Commits · f0093e45c4fb5a765ddf94367ddd675f3efb6bc3 · Roger Ferrer / llvm-epi · GitLab

Sep 25, 2017
- [SimplifyIndvar] Replace the srem used by IV if we can prove both of its operands are non-negative · f0093e45
  Hongbin Zheng authored Sep 25, 2017
  f0093e45
- revert r314117 because there are bogus clang tests that depend on the optimizer · aa7f750b
  Sanjay Patel authored Sep 25, 2017
  aa7f750b
- [InstCombine] remove extract-of-select vector transform · 9639897d
  Sanjay Patel authored Sep 25, 2017
  9639897d
- [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF8 stride 4): · 4a97df01
  Michael Zuckerman authored Sep 25, 2017
  4a97df01
- [SLP] Support for horizontal min/max reduction. · ccce7afe
  Alexey Bataev authored Sep 25, 2017
  ccce7afe
Sep 22, 2017
- [InstCombine] Teach foldICmpUsingKnownBits to simplify SLE/SGE/ULE/UGE to... · ea927bae
  Craig Topper authored Sep 22, 2017
  ea927bae
- [InstCombine] Add test cases for known bits simplifications for comparisons... · 73a99890
  Craig Topper authored Sep 22, 2017
  73a99890
- [InstCombine] Remove a FIXME from a test that was fixed in r314025. · 615729b3
  Craig Topper authored Sep 22, 2017
  615729b3
- [InstCombine] Add constant splat handling to one of the ICMP_SLT/SGT cases in... · 3f364aa9
  Craig Topper authored Sep 22, 2017
  3f364aa9
- Rework loop predication pass · 889dc1e3
  Artur Pilipenko authored Sep 22, 2017
  889dc1e3
Sep 21, 2017
- Re-land r313825: "[IR] Add llvm.dbg.addr, a control-dependent version of llvm.dbg.declare" · 0fe506bc
  Reid Kleckner authored Sep 21, 2017
  0fe506bc
- Revert r313825: "[IR] Add llvm.dbg.addr, a control-dependent version of llvm.dbg.declare" · 7d2f38d6
  Daniel Jasper authored Sep 21, 2017
  7d2f38d6
- Fixed reverted commit rL312318 · 29202f6d
  Strahinja Petrovic authored Sep 21, 2017
  29202f6d
- Revert "Re-enable "[IRCE] Identify loops with latch comparison against current IV value"" · 675e304e
  Serguei Katkov authored Sep 21, 2017
  675e304e
- [InstCombine] Teach getDemandedBitsLHSMask to handle constant splat vectors · 18887bf1
  Craig Topper authored Sep 20, 2017
  18887bf1
- [SimplifyCFG] don't create a no-op subtract · 73811a15
  Sanjay Patel authored Sep 20, 2017
  73811a15
Sep 20, 2017
- [IR] Add llvm.dbg.addr, a control-dependent version of llvm.dbg.declare · 3f547e87
  Reid Kleckner authored Sep 20, 2017
  3f547e87
- [SimplifyCFG] auto-generate full checks; NFC · 04308650
  Sanjay Patel authored Sep 20, 2017
  04308650
- [InstCombine] Handle (X & C2) < C1 --> (X & C2) == 0 · 562bf99e
  Craig Topper authored Sep 20, 2017
  562bf99e
- [InstCombine] Pre-commit test cases for D38065. · 9b593a69
  Craig Topper authored Sep 20, 2017
  9b593a69
- Revert r313771 "[SLP] Vectorize jumbled memory loads." · 57c3341a
  Hans Wennborg authored Sep 20, 2017
  57c3341a
- [InstCombine] Add select simplifications · aa103b3d
  Quentin Colombet authored Sep 20, 2017
  aa103b3d
- [SLP] Vectorize jumbled memory loads. · 2b281de5
  Mohammad Shahid authored Sep 20, 2017
  2b281de5
- [ThinLTO] Fix dead stripping analysis for SamplePGO · f625118e
  Teresa Johnson authored Sep 20, 2017
  f625118e
- Revert r313736: "[SLP] Vectorize jumbled memory loads." · 6a140234
  Alexander Kornienko authored Sep 20, 2017
  6a140234
- [SLP] Vectorize jumbled memory loads. · f8db9bd8
  Mohammad Shahid authored Sep 20, 2017
  f8db9bd8
- Tighten the invariants around LoopBase::invalidate · 09613b12
  Sanjoy Das authored Sep 20, 2017
  09613b12
Sep 19, 2017
- Import all inlined indirect call targets for SamplePGO. · 62b9c33e
  Dehao Chen authored Sep 19, 2017
  62b9c33e
- Handle profile mismatch correctly for SamplePGO. · b6e60c8b
  Dehao Chen authored Sep 19, 2017
  b6e60c8b
- [SLP] Reduce test, NFC. · 10ad32b7
  Alexey Bataev authored Sep 19, 2017
  10ad32b7
Sep 18, 2017
- [InstCombine] auto-generate complete checks; NFC · 709a804a
  Sanjay Patel authored Sep 18, 2017
  709a804a
- [SLP] Add a test for PR34635, NFC. · 286fe620
  Alexey Bataev authored Sep 18, 2017
  286fe620
- [ARM] Implement isTruncateFree · 71efbe4c
  Sam Parker authored Sep 18, 2017
  71efbe4c
Sep 16, 2017
- [X86] Remove VPERM2F128/VPERM2I128 intrinsics and autoupgrade to native shuffles. · f264fcc7
  Craig Topper authored Sep 16, 2017
  f264fcc7
- [SLP] Revert r312791 and other necessary commits, except for TTI and · beb22b54
  Chandler Carruth authored Sep 15, 2017
  beb22b54
Sep 15, 2017
- [ConstantFold] Return the correct type when folding a GEP with vector indices. · dee018c5
  Davide Italiano authored Sep 15, 2017
  dee018c5
- This patch fixes https://bugs.llvm.org/show_bug.cgi?id=32352 · b5ab895e
  Vivek Pandya authored Sep 15, 2017
  b5ab895e
- This reverts r313381 · df8598dc
  Vivek Pandya authored Sep 15, 2017
  df8598dc
- This patch fixes https://bugs.llvm.org/show_bug.cgi?id=32352 · 00d88744
  Vivek Pandya authored Sep 15, 2017
  00d88744
- Revert r313343 "[X86] PR32755 : Improvement in CodeGen instruction selection for LEAs." · 534bfbd3
  Hans Wennborg authored Sep 15, 2017
  534bfbd3