Commits · fc01b9bdf8b55f6b09f1dcaedf78dad62ff205c1 · Roger Ferrer / llvm-epi · GitLab

May 22, 2021
- [CostModel][X86] Align v4i64 MUL costs on AVX1 targets with worst case · fc01b9bd
  Simon Pilgrim authored May 22, 2021
  fc01b9bd
- [HIP] support ThinLTO · bf612458
  Yaxun (Sam) Liu authored Mar 29, 2021
  bf612458
- Reapply [InstCombine] Fold multiuse shr eq zero · 9a9421a4
  Nikita Popov authored Apr 19, 2021
  9a9421a4
- [CostModel][X86] vXi8 MUL is always promoted to vXi16 · 7a898477
  Simon Pilgrim authored May 22, 2021
  
  7a898477
- [CVP] Add test for PR50399 (NFC) · 069174a6
  Nikita Popov authored May 22, 2021
  
  069174a6
- Revert "[NewPM] Only invalidate modified functions' analyses in CGSCC passes" · f7788e1b
  Arthur Eubanks authored May 21, 2021
  f7788e1b
May 21, 2021
- [InstSimplify] add tests for rem-of-mul; NFC · 3c4b7948
  David Goldblatt authored May 21, 2021
  3c4b7948
- precommit tests for D102934 and D102928 · 317c105c
  Philip Reames authored May 21, 2021
  
  317c105c
- [CostModel][X86] Improve f64/v2f64/v4f64 FMUL costs on AVX1 targets to account for slower btver2 · fe6c11c5
  Simon Pilgrim authored May 21, 2021
  fe6c11c5
- Revert "[LoopUnrollAndJam] Change LoopUnrollAndJamPass to LoopNest pass" · fd53cb41
  maekawatoshiki authored May 22, 2021
  fd53cb41
- [CostModel][X86] Improve fneg costs · 2fca5558
  Simon Pilgrim authored May 21, 2021
  2fca5558
- [LoopUnrollAndJam] Change LoopUnrollAndJamPass to LoopNest pass · cea7a3fe
  maekawatoshiki authored May 21, 2021
  cea7a3fe
- [SLP]Improve handling of compensate external uses cost. · 8dab2595
  Alexey Bataev authored Apr 29, 2021
  8dab2595
- [SLP][NFC]Add a test for diamond match of broadcast tree nodes. · 117a247e
  Alexey Bataev authored May 21, 2021
  
  117a247e
- [VectorCombine] Add positive test for scalarizing multiple extracts. · 4efb4f67
  Florian Hahn authored May 21, 2021
  4efb4f67
- [DebugInfo] Salvage dbg.value() during ADCE · cd49b3ae
  Djordje Todorovic authored May 21, 2021
  cd49b3ae
- [CostModel][X86] Tweak fptoui v4f32->v4i32 + v8f32->v8i32 SSE/AVX costs · 3ae7f7ae
  Simon Pilgrim authored May 21, 2021
  3ae7f7ae
- [CostModel][X86] Add uitpfp v4f32->v4i32 + v8f32->v8i32 SSE/AVX costs · eb6429d0
  Simon Pilgrim authored May 21, 2021
  eb6429d0
- 3rd Reapply "[DebugInfo] Use variadic debug values to salvage BinOps and GEP... · 36ec97f7
  Stephen Tozer authored May 20, 2021
  36ec97f7
- [InstSimplify] Properly constrain {insert,extract}_subvector intrinsic fold · 5a476987
  Joe Ellis authored May 21, 2021
  5a476987
- [IR] Memory intrinsics are not unconditionally `nosync` · a888e492
  Nicolai Hähnle authored May 12, 2021
  a888e492
- [tests] Update Transforms/DeadStoreElim/multiblock-malloc-free.ll · 77b83d30
  Nicolai Hähnle authored May 19, 2021
  77b83d30
May 20, 2021
- Revert "[Remarks] Add analysis remarks for memset/memcpy/memmove lengths" · 0af3105b
  Jon Roelofs authored May 20, 2021
  0af3105b
- [FPEnv] EarlyCSE support for constrained intrinsics, default FP environment edition · f21f1eea
  Kevin P. Neal authored May 20, 2021
  f21f1eea
- [CodeGen][AArch64][SVE] Canonicalize intrinsic rdffr{ => _z} · 2d574a11
  Peter Waller authored May 12, 2021
  2d574a11
- [GlobalOpt] recompute alignments for loads and stores of updated globals · f34311c4
  Sanjay Patel authored May 20, 2021
  f34311c4
- [GlobalOpt] adjust test to show load problems; NFC · ee4055cf
  Sanjay Patel authored May 20, 2021
  ee4055cf
- [SLP]Try to vectorize tiny trees with shuffled gathers of extractelements. · 182162b6
  Alexey Bataev authored Apr 28, 2021
  182162b6
- [NFC][Coroutines] Autogenerate a few tests for ease of further updates · 4a35c51c
  Roman Lebedev authored May 20, 2021
  
  4a35c51c
- [NFC][CHR] Autogenerate checklines in a few tests for ease of updates · eeeeff0d
  Roman Lebedev authored May 20, 2021
  
  eeeeff0d
- [NFC][PruneEH] Autogenerate checklines in a few tests for ease of updates · 7e3d0a5d
  Roman Lebedev authored May 20, 2021
  
  7e3d0a5d
- [NFC][SimplifyCFG] Autogenerate checklines in a few tests for ease of updates · 8294e94a
  Roman Lebedev authored May 20, 2021
  
  8294e94a
- [RISCV] Add legality check for vectorizing reduction · 1595994b
  Luke authored Mar 29, 2021
  1595994b
- [Remarks] Add analysis remarks for memset/memcpy/memmove lengths · 4bf69fb5
  Jon Roelofs authored May 18, 2021
  4bf69fb5
May 19, 2021
- [CSSPGO] Avoid deleting probe instruction in FoldValueComparisonIntoPredecessors · 6539a80b
  wlei authored May 18, 2021
  6539a80b
- [Diagnostics] Allow emitting analysis and missed remarks on functions · 2db182ff
  Joseph Huber authored May 19, 2021
  2db182ff
- Do actual DCE in LoopUnroll (try 4) · 449d14eb
  Philip Reames authored May 19, 2021
  449d14eb
- [CSSPGO] Overwrite branch weight annotated in previous pass. · 4ca6e37b
  Hongtao Yu authored May 16, 2021
  4ca6e37b
- Revert "Do actual DCE in LoopUnroll (try 3)" · 51785742
  Amy Huang authored May 19, 2021
  51785742
- [LV] Add -scalable-vectorization=<option> flag. · 4f86aa65
  Sander de Smalen authored Apr 08, 2021
  4f86aa65