Commits · 3c46abb2eab8c0e62f3fdf552a6c5d501aed2738 · Roger Ferrer / llvm-epi-0.8 · GitLab

This is an archived project. Repository and other project resources are read-only.

You need to sign in or sign up before continuing.

Sep 19, 2016
- [llvm-cov] Teach the coverage exporter about instantiation coverage · 3c46abb2
  Vedant Kumar authored Sep 19, 2016
  3c46abb2
- [llvm-cov] Make a helper method static for re-use (NFC) · 9cbf80af
  Vedant Kumar authored Sep 19, 2016
  9cbf80af
- [llvm-cov] Track function and instantiation coverage separately · 016111f7
  Vedant Kumar authored Sep 19, 2016
  016111f7
- [llvm-cov] Don't recompute the 'Covered' field from *CoverageInfo (NFC) · 673ad727
  Vedant Kumar authored Sep 19, 2016
  673ad727
- [llvm-cov] Make 'adjustColumnWidths' do less work · dab0ec12
  Vedant Kumar authored Sep 19, 2016
  dab0ec12
- [llvm-cov] Drop another redundant 'No.' suffix · 98ba34e5
  Vedant Kumar authored Sep 19, 2016
  98ba34e5
- [utils] Delete the 'check-coverage-regressions' script · 44bc28c9
  Vedant Kumar authored Sep 19, 2016
  44bc28c9
- Handle Invoke during sample profiler annotation: make it inlinable. · 41cde0b9
  Dehao Chen authored Sep 18, 2016
  41cde0b9
- Extend title underline · e1117102
  Xinliang David Li authored Sep 18, 2016
  e1117102
Sep 18, 2016
- [AVX-512] Don't lower CVTPD2PS intrinsics to ISD::FP_ROUND with an X86... · af5ee86b
  Craig Topper authored Sep 18, 2016
  af5ee86b
- [AVX-512] Stop lowering avx512_mask_sqrt intrinsics to ISD:FSQRT with a second... · c26cd684
  Craig Topper authored Sep 18, 2016
  c26cd684
- [libFuzzer] add -print_coverage=1 flag to print coverage directly from libFuzzer · b706b481
  Kostya Serebryany authored Sep 18, 2016
  b706b481
- Fix covered-switch-default warning · f33a6b71
  Simon Pilgrim authored Sep 18, 2016
  f33a6b71
- [CostModel][X86] Added scalar float op costs · 91780598
  Simon Pilgrim authored Sep 18, 2016
  91780598
- Rename tests · 34032cba
  Simon Pilgrim authored Sep 18, 2016
  34032cba
- [X86] Fix typo in comment. NFC · cc03165d
  Craig Topper authored Sep 18, 2016
  cc03165d
- [AVX-512] Add memory load patterns for the legacy SSE scalar fp to integer... · 8542041b
  Craig Topper authored Sep 18, 2016
  8542041b
- [AVX-512] Remove COPY_TO_REGCLASS from a few patterns that already had the correct register class. · 8c252bc4
  Craig Topper authored Sep 18, 2016
  8c252bc4
- Fix built bot failure · f0630d3d
  Xinliang David Li authored Sep 18, 2016
  f0630d3d
- [Profile] Implement select instruction instrumentation in IR PGO · 4ca1733a
  Xinliang David Li authored Sep 18, 2016
  4ca1733a
- clang-format: [JS] Do not wrap taze annotation comments. · fbbe75b1
  Martin Probst authored Sep 18, 2016
  fbbe75b1
- clang-format: [JS] ASI insertion after boolean literals. · b9316ff8
  Martin Probst authored Sep 18, 2016
  b9316ff8
- CodeGen: mark ObjC cstring literals as unnamed_addr · 3f307518
  Saleem Abdulrasool authored Sep 18, 2016
  3f307518
- CodeGen: mark ObjC cstring literals as constant · 0c54dc86
  Saleem Abdulrasool authored Sep 18, 2016
  0c54dc86
- [Loop Vectorizer] Consecutive memory access - fixed and simplified · 5f8cc0c3
  Elena Demikhovsky authored Sep 18, 2016
  5f8cc0c3
- [X86][SSE] Improve recognition of uitofp conversions that can be performed as sitofp · 6c21e6a5
  Simon Pilgrim authored Sep 18, 2016
  6c21e6a5
- [Loop vectorizer] Simplified GEP cloning. NFC. · a1a0e7dd
  Elena Demikhovsky authored Sep 18, 2016
  a1a0e7dd
- GPGPU: add missing REQUIRES line to test case · 05ee64e6
  Tobias Grosser authored Sep 18, 2016
  05ee64e6
- GPGPU: Do not run mostly sequential kernels in GPU · bc653f20
  Tobias Grosser authored Sep 18, 2016
  bc653f20
- GPGPU: Dynamically ensure 'sufficient compute' · 82f2af35
  Tobias Grosser authored Sep 18, 2016
  82f2af35
- GPGPU: Make test cases independent of register numbering [NFC] · cfdee658
  Tobias Grosser authored Sep 18, 2016
  cfdee658
- Change the order of the splitted store from high - low to low - high. · ab24cd18
  Wei Mi authored Sep 18, 2016
  ab24cd18
- [libFuzzer] use 'if guard' instead of 'if guard >= 0' with trace-pc; change... · 8e781a88
  Kostya Serebryany authored Sep 18, 2016
  8e781a88
- [llvm-objump] Simplify the code. NFCI. · 1bdaa20b
  Davide Italiano authored Sep 18, 2016
  1bdaa20b
- [lib/LTO] Try harder to reduce code duplication. NFCI. · a416d113
  Davide Italiano authored Sep 17, 2016
  a416d113
- [X86][SSE] Added vector udiv combine tests · aeb764a7
  Simon Pilgrim authored Sep 17, 2016
  aeb764a7
Sep 17, 2016
- [X86][SSE] Added vector fcopysign combine tests · 2cc29002
  Simon Pilgrim authored Sep 17, 2016
  2cc29002
- [ThinLTO] Ensure anonymous globals renamed even at -O0 · fbb431b2
  Teresa Johnson authored Sep 17, 2016
  fbb431b2
- [X86][SSE] Added vector mul combine tests · d4473f11
  Simon Pilgrim authored Sep 17, 2016
  d4473f11
- GPGPU: Store back non-read-only scalars · 51dfc275
  Tobias Grosser authored Sep 17, 2016
  51dfc275