Commits · ddd3807e6952dedfd842b3b10076a614f75c1979 · Roger Ferrer / llvm-epi · GitLab

Feb 02, 2022
- [AMDGPU] Use new target MMO flag MONoClobber · ddd3807e
  Jay Foad authored Feb 02, 2022
  ddd3807e
Feb 01, 2022
- [AMDGPU] Check atomics aliasing in the clobbering annotation · 79606ee8
  Stanislav Mekhanoshin authored Jan 31, 2022
  79606ee8
- [AMDGPU] Allow scalar loads after barrier · c2b18a3c
  Stanislav Mekhanoshin authored Jan 27, 2022
  c2b18a3c
- [AMDGPU][test] Add dso_local to prevent preemptible alias resolution · 1494d064
  Fangrui Song authored Feb 01, 2022
  
  1494d064
- [StructurizeCFG] Clean up some boolean not instructions · d2e5d351
  Jay Foad authored Jan 31, 2022
  d2e5d351
- AMDGPU {NFC}: Add code object v5 support and generate metadata for implicit kernel args · 1194b9cd
  Changpeng Fang authored Jan 31, 2022
  1194b9cd
Jan 31, 2022
- Revert "[Local] invertCondition: try modifying an existing ICmpInst" · 8faad296
  Jay Foad authored Jan 31, 2022
  8faad296
- [AMDGPU] Add test for a problem with noclobber metadata · ae68b3a4
  Jay Foad authored Jan 28, 2022
  ae68b3a4
- [Local] invertCondition: try modifying an existing ICmpInst · a6b54dda
  Jay Foad authored Jan 28, 2022
  a6b54dda
Jan 27, 2022
- AMDGPU: Handle addrspacecast of constant 32-bit to flat · 33b45ee4
  Matt Arsenault authored Jan 26, 2022
  33b45ee4
- AMDGPU: Fix broken check lines in test · d77c7c80
  Matt Arsenault authored Jan 26, 2022
  
  d77c7c80
- AMDGPU/GlobalISel: Fix assert on invalid cond code for llvm.amdgcn.icmp · aa88b653
  Matt Arsenault authored Jan 18, 2022
  
  aa88b653
- AMDGPU/GlobalISel: Fix flat_scratch_init handling for shaders · f482e869
  Matt Arsenault authored Jan 18, 2022
  f482e869
- [AMDGPU] SILoadStoreOptimizer: Allow merging across a swizzled access · 185cb8e8
  Jay Foad authored Jan 26, 2022
  185cb8e8
- [AMDGPU] Precommit test for swizzled store aliasing two loads · c5d2b97a
  Jay Foad authored Jan 27, 2022
  
  c5d2b97a
- [AMDGPU] Update MachineMemOperands syntax in commented out tests · 15b11e00
  Jay Foad authored Jan 27, 2022
  
  15b11e00
- [AMDGPU] Remove unused CI check lines · b30d9df4
  Jay Foad authored Jan 27, 2022
  
  b30d9df4
- [AMDGPU] Remove unused GFX6 check lines · 3b259a68
  Jay Foad authored Jan 27, 2022
  
  3b259a68
Jan 26, 2022
- [AMDGPU] Prevent aliasing of SrcC and Dst in MAI · dbf278b9
  Stanislav Mekhanoshin authored Jan 21, 2022
  dbf278b9
- AMDGPU/GlobalISel: Fold wave address into mubuf addressing modes · 045be6ff
  Matt Arsenault authored Jan 12, 2022
  
  045be6ff
- GlobalISel: Avoid crash on asm with lying result types · 2d670de8
  Matt Arsenault authored Jan 19, 2022
  2d670de8
- AMDGPU/GlobalISel: Mostly fix BFI patterns · 09fc311a
  Matt Arsenault authored Jan 19, 2022
  09fc311a
- AMDGPU: Add some additional test coverage for BFI matching · eb88e793
  Matt Arsenault authored Jan 19, 2022
  eb88e793
- AMDGPU: Switch bfi pattern test to generated checks and add gfx10 · 2f33396e
  Matt Arsenault authored Jan 19, 2022
  
  2f33396e
- AMDGPU: Emit user sgpr count directives in text asm · e6564f39
  Matt Arsenault authored Jan 21, 2022
  e6564f39
- [AMDGPU] SILoadStoreOptimizer: Precommit tests for merging across a swizzled access · 0b9ee8ec
  Jay Foad authored Jan 26, 2022
  
  0b9ee8ec
- [AMDGPU][SIWholeQuadMode] Use the right VCC register to activate the correct lanes. · aa418b91
  Konstantina authored Jan 24, 2022
  aa418b91
- [AMDGPUHSAMetadataStreamer] Do not assume ABI alignment for pointers · a5e324e3
  Nikita Popov authored Jan 26, 2022
  a5e324e3
- [AMDGPU] Enable divergence-driven XNOR selection · 5157f984
  alex-t authored Dec 24, 2021
  5157f984
- [AMDGPU][GlobalISel] Combine unmerge of undef · 4723f3cf
  Sebastian Neubauer authored Jan 25, 2022
  4723f3cf
- [AMDGPU][NFC] Pre-commit regenerated test · 66804666
  Sebastian Neubauer authored Jan 25, 2022
  
  66804666
- [LSV] Vectorize loads of vectors by turning it into a larger vector · 0776f6e0
  Benjamin Kramer authored Jan 13, 2022
  0776f6e0
Jan 24, 2022
- [AMDGPU] Make v8i16/v8f16 legal · bb1fe369
  Stanislav Mekhanoshin authored Jan 19, 2022
  bb1fe369
- [AMDGPU] Remove cndmask from readsExecAsData · c27f8fb9
  Stanislav Mekhanoshin authored Jan 21, 2022
  c27f8fb9
- AMDGPU: Fix assertion on fixed stack objects with VGPR->AGPR spills · 18aabae8
  Matt Arsenault authored Jan 20, 2022
  18aabae8
- Reapply "Revert "GlobalISel: Add G_ASSERT_ALIGN hint instruction" · 99e8e173
  Matt Arsenault authored Jan 14, 2022
  99e8e173
- [AMDGPU][GlobalISel] Remove the post ':' part of vreg operands in fsh combine tests. · 912af6b5
  Abinav Puthan Purayil authored Jan 24, 2022
  
  912af6b5
- [AMDGPU][GlobalISel] Add more sign/zero/any-extension tests · aa50b93e
  Jay Foad authored Jan 21, 2022
  aa50b93e
- [AMDGPU][GlobalISel] Regenerate checks in inst-select-*ext.mir · 906ebd58
  Jay Foad authored Jan 21, 2022
  
  906ebd58
- [GlobalISel] Fold or of shifts with constant amount to funnel shift. · 68b70d17
  Abinav Puthan Purayil authored Jan 03, 2022
  68b70d17