISel::matchBEXTRFromAndImm(): if can't use BEXTR, fallback to... · 7c3d6f5a
  Roman Lebedev authored Sep 22, 2019
  7c3d6f5a
Sep 22, 2019
- [NFC][X86] Add BEXTR test with load and 33-bit mask (PR43381 / D67875) · 24159592
  Roman Lebedev authored Sep 22, 2019
  24159592
- [X86] Update commutable EVEX vcmp patterns to use timm instead of imm. · a1d86857
  Craig Topper authored Sep 22, 2019
  a1d86857
- [X86] Add more tests for commuting evex vcmp instructions during isel to fold a load. · ac847712
  Craig Topper authored Sep 22, 2019
  ac847712
- [X86] Add test memset and memcpy testcases for D67874. NFC · 38014c55
  Craig Topper authored Sep 22, 2019
  38014c55
Sep 21, 2019
- [NFC][X86] Adjust check prefixes in bmi.ll (PR43381) · 854b0f0f
  Roman Lebedev authored Sep 21, 2019
  854b0f0f
- [AArch64][GlobalISel] Implement selection for G_SHL of <2 x i64> · 9c7d599d
  Amara Emerson authored Sep 21, 2019
  9c7d599d
- [AArch64][GlobalISel] Selection support for G_ASHR of <2 x s64> · a59a8868
  Amara Emerson authored Sep 21, 2019
  a59a8868
- [AArch64][GlobalISel] Make <4 x s32> G_ASHR and G_LSHR legal. · fae979bc
  Amara Emerson authored Sep 21, 2019
  fae979bc
- [MachinePipeliner] Improve the TargetInstrInfo API analyzeLoop/reduceLoopCount · 8a74eca3
  James Molloy authored Sep 21, 2019
  8a74eca3
- [X86] Use sse_load_f32/f64 and timm in patterns for memory form of vgetmantss/sd. · 04682939
  Craig Topper authored Sep 21, 2019
  04682939
- [X86] Add test case to show failure to fold load with getmantss due to isel... · 4fa12ac9
  Craig Topper authored Sep 21, 2019
  4fa12ac9
- AMDGPU/GlobalISel: Allow selection of scalar min/max · eb6eb694
  Matt Arsenault authored Sep 21, 2019
  eb6eb694
- [GlobalISel] Defer setting HasCalls on MachineFrameInfo to selection time. · 7ac10399
  Amara Emerson authored Sep 20, 2019
  7ac10399
- [SystemZ] Support z15 processor name · 819c1651
  Ulrich Weigand authored Sep 20, 2019
  819c1651