Commits · 9a6817b7edb44669cec14a3fbde8a93dfb217c1a · Roger Ferrer / llvm-epi · GitLab

Nov 16, 2021
- [ARM] Add datalayout to costmodel tests. NFC · 309f1e4a
  David Green authored Nov 16, 2021
  309f1e4a
- [CGP][PowerPC] Pre-commit test case for D113872. NFC. · c0da8a4e
  Kai Luo authored Nov 16, 2021
  
  c0da8a4e
- [STATEPOINT] Force implicit-def for lr register. · 0ecb12a2
  Serguei Katkov authored Oct 05, 2021
  0ecb12a2
- [GlobalISel] Add a store-merging optimization pass and enable for AArch64. · dc84770d
  Amara Emerson authored Nov 15, 2021
  dc84770d
- [llvm-profgen] Add switch to allow use of first loadable segment for calculating offset · f7976edc
  Wenlei He authored Nov 11, 2021
  f7976edc
- [RISCV] Override TargetLowering::hasAndNot for Zbb. · 391b0ba6
  Craig Topper authored Nov 15, 2021
  391b0ba6
- [RISCV] Add test cases to prepare for overring TargetLowering::hasAndNot. NFC · d90eeab0
  Craig Topper authored Nov 15, 2021
  d90eeab0
- [X86] Fix crash with inline asm using wrong register name · b484fa82
  Fabian Wolff authored Nov 16, 2021
  b484fa82
- AMDGPU: Mark prolog/epilog SCC defs as dead · 659887b4
  Matt Arsenault authored Nov 13, 2021
  659887b4
- AMDGPU: Regenerate test checks · e6bfbd7e
  Matt Arsenault authored Nov 13, 2021
  
  e6bfbd7e
- [DAGCombiner] Prevent unfoldMaskedMerge from creating an AND with two inverted inputs. · 233def40
  Craig Topper authored Nov 15, 2021
  233def40
- [InstSimplify] Fold A|B | (A^B) --> A|B · 62c51a72
  Mehrnoosh Heidarpour authored Nov 15, 2021
  62c51a72
- [RISCV] Optimize immediate materialisation with SH*ADD · 4c3d916c
  Ben Shi authored Nov 11, 2021
  4c3d916c
- [RISCV][test] Add more tests of immediate materialisation · 39256ed5
  Ben Shi authored Nov 10, 2021
  39256ed5
Nov 15, 2021
- Revert "[InstSimplify] Fold A|B | (A^B) --> A|B" · 833cdb0a
  Stanislav Mekhanoshin authored Nov 15, 2021
  833cdb0a
- [NewPM] Only invalidate modified functions' analyses in CGSCC passes + turn on... · 19867de9
  Arthur Eubanks authored May 03, 2021
  19867de9
- [InstSimplify] Fold A|B | (A^B) --> A|B · 193c40e9
  Stanislav Mekhanoshin authored Nov 15, 2021
  193c40e9
- [SystemZ] Support symbolic displacements. · 1c3ef9ef
  Jonas Paulsson authored Nov 05, 2021
  1c3ef9ef
- [asm] Correctly handle special names in variants · 833393e0
  Nico Weber authored Nov 15, 2021
  833393e0
- [PowerPC] Fix 32bit vector insert instructions for ISA3.1 · f50c6c17
  Lei Huang authored Nov 12, 2021
  f50c6c17
- Fix a misleading FIXME in an unroll test · da327e72
  Philip Reames authored Nov 15, 2021
  
  da327e72
- [SLP][NFC]Add a test for extra shuffle emission, NFC. · 2d0cab9d
  Alexey Bataev authored Nov 15, 2021
  
  2d0cab9d
- [RISCV] Teach needVSETVLIPHI to handle mask register instructions. · f59307bf
  Craig Topper authored Nov 15, 2021
  f59307bf
- [NFC][X86][Costmodel] Improve test coverage for {i8,i16,i32,i64}->i1 vector trunc · 7114c60e
  Roman Lebedev authored Nov 15, 2021
  
  7114c60e
- [NFC][X86][Costmodel] Improve test coverage for i1->{i8,i16,i32,i64} vector *ext · 949103dc
  Roman Lebedev authored Nov 15, 2021
  
  949103dc
- [InstCombine] Fold (A^B)|~A-->~(A&B) · 7daa95c8
  Mehrnoosh Heidarpour authored Nov 15, 2021
  7daa95c8
- [NFC][X86][Costmodel] Add i1 replication shuffle costmodel test coverage · bc35d5fe
  Roman Lebedev authored Nov 15, 2021
  
  bc35d5fe
- [x86] fold vector (X > -1) & Y to shift+andn (2nd try) · 3d01507c
  Sanjay Patel authored Nov 15, 2021
  3d01507c
- [x86] add test for vector signbit mask fold (PR52504); NFC · 6efe64cf
  Sanjay Patel authored Nov 15, 2021
  6efe64cf
- [X86][Costmodel] `getReplicationShuffleCost()`: promote 8 bit-wide elements to... · 5c7255fe
  Roman Lebedev authored Nov 15, 2021
  5c7255fe
- [X86][Costmodel] `trunc v32i16 to v64i8` can appear after legalization, cost... · a468c39c
  Roman Lebedev authored Nov 15, 2021
  a468c39c
- [X86][Costmodel] `trunc v8i64 to v16i8/v32i8/v64i8` can appear after... · 9e57d9b0
  Roman Lebedev authored Nov 15, 2021
  9e57d9b0
- [X86][Costmodel] `trunc v16i32 to v32i8/v64i8` can appear after legalization,... · 0116c708
  Roman Lebedev authored Nov 15, 2021
  0116c708
- [X86] getAVX512Node() - find constant broadcasts to encourage load-folding · ea9e6aa4
  Simon Pilgrim authored Nov 15, 2021
  ea9e6aa4
- [SLP]Improve splat detection. · 036207d5
  Alexey Bataev authored Nov 12, 2021
  036207d5
- [NFC][InstSimplify] add test cases with base results for or-xor fold · a7f7cf11
  Mehrnoosh Heidarpour authored Nov 15, 2021
  a7f7cf11
- [SLP]Do not create unused gather nodes for scalar arguments of vector intrinsics. · 6fb5bed7
  Alexey Bataev authored Nov 12, 2021
  6fb5bed7
- [IVDescriptor] Make sure the sign is included for negative extension. · 112c1c34
  Florian Hahn authored Nov 15, 2021
  112c1c34
- [llvm-nm][test] Move X86 lit.local.cfg into the X86 subfolder · 254aa65d
  James Henderson authored Nov 15, 2021
  254aa65d
- Revert "[x86] fold vector (X > -1) & Y to shift+andn" · 5be64d41
  Hans Wennborg authored Nov 15, 2021
  5be64d41