Commits · 0e0e2d5d2640c629741cab88ae3d9736f6900bb4 · Roger Ferrer / llvm-epi-0.8 · GitLab

This project is archived. Its data is read-only.

Jul 21, 2016
- [OptDiag,LV] Add hotness attribute to the derived analysis remarks · 0e0e2d5d
  Adam Nemet authored Jul 20, 2016
  0e0e2d5d
Jul 20, 2016
- [OptDiag,LV] Add hotness attribute to analysis remarks · 5b3a5cf6
  Adam Nemet authored Jul 20, 2016
  5b3a5cf6
- [LSV] Don't move stores across may-load instrs, and loosen restrictions on moving loads. · a272c12b
  Justin Lebar authored Jul 20, 2016
  a272c12b
- [LSV] Vectorize up to side-effecting instructions. · 62b03e34
  Justin Lebar authored Jul 20, 2016
  62b03e34
- [LV] Add hotness attribute to missed-optimization remarks · 67c8929a
  Adam Nemet authored Jul 20, 2016
  67c8929a
- [LSV] Don't assume that loads/stores appear in address order in the BB. · 6114b378
  Justin Lebar authored Jul 20, 2016
  6114b378
- [LSV] Insert stores at the right point. · 8778c626
  Justin Lebar authored Jul 19, 2016
  8778c626
- [LSV] Use make_range, and reformat a DEBUG message. NFC · 2cf2c228
  Justin Lebar authored Jul 19, 2016
  2cf2c228
- [LSV] Nix two global (ish) variables in the LoadStoreVectorizer. NFC · 4ee8a2d0
  Justin Lebar authored Jul 19, 2016
  4ee8a2d0
Jul 19, 2016
- Recommit the patch "Use uniforms set to populate VecValuesToIgnore". · 79997a24
  Wei Mi authored Jul 19, 2016
  79997a24
Jul 18, 2016
- Revert rL275912. · f9afff71
  Wei Mi authored Jul 18, 2016
  f9afff71
- Use uniforms set to populate VecValuesToIgnore. · 1fd25726
  Wei Mi authored Jul 18, 2016
  1fd25726
Jul 15, 2016
- [LV] Swap A and B in interleaved access analysis (NFC) · f855346f
  Matthew Simpson authored Jul 15, 2016
  f855346f
Jul 14, 2016
- [LV] Rename StrideAccesses to AccessStrideInfo (NFC) · 96e881de
  Matthew Simpson authored Jul 14, 2016
  96e881de
- [LV] Allow interleaved accesses in loops with predicated blocks · 65ca32b8
  Matthew Simpson authored Jul 14, 2016
  65ca32b8
- [LV] Avoid unnecessary IV scalar-to-vector-to-scalar conversions · 3c3b4a25
  Matthew Simpson authored Jul 14, 2016
  3c3b4a25
Jul 13, 2016
- Extended LoadStoreVectorizer to vectorize subchains. · 640a61cd
  Alina Sbirlea authored Jul 13, 2016
  640a61cd
- [LoopVectorize] Further cleanups · 81d877b3
  David Majnemer authored Jul 13, 2016
  81d877b3
- [LV] Do not invalidate use-lists we're iterating over. · 51078b81
  Michael Kuperstein authored Jul 12, 2016
  51078b81
Jul 12, 2016
- [LV] Remove wrong assumption about LCSSA · a99c46cc
  Michael Kuperstein authored Jul 12, 2016
  a99c46cc
- [LoopVectorize] Assorted cleanups · 9330b784
  David Majnemer authored Jul 12, 2016
  9330b784
- Correct ordering of loads/stores. · cbc6ac2a
  Alina Sbirlea authored Jul 11, 2016
  cbc6ac2a
Jul 11, 2016
- Add TLI.allowsMisalignedMemoryAccesses to LoadStoreVectorizer · 327955e0
  Alina Sbirlea authored Jul 11, 2016
  327955e0
Jul 10, 2016
- Give helper classes/functions internal linkage. NFC. · 4d09892e
  Benjamin Kramer authored Jul 10, 2016
  4d09892e
- [PM] Port LoopVectorize to the new PM. · db90d4d9
  Sean Silva authored Jul 09, 2016
  db90d4d9
Jul 09, 2016
- [PM] Fix a think-o. mv {Scalar,Vectorize}/SLPVectorize.h · 0dacbd8f
  Sean Silva authored Jul 09, 2016
  0dacbd8f
Jul 08, 2016
- Rename LoopAccessAnalysis to LoopAccessLegacyAnalysis /NFC · 7853c1dd
  Xinliang David Li authored Jul 08, 2016
  7853c1dd
- Remove duplicate inclusion /NFC · 8c3554fa
  Xinliang David Li authored Jul 08, 2016
  8c3554fa
Jul 07, 2016
- Add a missing semicolon. · a7e11a5d
  Rui Ueyama authored Jul 07, 2016
  a7e11a5d
- Clang-format LoadStoreVectorizer · 598f8aad
  Alina Sbirlea authored Jul 07, 2016
  598f8aad
- Fixed a bug in vectorizing GEP before gather/scatter intrinsic. · fc1e969d
  Elena Demikhovsky authored Jul 07, 2016
  fc1e969d
Jul 06, 2016
- [LV] Don't widen trivial induction variables · 433cb1df
  Matthew Simpson authored Jul 06, 2016
  433cb1df
Jul 05, 2016
- [LV] Refactor integer induction widening (NFC) · 89188729
  Matthew Simpson authored Jul 05, 2016
  89188729
Jul 02, 2016
- LoadStoreVectorizer: Fix warning about extra semicolon · 3add3a40
  Matt Arsenault authored Jul 01, 2016
  3add3a40
Jul 01, 2016
- Address two correctness issues in LoadStoreVectorizer · 8d8aa5dd
  Alina Sbirlea authored Jul 01, 2016
  8d8aa5dd
- [PM] refactor LoopAccessInfo code part-2 · 94734eef
  Xinliang David Li authored Jul 01, 2016
  94734eef
- LoadStoreVectorizer: improvements: better pointer analysis · a8576706
  Matt Arsenault authored Jul 01, 2016
  a8576706
- 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