Commits · dbbaca0e1b5d9a6e36ffdf523c946f02fd6c2357 · Roger Ferrer / llvm-epi · GitLab

Feb 24, 2016
- [InstCombine] enable optimization of casted vector xor instructions · dbbaca0e
  Sanjay Patel authored Feb 24, 2016
  dbbaca0e
- add test to show missing bitcasted vector xor fold · 3a74fb51
  Sanjay Patel authored Feb 24, 2016
  3a74fb51
- [SimplifyCFG] Do not blindly remove unreachable blocks · ec72e372
  David Majnemer authored Feb 24, 2016
  ec72e372
Feb 23, 2016
- minimize test and use FileCheck · 55a2b244
  Sanjay Patel authored Feb 23, 2016
  55a2b244
- [WinEH] Don't inline an 'unwinds to caller' cleanupret into funclets which locally unwind · 223538f7
  David Majnemer authored Feb 23, 2016
  223538f7
- More detailed dependence test between volatile and non-volatile accesses · e261e5ac
  Krzysztof Parzyszek authored Feb 22, 2016
  e261e5ac
Feb 22, 2016
- Set function entry count as 0 if sample profile is not found for the function. · 6c73b499
  Dehao Chen authored Feb 22, 2016
  6c73b499
- [LoopDataPrefetch] Make it testable with opt · fb31d580
  Adam Nemet authored Feb 22, 2016
  fb31d580
- [LoopUnrolling] Fix a bug introduced in r259869 (PR26688). · d734bea8
  Michael Zolotukhin authored Feb 22, 2016
  d734bea8
- [RS4GC] "Constant fold" the rs4gc-split-vector-values flag · ce38c2dd
  Philip Reames authored Feb 22, 2016
  ce38c2dd
- ARM: sink atomic release barrier as far as possible into cmpxchg. · d32f8e60
  Tim Northover authored Feb 22, 2016
  d32f8e60
- [RS4GC] Revert optimization attempt due to memory corruption · 79fa9b75
  Philip Reames authored Feb 22, 2016
  79fa9b75
- Revert "[attrs] Handle convergent CallSites." · ccbd8f5a
  Justin Lebar authored Feb 22, 2016
  ccbd8f5a
- [attrs] Handle convergent CallSites. · 7bf9187a
  Justin Lebar authored Feb 22, 2016
  7bf9187a
- Allow setting MaxRerollIterations above 16 · 9914dbd1
  Elena Demikhovsky authored Feb 22, 2016
  9914dbd1
Feb 21, 2016
- [InstCombine] Added SSE41 roundss/roundsd demanded vector elements invec tests · b1cc4d6d
  Simon Pilgrim authored Feb 21, 2016
  b1cc4d6d
- [InstCombine] Added XOP frczss/vfrczsd demanded vector elements tests · e2ada8d3
  Simon Pilgrim authored Feb 21, 2016
  e2ada8d3
- [InstCombine] Added SSE41 roundss/roundsd demanded vector elements tests · 2ed8686a
  Simon Pilgrim authored Feb 21, 2016
  2ed8686a
- [InstCombine] SSE/SSE2 (u)comiss/(u)comisd comparison intrinsics only use the lowest vector element · 471efd24
  Simon Pilgrim authored Feb 20, 2016
  471efd24
Feb 20, 2016
- [InstCombine] Added SSE/SSE2 comparison intrinsics demanded vector elements tests · 6a0dca53
  Simon Pilgrim authored Feb 20, 2016
  6a0dca53
- [InstCombine] Added some SSE/SSE2 demanded vector elements tests · d4768fa3
  Simon Pilgrim authored Feb 20, 2016
  d4768fa3
- [SimplifyCFG] Merge together cleanuppads · 1efa23dd
  David Majnemer authored Feb 20, 2016
  1efa23dd
Feb 19, 2016
- Revert r255691 "[LoopVectorizer] Refine loop vectorizer's register usage... · a0f70905
  Hans Wennborg authored Feb 19, 2016
  a0f70905
- [LV] Vectorize first-order recurrences · 29c997c1
  Matthew Simpson authored Feb 19, 2016
  29c997c1
- [LV] Fix PR26600: avoid out of bounds loads for interleaved access vectorization · ad1dafb2
  Silviu Baranga authored Feb 19, 2016
  ad1dafb2
- [LPM] Factor all of the loop analysis usage updates into a common helper · 31088a9d
  Chandler Carruth authored Feb 19, 2016
  31088a9d
- Add profile summary support for sample profile. · 40ee23db
  Easwaran Raman authored Feb 19, 2016
  40ee23db
- [CaptureTracking] Add a test case for pointer cmpxchg · 36706852
  Philip Reames authored Feb 19, 2016
  36706852
- [IR] Extend cmpxchg to allow pointer type operands · 1960cfd3
  Philip Reames authored Feb 19, 2016
  1960cfd3
Feb 18, 2016
- [IR] Straighten out bundle overload of IRBuilder::CreateCall · b6c6dda4
  Reid Kleckner authored Feb 18, 2016
  b6c6dda4
- [CaptureTracking] Support atomicrmw and cmpxchg · bd09e86f
  Philip Reames authored Feb 18, 2016
  bd09e86f
- Reapply commit r259357 with a fix for PR26629 · 92821cb4
  Matthew Simpson authored Feb 18, 2016
  92821cb4
- [PM] Port the PostOrderFunctionAttrs pass to the new pass manager and · 9c4ed175
  Chandler Carruth authored Feb 18, 2016
  9c4ed175
Feb 17, 2016
- [LIR] Avoid turning non-temporal stores into memset · 57e1a3e6
  Haicheng Wu authored Feb 17, 2016
  57e1a3e6
- Create masked gather and scatter intrinsics in Loop Vectorizer. · 88e76cad
  Elena Demikhovsky authored Feb 17, 2016
  88e76cad
- Fix load alignement when unpacking aggregates structs · 61a7d629
  Amaury Sechet authored Feb 17, 2016
  61a7d629
- Revert "Reapply commit r258404 with fix." · f48bcb2b
  David Majnemer authored Feb 17, 2016
  f48bcb2b
- [ObjCARC] Handle ARCInstKind::ClaimRV in OptimizeIndividualCalls. · 009d6065
  Frederic Riss authored Feb 17, 2016
  009d6065
- [AliasSetTracker] Teach AliasSetTracker about MemSetInst · 5cf99095
  Haicheng Wu authored Feb 17, 2016
  5cf99095
Feb 16, 2016
- Revert 260705, it appears to be causing pr26628 · 845435c8
  Philip Reames authored Feb 16, 2016
  845435c8