Commits · f50d4b6cdcbe0e8f20275bc040018bfb1d593eff · Roger Ferrer / llvm-epi · GitLab

Jul 06, 2016
- Work around PR28400 a bit harder. · f50d4b6c
  Sean Silva authored Jul 06, 2016
  f50d4b6c
- [TTI] The cost model should not assume vector casts get completely scalarized · aa71bdd3
  Michael Kuperstein authored Jul 06, 2016
  aa71bdd3
- [LV] Don't widen trivial induction variables · 433cb1df
  Matthew Simpson authored Jul 06, 2016
  433cb1df
- Vector GEP test: renamed + some comments · 971fbfda
  Elena Demikhovsky authored Jul 06, 2016
  971fbfda
- Fix handling of forward unreachable but reverse-reachable blocks in MemorySSA construction · fc7e651b
  Daniel Berlin authored Jul 06, 2016
  fc7e651b
Jul 03, 2016
- [InstCombine] enable vector select of bools -> logic folds · cbaac418
  Sanjay Patel authored Jul 03, 2016
  cbaac418
- add vector bool select tests and regenerate checks for scalar bool select tests · 42396ae0
  Sanjay Patel authored Jul 03, 2016
  42396ae0
- Remove dead TLI arg of isKnownNonNull and propagate deadness. NFC. · 45835e73
  Sean Silva authored Jul 02, 2016
  45835e73
Jul 02, 2016
- [PM] Port ConstantHoisting to the new Pass Manager · 071d8306
  Michael Kuperstein authored Jul 02, 2016
  071d8306
Jul 01, 2016
- Address two correctness issues in LoadStoreVectorizer · 8d8aa5dd
  Alina Sbirlea authored Jul 01, 2016
  8d8aa5dd
- Revert "add tests for bugs fixed by the GVN hoist pass" · e60719b3
  Duncan P. N. Exon Smith authored Jul 01, 2016
  e60719b3
- add tests for bugs fixed by the GVN hoist pass · 196ba4f8
  Sebastian Pop authored Jul 01, 2016
  196ba4f8
- LoadStoreVectorizer: Don't increase alignment with no align set · 0101ecad
  Matt Arsenault authored Jul 01, 2016
  0101ecad
- LoadStoreVectorizer: Check TTI for vec reg bit width · 370e8226
  Matt Arsenault authored Jul 01, 2016
  370e8226
- LoadStoreVectorizer: Fix assert when merging pointer ops · 42ad1705
  Matt Arsenault authored Jul 01, 2016
  42ad1705
- Revert "code hoisting pass based on GVN" · 9d1f1564
  Duncan P. N. Exon Smith authored Jul 01, 2016
  9d1f1564
- LoadStoreVectorizer: Use AA metadata · 241f34cd
  Matt Arsenault authored Jul 01, 2016
  241f34cd
- LoadStoreVectorizer: if one element of a vector is integer, default to · d7e8898b
  Matt Arsenault authored Jul 01, 2016
  d7e8898b
- LoadStoreVectorizer: Fix crashes on sub-byte types · 8a4ab5e1
  Matt Arsenault authored Jul 01, 2016
  8a4ab5e1
- code hoisting pass based on GVN · 5c5798c5
  Sebastian Pop authored Jul 01, 2016
  5c5798c5
- LoadStoreVectorizer: Check skipFunction first. · 079d0f19
  Matt Arsenault authored Jun 30, 2016
  079d0f19
- LoadStoreVectorizer: Skip optnone functions · 2cbe52b9
  Matt Arsenault authored Jun 30, 2016
  2cbe52b9
- Add LoadStoreVectorizer pass · 08debb02
  Matt Arsenault authored Jun 30, 2016
  08debb02
Jun 30, 2016
- SLPVectorizer: Move propagateMetadata to VectorUtils · 727e279a
  Matt Arsenault authored Jun 30, 2016
  727e279a
- Refine the set of UniformAfterVectorization instructions. · 95685fae
  Wei Mi authored Jun 30, 2016
  95685fae
- [DSE] Fix bug in partial overwrite tracking · 596a3bd9
  Jun Bum Lim authored Jun 30, 2016
  596a3bd9
- [InstCombine] shrink switch conditions better (PR24766) · 7c6eab57
  Sanjay Patel authored Jun 30, 2016
  7c6eab57
- [InstCombine] extend matchSelectFromAndOr() to work with i1 scalar types · 7ad98bab
  Sanjay Patel authored Jun 30, 2016
  7ad98bab
- add vector tests to show missing transform · 348111f4
  Sanjay Patel authored Jun 30, 2016
  348111f4
- regenerate checks · c3701e8b
  Sanjay Patel authored Jun 29, 2016
  c3701e8b
Jun 29, 2016
- StackColoring for SafeStack. · a5da256f
  Evgeniy Stepanov authored Jun 29, 2016
  a5da256f
- [InstCombine] Simplify and correct folding fcmps with the same children · aec68b26
  Tim Shen authored Jun 29, 2016
  aec68b26
- [InstCombine, NFC] Change the generated variable names by creating new instructions · 860a67eb
  Tim Shen authored Jun 29, 2016
  860a67eb
- [InstCombine] Add full tests for FoldAndOfFCmps and FoldOrOfFCmps · 4561e784
  Tim Shen authored Jun 29, 2016
  4561e784
- Reverted patch 273864 · 5e21c94f
  Elena Demikhovsky authored Jun 29, 2016
  5e21c94f
- Revert "[ValueTracking] Teach computeKnownBits for PHI nodes to compute sign... · df7454f9
  Craig Topper authored Jun 29, 2016
  df7454f9
- [ValueTracking] Teach computeKnownBits for PHI nodes to compute sign bit for a... · 2cc199ba
  Craig Topper authored Jun 29, 2016
  2cc199ba
- Revert "[InstCombine] Avoid combining the bitcast of a var that is used as... · 0c58837b
  Eric Christopher authored Jun 29, 2016
  0c58837b
- [ARM] Fix 28282: cost computation for constant hoisting · 5410eddd
  Weiming Zhao authored Jun 28, 2016
  5410eddd
Jun 28, 2016
- minimize regression tests and update checks · 3a0f2606
  Sanjay Patel authored Jun 28, 2016
  3a0f2606