Commits · ed2213e6efc1cc71a323f49585e859d6e41853ed · Roger Ferrer / llvm-epi · GitLab

Mar 14, 2016
- [CVP] Convert an SDiv to a UDiv if both operands are known to be nonnegative · d60ae33d
  Haicheng Wu authored Mar 14, 2016
  d60ae33d
Mar 12, 2016
- update test to use FileCheck · 610da4fb
  Sanjay Patel authored Mar 12, 2016
  610da4fb
- [x86, InstCombine] delete x86 SSE2 masked store with zero mask · c4acbae6
  Sanjay Patel authored Mar 12, 2016
  c4acbae6
- [LoopUnroll] Convert some existing tests to unit-tests. · 789ac453
  Michael Zolotukhin authored Mar 12, 2016
  789ac453
Mar 11, 2016
- Introduce @llvm.experimental.deoptimize · b51325db
  Sanjoy Das authored Mar 11, 2016
  b51325db
- [PGO] Skip value profile instrumentation of inline asm · e5a9a275
  Vedant Kumar authored Mar 11, 2016
  e5a9a275
- [PM] Port GVN to the new pass manager, wire it up, and teach a couple of · 89c45a16
  Chandler Carruth authored Mar 11, 2016
  89c45a16
- Remove llvm::getDISubprogram in favor of Function::getSubprogram · adebb937
  Pete Cooper authored Mar 11, 2016
  adebb937
- [LLE] Add missed LoopSimplify dependence · efb23413
  Adam Nemet authored Mar 10, 2016
  efb23413
Mar 10, 2016
- Support arbitrary addrspace pointers in masked load/store intrinsics · 3c8fc57e
  Artur Pilipenko authored Mar 10, 2016
  3c8fc57e
- [SROA] Fix PR25873, which Andrea Di Biagio analyzed the daylights out · 37f1f122
  Chandler Carruth authored Mar 10, 2016
  37f1f122
- [CGP] Duplicate addressing computation in cold paths if required to sink addressing mode · ac115ed7
  Philip Reames authored Mar 09, 2016
  ac115ed7
Mar 09, 2016
- [LICM] Store promotion when memory is thread local · b54c8e6e
  Philip Reames authored Mar 09, 2016
  b54c8e6e
- [x86] fix cost model inaccuracy for vector memory ops · 9f6c4d50
  Sanjay Patel authored Mar 09, 2016
  9f6c4d50
- [InstCombine] (icmp sgt smin(PosA, B) 0) -> (icmp sgt B 0) · ec8a8b54
  Philip Reames authored Mar 09, 2016
  ec8a8b54
- [LLE] Add missing check for unit stride · 660748ca
  Adam Nemet authored Mar 09, 2016
  660748ca
- InstCombine: Restrict computeKnownBits() on all Values to OptLevel > 2 · c31032d6
  Matthias Braun authored Mar 09, 2016
  c31032d6
- add a test RUN to show unexpected behavior · 14f598e5
  Sanjay Patel authored Mar 09, 2016
  14f598e5
- Reland r262337 "calculate builtin_object_size if arg is a removable pointer" · 921c2b4e
  Petar Jovanovic authored Mar 09, 2016
  921c2b4e
Mar 08, 2016
- Remove pr25342 test-case. · 3452d33a
  Junmo Park authored Mar 08, 2016
  3452d33a
- Revert revisions 262636, 262643, 262679, and 262682. · b1bd398c
  Easwaran Raman authored Mar 08, 2016
  b1bd398c
Mar 04, 2016
- [LVI] Fix a bug which prevented use of !range metadata within a query · a0c9f6e7
  Philip Reames authored Mar 04, 2016
  a0c9f6e7
- [InstCombine] Combine A->B->A BitCast · 92e9d0e8
  Guozhi Wei authored Mar 03, 2016
  92e9d0e8
Mar 03, 2016
- [ValueTracking] Remove dead code from an old experiment · 146307eb
  Philip Reames authored Mar 03, 2016
  146307eb
- [InstCombine] transform bitcasted bitwise logic ops with constants (PR26702) · 9bba7508
  Sanjay Patel authored Mar 03, 2016
  9bba7508
- Infrastructure for PGO enhancements in inliner · 3035719c
  Easwaran Raman authored Mar 03, 2016
  3035719c
- [LoopUtils, LV] Fix PR26734 · b840a6d6
  Matthew Simpson authored Mar 03, 2016
  b840a6d6
Mar 02, 2016
- Explode store of arrays in instcombine · 3b8b2ea2
  Amaury Sechet authored Mar 02, 2016
  3b8b2ea2
- Unpack array of all sizes in InstCombine · 7cd3fe7d
  Amaury Sechet authored Mar 02, 2016
  7cd3fe7d
- Add another test for the GlobalOpt change in r212079. · 9ab86aab
  Bob Wilson authored Mar 02, 2016
  9ab86aab
- revert r262424 because there's a *clang test* for AArch64 that checks -O3 asm output · 5e4c46de
  Sanjay Patel authored Mar 02, 2016
  5e4c46de
- [InstCombine] convert 'isPositive' and 'isNegative' vector comparisons to shifts (PR26701) · 147e9279
  Sanjay Patel authored Mar 01, 2016
  147e9279
Mar 01, 2016
- Perform InstructioinCombiningPass before SampleProfile pass. · 1012be12
  Dehao Chen authored Mar 01, 2016
  1012be12
- Fix an issue where fast math flags were dropped during scalarization. · 7ea02fc7
  Owen Anderson authored Mar 01, 2016
  7ea02fc7
- Revert "calculate builtin_object_size if argument is a removable pointer" · 6315f3f9
  Petar Jovanovic authored Mar 01, 2016
  6315f3f9
- calculate builtin_object_size if argument is a removable pointer · 8aef99aa
  Petar Jovanovic authored Mar 01, 2016
  8aef99aa
- [LLE] Add testcase for the fix in r262267 · 94877519
  Adam Nemet authored Mar 01, 2016
  94877519
- [x86, InstCombine] transform more x86 masked loads to LLVM intrinsics · 6f2c01f7
  Sanjay Patel authored Feb 29, 2016
  6f2c01f7
- [x86, InstCombine] transform x86 AVX masked loads to LLVM intrinsics · 98a71505
  Sanjay Patel authored Feb 29, 2016
  98a71505
Feb 29, 2016
- [InstSimplify] Restore fsub 0.0, (fsub 0.0, X) ==> X optzn · 6bb15021
  Benjamin Kramer authored Feb 29, 2016
  6bb15021