Commits · e08d9c7c8748670b4fa0e215ac11c89675bda6ab · Roger Ferrer / llvm-epi · GitLab

Dec 11, 2016
- instr-combiner: sum up all latencies of the transformed instructions · e08d9c7c
  Sebastian Pop authored Dec 11, 2016
  e08d9c7c
- [X86][XOP] Add target shuffle tests showing missing PSHUFPD combine. · 8766a76f
  Simon Pilgrim authored Dec 11, 2016
  8766a76f
- [SCEVExpander] Explicitly expand AddRec starts into loop preheader · ba1bf875
  Sanjoy Das authored Dec 11, 2016
  ba1bf875
- [X86] Regcall - Adding support for mask types · 9683ecbf
  Oren Ben Simhon authored Dec 11, 2016
  9683ecbf
- [FileCheck] Re-implement the logic to find each check prefix in the · 726774cb
  Chandler Carruth authored Dec 11, 2016
  726774cb
- [FileCheck] Remove a parameter that was simply always set to · b03c166a
  Chandler Carruth authored Dec 11, 2016
  b03c166a
- [FileCheck] Clean up doxygen comments throughout. NFC. · 4dabac20
  Chandler Carruth authored Dec 11, 2016
  4dabac20
- [FileCheck] Run clang-format over this code. NFC. · e8f2fb20
  Chandler Carruth authored Dec 11, 2016
  e8f2fb20
- Refactor FileCheck some to reduce memory allocation and copying. Also · 20247900
  Chandler Carruth authored Dec 11, 2016
  20247900
- [X86][InstCombine] Add support for scalar FMA intrinsics to SimplifyDemandedVectorElts. · 23ebd956
  Craig Topper authored Dec 11, 2016
  23ebd956
- [X86][InstCombine] Add the test cases for r289370, r289371, and r289372. · 7a230f42
  Craig Topper authored Dec 11, 2016
  7a230f42
- Tweak the core loop in StringRef::find to avoid calling memcmp on every · ecbe6196
  Chandler Carruth authored Dec 11, 2016
  ecbe6196
- [X86][InstCombine] Teach InstCombineCalls to simplify demanded elements for scalar FMA intrinsics. · 61b280e7
  Craig Topper authored Dec 11, 2016
  61b280e7
- [AVX-512][InstCombine] Teach InstCombineCalls how to simplify demanded for... · d9639536
  Craig Topper authored Dec 11, 2016
  d9639536
- [AVX-512][InstCombine] Teach InstCombineCalls how to simplify demanded... · 790d0fa5
  Craig Topper authored Dec 11, 2016
  790d0fa5
- [AVR] Add calling convention CodeGen tests · bf1d2eda
  Dylan McKay authored Dec 11, 2016
  bf1d2eda
- [libFuzzer] don't depend on time in a test · 441e6310
  Kostya Serebryany authored Dec 11, 2016
  441e6310
- [AVR] Add a test to validate a simple 'blinking led' program · 72967a56
  Dylan McKay authored Dec 11, 2016
  72967a56
- [AVX-512][InstCombine] Add 512-bit vpermilvar intrinsics to InstCombineCalls... · 58917f35
  Craig Topper authored Dec 11, 2016
  58917f35
- [X86] Fix a comment to say 'an FMA' instead of 'a FMA'. NFC · e7166ce2
  Craig Topper authored Dec 11, 2016
  e7166ce2
- [X86] Remove masking from 512-bit VPERMIL intrinsics in preparation for being... · 1f1b4412
  Craig Topper authored Dec 11, 2016
  1f1b4412
- [AVR] Fix a signed vs unsigned compiler warning · 139c0c7c
  Dylan McKay authored Dec 11, 2016
  139c0c7c
- [X86][InstCombine] Teach InstCombineCalls to turn pshufb intrinsic into a... · 9a63d7ad
  Craig Topper authored Dec 11, 2016
  9a63d7ad
- [AVR] Remove incorrect comment · 658bb096
  Dylan McKay authored Dec 10, 2016
  658bb096
- [X86] Remove masking from 512-bit PSHUFB intrinsics in preparation for being... · edab02b5
  Craig Topper authored Dec 10, 2016
  edab02b5
Dec 10, 2016
- [InstCombine] add helper for shift-by-shift folds; NFCI · 4c48bbe9
  Sanjay Patel authored Dec 10, 2016
  4c48bbe9
- [X86][SSE] Add tests for sign extended vXi64 multiplication · b71b2142
  Simon Pilgrim authored Dec 10, 2016
  b71b2142
- [X86][SSE] Ensure UNPCK inputs are a consistent value type in LowerHorizontalByteSum · a03e350e
  Simon Pilgrim authored Dec 10, 2016
  a03e350e
- [AVX-512] Remove 128/256 masked vpermil instrinsics and autoupgrade to a... · abe7c5b5
  Craig Topper authored Dec 10, 2016
  abe7c5b5
- [X86][IR] Move the autoupgrading of store intrinsics out of the main nested... · a4744d17
  Craig Topper authored Dec 10, 2016
  a4744d17
- AMDGPU: Fix asan errors when folding operands · fbc72885
  Matt Arsenault authored Dec 10, 2016
  fbc72885
- [X86][SSE] Move ZeroVector creation into the shuffle pattern case where its actually used. · fb58550d
  Simon Pilgrim authored Dec 10, 2016
  fb58550d
- [AVX-512] Add support for lowering (v2i64 (fp_to_sint (v2f32))) to vcvttps2uqq... · 18b57da4
  Craig Topper authored Dec 10, 2016
  18b57da4
- [X86] Clarify indentation. NFC · 8e288e0b
  Craig Topper authored Dec 10, 2016
  8e288e0b
- [X86] Combine LowerFP_TO_SINT and LowerFP_TO_UINT. They only differ by a... · 85f0e57c
  Craig Topper authored Dec 10, 2016
  85f0e57c
- [InstSimplify] improve function name; NFC · 35289c62
  Sanjay Patel authored Dec 10, 2016
  35289c62
- [mips] Eliminate else-after-return. NFC · edd7a7bb
  Simon Atanasyan authored Dec 10, 2016
  edd7a7bb
- [SelectionDAG] Add ability for computeKnownBits to peek through bitcasts from... · 54945a12
  Simon Pilgrim authored Dec 10, 2016
  54945a12
- [X86][XOP] Add permil2ps buildvector combine test · 90a040e7
  Simon Pilgrim authored Dec 10, 2016
  90a040e7
- [AVR] Add a stub README file · 41258cf0
  Dylan McKay authored Dec 10, 2016
  41258cf0