Commits · f941ec0ecc2d177f54a94ad429278947556a3ea9 · Roger Ferrer / llvm-epi · GitLab

Apr 21, 2017
- add skylake · 00b51bf1
  Clement Courbet authored Apr 21, 2017
  00b51bf1
- add 32 bit tests · 2430e25b
  Clement Courbet authored Apr 21, 2017
  2430e25b
- use repmovsb when optimizing forminsize · d5f6182b
  Clement Courbet authored Apr 21, 2017
  d5f6182b
- Rename FastString flag. · 203fc177
  Clement Courbet authored Apr 21, 2017
  203fc177
- add more tests · a7c233fb
  Clement Courbet authored Apr 21, 2017
  a7c233fb
- X86 memcpy: use REPMOVSB instead of REPMOVS{Q,D,W} for inline copies · 1ce3b82d
  Clement Courbet authored Apr 21, 2017
  1ce3b82d
Apr 20, 2017
- Temporarily revert r299221 to fix nondeterminism in ThinLTO builder. · 2cc97d92
  Galina Kistanova authored Apr 19, 2017
  2cc97d92
- X86FrameLowering: Fix getFrameIndexReference() for 'fixed' objects · 372ee597
  Matthias Braun authored Apr 19, 2017
  372ee597
- [DAG] add splat vector support for 'or' in SimplifyDemandedBits · 0658a95a
  Sanjay Patel authored Apr 19, 2017
  0658a95a
Apr 19, 2017
- [DAG] add splat vector support for 'xor' in SimplifyDemandedBits · ae382bb6
  Sanjay Patel authored Apr 19, 2017
  ae382bb6
- Update the madd.ll test with utils/update_llc_test_checks.py (NFC) · cfe5b2bf
  Dehao Chen authored Apr 19, 2017
  cfe5b2bf
- PR32710: Disable using PMADDWD for unsigned short. · 58601674
  Dehao Chen authored Apr 19, 2017
  58601674
- [DAG] add splat vector support for 'and' in SimplifyDemandedBits · ded7d59f
  Sanjay Patel authored Apr 19, 2017
  ded7d59f
- [GlobalIsel][X86] support G_TRUNC selection. · 4fdf1e48
  Igor Breger authored Apr 19, 2017
  4fdf1e48
- [X86] Add D32039/PR31357 tests to show current BSWAP codegen · 94d5dba8
  Simon Pilgrim authored Apr 19, 2017
  94d5dba8
- [X86][SSE] Add scheduling latency/throughput tests for (most) SSE2 instructions · b13ab635
  Simon Pilgrim authored Apr 19, 2017
  b13ab635
- [GlobalISel][X86] Split select tests. NFC. · 746b3c33
  Igor Breger authored Apr 19, 2017
  746b3c33
- [x86] add tests for potential andn optimization; NFC · ff981f92
  Sanjay Patel authored Apr 18, 2017
  ff981f92
Apr 18, 2017
- [X86] Keep EXTRACT_VECTOR_ELT result type as f128 for Android x86_64. · 877923a8
  Chih-Hung Hsieh authored Apr 18, 2017
  877923a8
- [X86][SSE] Add scheduling latency/throughput tests for (most) SSE1 instructions · 9398649f
  Simon Pilgrim authored Apr 18, 2017
  9398649f
- [DAG] Improve store merge candidate pruning. · 855ef456
  Nirav Dave authored Apr 18, 2017
  855ef456
- Add base-index-based store merge test · e50544cd
  Nirav Dave authored Apr 18, 2017
  e50544cd
- Add store Merge test. · b9776849
  Nirav Dave authored Apr 18, 2017
  b9776849
- Change the testcase tail-merge-after-mbp.ll to tail-merge-after-mbp.mir · 68f82a31
  Haicheng Wu authored Apr 17, 2017
  68f82a31
Apr 16, 2017
- [X86] Remove special handling for 16 bit for A asm constraints. · f5f593b6
  Benjamin Kramer authored Apr 16, 2017
  f5f593b6
- Use correct registers for "A" inline asm constraint · 909b3376
  Dimitry Andric authored Apr 15, 2017
  909b3376
Apr 14, 2017
- [X86][SSE] Update MOVNTDQA non-temporal loads to generic implementation (LLVM) · 5a22eaa2
  Simon Pilgrim authored Apr 14, 2017
  5a22eaa2
- Fix for PR#30562: Selection DAG error: Detected cycle in SelectionDAG. · 4e7bcd52
  Andrew V. Tischenko authored Apr 14, 2017
  4e7bcd52
- This patch closes PR#32216: Better testing of schedule model instruction latencies/throughputs. · 75745d0c
  Andrew V. Tischenko authored Apr 14, 2017
  75745d0c
Apr 12, 2017
- [GlobalIsel][X86] support G_CONSTANT selection. · 3b97ea39
  Igor Breger authored Apr 12, 2017
  3b97ea39
Apr 11, 2017
- [DAGCombine] Add more test cases for shuffle of splat. NFC. · f720c036
  Zvi Rackover authored Apr 11, 2017
  f720c036
- [x86] Relax the check in areLoadsFromSameBasePtr · ddb9ae19
  Easwaran Raman authored Apr 11, 2017
  ddb9ae19
- [X86] Create the correct ADC/SBB SDNode when lowering add. · 8455f7d6
  Davide Italiano authored Apr 11, 2017
  8455f7d6
- [ARM, x86] add tests to show possible improvement for bool math; NFC · 8f200116
  Sanjay Patel authored Apr 10, 2017
  8f200116
- CodeGen: BlockPlacement: Don't always tail-duplicate with no other successor. · 7e8be286
  Kyle Butt authored Apr 10, 2017
  7e8be286
- CodeGen: BlockPlacement: Minor probability changes. · ee51a201
  Kyle Butt authored Apr 10, 2017
  ee51a201
- CodeGen: BranchFolding: Merge identical blocks, even if they are short. · a12bd756
  Kyle Butt authored Apr 10, 2017
  a12bd756
Apr 10, 2017
- Add address space mangling to lifetime intrinsics · f10061ec
  Matt Arsenault authored Apr 10, 2017
  f10061ec
- [X86][MMX] Add fast-isel support for MMX non-temporal writes · b6702eae
  Simon Pilgrim authored Apr 10, 2017
  b6702eae
Apr 07, 2017
- Use PMADDWD to expand reduction in a loop · 58fa7244
  Dehao Chen authored Apr 07, 2017
  58fa7244