Commits · d0aa53b9ae0b1bd1b093f3d40e0a0e9e128dedd2 · Roger Ferrer / llvm-epi · GitLab

Jan 04, 2017
- [AVX-512] Add support for detecting 512-bit shuffles that contain a 128-bit... · d0aa53b9
  Craig Topper authored Jan 04, 2017
  d0aa53b9
- [AVX-512] Add more test cases for shuffles that should be handled with... · a3b9a4ed
  Craig Topper authored Jan 04, 2017
  a3b9a4ed
- [AVX-512] Fix a typo in a couple case names to match their behavior. · 9e065c5b
  Craig Topper authored Jan 04, 2017
  9e065c5b
- [AVX-512] Add avx512dq to the vector-shuffle-512-v16.ll test command lines in... · 42e8e33c
  Craig Topper authored Jan 04, 2017
  42e8e33c
- [InstCombine] Add a test for r290733 · b5e365c9
  David Majnemer authored Jan 04, 2017
  b5e365c9
- [InstCombine] Move casts around shift operations · cb892e90
  David Majnemer authored Jan 04, 2017
  cb892e90
- [InstCombine] Combine adds across a zext · 022d2a56
  David Majnemer authored Jan 04, 2017
  022d2a56
Jan 03, 2017
- InstCombine: Fold fabs on select of constants · 56ff4839
  Matt Arsenault authored Jan 03, 2017
  56ff4839
- [InstCombine] tighten checks for tests of assume -> metadata transform; NFC · ada846af
  Sanjay Patel authored Jan 03, 2017
  ada846af
- [X86][SSE] Add extra truncated arithmetic tests for D28219 · 1145989a
  Simon Pilgrim authored Jan 03, 2017
  1145989a
- Add llvm-bcanalyzer support for new metadata node types. · 36daf63b
  Adrian Prantl authored Jan 03, 2017
  36daf63b
- Enable disabled loopidiom test. Apparently we handle it now · 883dd1b6
  Xin Tong authored Jan 03, 2017
  883dd1b6
- [AVX-512] Teach shuffle lowering to use vinsert instructions for shuffles... · 9496e3f9
  Craig Topper authored Jan 03, 2017
  9496e3f9
- [AVX-512] Teach EVEX to VEX conversion pass to handle VINSERT and VEXTRACT instructions. · fa875a1d
  Craig Topper authored Jan 03, 2017
  fa875a1d
- [AVX-512] Re-generate tests that were updated for r290663 without using... · 15d116ab
  Craig Topper authored Jan 03, 2017
  15d116ab
- [AVX-512] Add support for pushing bitcasts through INSERT_SUBVEC in order to... · c8491721
  Craig Topper authored Jan 03, 2017
  c8491721
- [AVX-512] Remove vinsert intrinsics and autoupgrade to native shufflevectors.... · 0cda8bbf
  Craig Topper authored Jan 03, 2017
  0cda8bbf
- [AVX-512] Remove vextract intrinsics and autoupgrade to native shufflevectors.... · 4d47c6ae
  Craig Topper authored Jan 03, 2017
  4d47c6ae
- InstCombine: Add fma with constant transforms · b264c949
  Matt Arsenault authored Jan 03, 2017
  b264c949
- InstCombine: Add fma + fabs/fneg transforms · 1cc294c8
  Matt Arsenault authored Jan 03, 2017
  1cc294c8
Jan 02, 2017
- Make sure total loop body weight is preserved in loop peeling · 2940231f
  Xin Tong authored Jan 02, 2017
  2940231f
- NewGVN: Add a test case for equivalent phis. · aa0ec1e9
  Daniel Berlin authored Jan 02, 2017
  aa0ec1e9
- NewGVN: Add forgotten testcase for PR 31483 · 43a5f998
  Daniel Berlin authored Jan 02, 2017
  43a5f998
- [cmake] Add LLVM_ENABLE_DIA_SDK option, and expose it in LLVMConfig · 89b6f16b
  Michal Gorny authored Jan 02, 2017
  89b6f16b
- Emit .cfi_sections before the first .cfi_startproc · 7b83732a
  Joerg Sonnenberger authored Jan 02, 2017
  7b83732a
- Fixed shuffle-reverse cost on AVX-512. · d96200d6
  Elena Demikhovsky authored Jan 02, 2017
  d96200d6
- AVX-512 Loop Vectorizer: Cost calculation for interleave load/store patterns. · 21706cbd
  Elena Demikhovsky authored Jan 02, 2017
  21706cbd
- Reapply "[CodeGen] Fix invalid DWARF info on Win64" · f7d84ee6
  Keno Fischer authored Jan 02, 2017
  f7d84ee6
Jan 01, 2017
- [InstCombine] add explanatory comment to test; NFC · 0e3ae439
  Sanjay Patel authored Jan 01, 2017
  0e3ae439
- [InstCombine] add test to show potential nonnull attribute propagation; NFC · 07537c2b
  Sanjay Patel authored Jan 01, 2017
  07537c2b
Dec 31, 2016
- [ValueTracking] add tests for known-nonnull-at; NFC · 5865d12e
  Sanjay Patel authored Dec 31, 2016
  5865d12e
- [AVR] Optimize 16-bit ANDs with '1' · 97cf837b
  Dylan McKay authored Dec 31, 2016
  97cf837b
- [InstCombine][AVX-512] Teach InstCombine that llvm.x86.avx512.vcomi.sd and... · d00db692
  Craig Topper authored Dec 31, 2016
  d00db692
Dec 30, 2016
- [X86][AVX512DQ] Add truncated math tests for AVX512DQ. · c5fde8d7
  Simon Pilgrim authored Dec 30, 2016
  c5fde8d7
- [X86][SSE] Fix truncated math test names. · 85af9735
  Simon Pilgrim authored Dec 30, 2016
  85af9735
- [X86][AVX512] Regenerate test - missing shuffle comments · 71237416
  Simon Pilgrim authored Dec 30, 2016
  71237416
- [MemDep] Handle gep with zeros for invariant.group · da362150
  Piotr Padlewski authored Dec 30, 2016
  da362150
- Fix test change in r290736: restore index generation · e2770c0b
  Mehdi Amini authored Dec 30, 2016
  e2770c0b
- [LICM] When promoting scalars, allow inserting stores to thread-local allocas. · 76e06c88
  Michael Kuperstein authored Dec 30, 2016
  76e06c88
- Use continuous boosting factor for complete unroll. · cc76344e
  Dehao Chen authored Dec 30, 2016
  cc76344e