Commits · db7d8678bcdc440ffd0529f0d970cf10cbcbe984 · Roger Ferrer / llvm-epi · GitLab

Mar 07, 2023
- [SCEV] Strengthen nowrap flags via ranges for ARs on construction. · 7019624e
  Florian Hahn authored Mar 07, 2023
  7019624e
- [RISCV] Enable ordered reduction. · d5c404d1
  Mel Chen authored Feb 20, 2023
  d5c404d1
- [AArch64] Don't map llvm sqrt intrinsics to veclib functions · 92e0ab93
  Graham Hunter authored Mar 06, 2023
  92e0ab93
Mar 03, 2023
- [LV] Allow scalarization of function calls when masking is required · a1803445
  Graham Hunter authored Sep 16, 2022
  a1803445
Mar 01, 2023
- [RISCV] Pre-commit test case for ordered reduction, NFC · 9d0703a6
  Mel Chen authored Feb 20, 2023
  9d0703a6
- [LoopVectorize] Use overflow-check analysis to improve tail-folding. · c41b41eb
  Sander de Smalen authored Jan 30, 2023
  c41b41eb
- [LoopVectorize] Remove runtime check and scalar tail loop when tail-folding. · fe1b51ff
  Sander de Smalen authored Jan 19, 2023
  fe1b51ff
- NFC: Use generate_test_checks script for LV tests which seem to have been auto-generated. · de111ae7
  Sander de Smalen authored Jan 19, 2023
  
  de111ae7
Feb 28, 2023
- [LV] Reland testcase in 0ec4cae1 · ac67ec3a
  sgokhale authored Feb 28, 2023
  
  ac67ec3a
- [LV] Modify test case for commit 4f9a5447 · 0ec4cae1
  sgokhale authored Feb 28, 2023
  0ec4cae1
- [LV] Reland "Update logic for calculating register usage due to invariants" · 4f9a5447
  sgokhale authored Feb 28, 2023
  4f9a5447
- Revert "[LV] Update logic for calculating register usage due to invariants" · 3c8ddbde
  sgokhale authored Feb 28, 2023
  3c8ddbde
- [LV] Update logic for calculating register usage due to invariants · d1628266
  sgokhale authored Feb 27, 2023
  d1628266
Feb 22, 2023
- [LV][RISCV] Don't interleave scalable vector loops · 15f9cf16
  Luke Lau authored Feb 21, 2023
  15f9cf16
Feb 20, 2023
- [RISCV][NFC] Add test for different LMULs in vectorizer · 0b336e9e
  Luke Lau authored Feb 10, 2023
  0b336e9e
- [LV] Harden the test of the minmax with index pattern. (NFC) · 3e84fc85
  Mel Chen authored Feb 13, 2023
  3e84fc85
Feb 15, 2023
- [PseudoProbe] Refactoring a test · c38c8d67
  Hongtao Yu authored Feb 15, 2023
  c38c8d67
Feb 14, 2023
- [LV] Synthesize all true masks for masked vector function variants · 0fa5df19
  Graham Hunter authored Sep 09, 2022
  0fa5df19
Feb 13, 2023
- [VPlan] Fix iterator invalidation in adjustFixedOrderRecurrences. · af3c25dc
  Florian Hahn authored Feb 13, 2023
  af3c25dc
Feb 09, 2023
- Reland D143267: [LoopVectorize] Use DataLayout::getIndexType instead of i32... · 5a115452
  Sander de Smalen authored Feb 08, 2023
  5a115452
Feb 08, 2023
- [LV] Perform recurrence sinking directly on VPlan. · c83fdc90
  Florian Hahn authored Feb 08, 2023
  c83fdc90
- Revert "[LoopVectorize] Use DataLayout::getIndexType instead of i32 for non-constant GEP indices." · 1f01cdda
  Sander de Smalen authored Feb 08, 2023
  1f01cdda
- [RISCV] Allow mismatched SmallDataLimit and use Min for conflicting values · af128791
  Fangrui Song authored Feb 07, 2023
  af128791
Feb 07, 2023
- [LV] Remove unused load from RISCV test (NFC). · a69f2349
  Florian Hahn authored Feb 07, 2023
  a69f2349
- [LoopVectorize] Use DataLayout::getIndexType instead of i32 for non-constant GEP indices. · e6eb84a1
  Sander de Smalen authored Feb 06, 2023
  e6eb84a1
Feb 06, 2023
- [LV] Also check interleaving only in select-min-index.ll · 40621ff4
  Florian Hahn authored Feb 06, 2023
  40621ff4
Feb 04, 2023
- [LV] Add users for loads to make tests more robust. · a9ac22b5
  Florian Hahn authored Feb 04, 2023
  a9ac22b5
- [LV] Add initial tests for sinking loads past other instructions. · fd5bccb8
  Florian Hahn authored Feb 04, 2023
  fd5bccb8
Jan 31, 2023
- [InstCombine][LV] Fold (add (zext (add X, -1)), 1) -> (zext X) if X is non-zero. · df76ff98
  Craig Topper authored Jan 30, 2023
  df76ff98
Jan 30, 2023
- [LoopVectorize] Enable integer Mul and Add as select reduction patterns · 8ff47f60
  Matt Devereau authored Jan 27, 2023
  8ff47f60
Jan 26, 2023
- Revert "[LoopVectorize] Enable integer Mul and Add as select reduction patterns" · 4468e27d
  Matt Devereau authored Jan 26, 2023
  4468e27d
Jan 25, 2023
- [LoopVectorize] Enable integer Mul and Add as select reduction patterns · f9010385
  Matt Devereau authored Jan 16, 2023
  f9010385
Jan 21, 2023
- [VPlan] Consider all recipes in replicate blocks as sink candidates. · fb40c34b
  Florian Hahn authored Jan 21, 2023
  fb40c34b
Jan 20, 2023
- [AArch64] Enable libm vectorized functions via SLEEF · c4fa504f
  Daniel Kiss authored Jan 20, 2023
  c4fa504f
Jan 18, 2023
- [AsmParser] Remove typed pointer auto-detection · 9ed2f14c
  Nikita Popov authored Dec 14, 2022
  9ed2f14c
Jan 11, 2023
- [AArch64] Set MaxInterleaveFactor for Apple A14, A15, A16. · 830d0bc5
  Florian Hahn authored Jan 11, 2023
  830d0bc5
- [AArch64] Add tests for selecting interleave counts for different CPUs. · 42f67837
  Florian Hahn authored Jan 10, 2023
  42f67837
- [IRBuilder] Use canonical i64 type for insertelement index used by vector splats. · eae26b66
  Paul Walker authored Jan 04, 2023
  eae26b66
- [LoopVectorize] Clear cache of `LoopAccessInfoManager` · 3c5e0d87
  Miguel Saldivar authored Jan 11, 2023
  3c5e0d87
Jan 08, 2023
- [VPlan] Keep entries in worklist in sinkScalarOperands. · 78914e8c
  Florian Hahn authored Jan 08, 2023
  78914e8c