Commits · bcde5afcac47f430c8a68e92d809f02bd5786867 · Roger Ferrer / llvm-epi-0.8 · GitLab

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

Oct 30, 2018
- [InstSimplify] add tests for fcmp folds; NFC · bcde5afc
  Sanjay Patel authored Oct 30, 2018
  bcde5afc
- [InstCombine] try to turn shuffle into insertelement · b12e4100
  Sanjay Patel authored Oct 30, 2018
  b12e4100
- [LoopVectorizer] Fix for cost values of memory accesses. · 1f067c94
  Jonas Paulsson authored Oct 30, 2018
  1f067c94
- [SROA] Use offset sizes from the DataLayout instead of the pointer siezes. · f96383c9
  Nicola Zaghen authored Oct 30, 2018
  f96383c9
Oct 29, 2018
- [InstSimplify] add tests for abs/nabs+icmp folding; NFC · 33603198
  Sanjay Patel authored Oct 29, 2018
  33603198
- [LoopUnroll] NFC. Factor out runtime-loop.ll common test behavior. · f923e862
  Fedor Sergeev authored Oct 29, 2018
  f923e862
- [HotColdSplitting] Allow outlining single-block cold regions · dd4be53b
  Vedant Kumar authored Oct 29, 2018
  dd4be53b
Oct 28, 2018
- Revert r344172: [LV] Add a new reduction pattern match · 53bd4f48
  Renato Golin authored Oct 27, 2018
  53bd4f48
Oct 27, 2018
- [Local] Keep K's range if K does not move when combining metadata. · fc7654a6
  Florian Hahn authored Oct 27, 2018
  fc7654a6
Oct 26, 2018
- [ValueTracking] peek through shuffles in ComputeNumSignBits (PR37549) · cc9e401e
  Sanjay Patel authored Oct 26, 2018
  cc9e401e
- Pointer types were treated as zero-size by MergeICmps · 3cc0e935
  Christy Lee authored Oct 26, 2018
  3cc0e935
- [SimpleLoopUnswitch] Unswitch by experimental.guard intrinsics · 619a8346
  Max Kazantsev authored Oct 26, 2018
  619a8346
Oct 25, 2018
- [FPEnv] Last BinaryOperator::isFNeg(...) to m_FNeg(...) changes · 384a74b0
  Cameron McInally authored Oct 25, 2018
  384a74b0
- [ARM] Use Cortex-A57 sched model for Cortex-A72 · a16667e7
  Sam Parker authored Oct 25, 2018
  a16667e7
- [CostModel][X86] Add realistic vXi64 uitofp vXf64 costs · 53e8e145
  Simon Pilgrim authored Oct 25, 2018
  53e8e145
- [CostModel][X86] Add realistic i64 uitofp f64 scalar costs · 0573b8d8
  Simon Pilgrim authored Oct 25, 2018
  0573b8d8
- Add -instcombine-code-sinking option · 1f6ca0ba
  Gabor Buella authored Oct 25, 2018
  1f6ca0ba
- Update MemorySSA in LoopRotate. · ad4d0182
  Alina Sbirlea authored Oct 24, 2018
  ad4d0182
- [HotColdSplitting] Identify larger cold regions using domtree queries · c2990068
  Vedant Kumar authored Oct 24, 2018
  c2990068
- [InstCombine] add test for fptrunc with vector with undef elt; NFC · e9f9a2a2
  Sanjay Patel authored Oct 24, 2018
  e9f9a2a2
Oct 24, 2018
- [hot-cold-split] Name split functions with ".cold" suffix · c8dba682
  Teresa Johnson authored Oct 24, 2018
  c8dba682
- [InstCombine] add test for ComputeNumSignBits with shuffle; NFC · d1fe437c
  Sanjay Patel authored Oct 24, 2018
  d1fe437c
- [InstCombine] add test for select with shuffled condition (PR37549); NFC · 2169b9c9
  Sanjay Patel authored Oct 24, 2018
  2169b9c9
- [InstCombine] try harder to form select from logic ops (2nd try) · 3b206305
  Sanjay Patel authored Oct 24, 2018
  3b206305
- [FPEnv] Convert more BinaryOperator::isFNeg(...) to m_FNeg(...) · 678f43f6
  Cameron McInally authored Oct 24, 2018
  678f43f6
- Revert r345114 · c523036f
  Gil Rapaport authored Oct 24, 2018
  c523036f
- [LV] Don't have fold-tail under optsize invalidate interleave-groups when · 5114390e
  Dorit Nuzman authored Oct 24, 2018
  5114390e
- [LSR] Combine unfolded offset into invariant register · 5012e7f6
  Gil Rapaport authored Oct 24, 2018
  5012e7f6
- [PM] keeping history when original SCC split and then merge into itself · 80a0c97e
  Wei Mi authored Oct 23, 2018
  80a0c97e
Oct 23, 2018
- [HotColdSplitting] Attach MinSize to outlined code · 50315461
  Vedant Kumar authored Oct 23, 2018
  50315461
- [DebugInfo][GlobalOpt] Fix -debugify for globalopt shrinking globals to booleans. · 2fed6ac1
  Jordan Rupprecht authored Oct 23, 2018
  2fed6ac1
- [Reassociate] replace fake binop queries with 'match' API · 5b6b090c
  Sanjay Patel authored Oct 23, 2018
  5b6b090c
- [SLPVectorizer] Add basic support for mul/and/or/xor horizontal reductions · 532a0f12
  Simon Pilgrim authored Oct 23, 2018
  532a0f12
- [InstCombine] use 'match' to handle vectors and simplify code · 747feb28
  Sanjay Patel authored Oct 23, 2018
  747feb28
- [InstCombine] swap select profile metadata when swapping select ops · ad76c682
  Sanjay Patel authored Oct 23, 2018
  ad76c682
- [InstCombine] add/move tests for select with inverted condition; NFC · 54da0057
  Sanjay Patel authored Oct 23, 2018
  54da0057
- [SLSR] use 'match' to simplify code; NFC · 5141435d
  Sanjay Patel authored Oct 23, 2018
  5141435d
- [SLSR] auto-generate full test assertions; NFC · d3d2a0b5
  Sanjay Patel authored Oct 23, 2018
  d3d2a0b5
- [Reassociate] remove bogus tests; NFC · 76762540
  Sanjay Patel authored Oct 22, 2018
  76762540
- [Reassociate] add vector tests with undef elements; NFC · 21a62e23
  Sanjay Patel authored Oct 22, 2018
  21a62e23