Commits · c37414355feba2f91ffe6a4ae009db0257195369 · Roger Ferrer / llvm-epi · GitLab

Sep 14, 2023
- [mlir][sparse] remove the MLIR PyTACO tests (#66302) · 0f65df73
  Aart Bik authored Sep 13, 2023
  0f65df73
- [mlir][vector] Implement Workaround Lowerings for Masked `fm**imum` Reductions · 8f5d5194
  Daniil Dudkin authored Sep 13, 2023
  8f5d5194
- [mlir][vector] Bring back `maxf`/`minf` reductions · 709b2742
  Daniil Dudkin authored Sep 13, 2023
  709b2742
- [mlir][vector] Rename vector reductions: `maxf` → `maximumf`, `minf` → `minimumf` · 4a831250
  Daniil Dudkin authored Sep 13, 2023
  4a831250
- [mlir][sparse] remove sparse output python example (#66298) · 9918d255
  Aart Bik authored Sep 13, 2023
  9918d255
- [mlir][arith] Rename `AtomicRMWKind`'s `maxf` → `maximumf`, `minf` → `minimumf` (#66135) · c46a0433
  Daniil Dudkin authored Sep 14, 2023
  c46a0433
Sep 13, 2023
- [mlir][sparse] fix typo in sparse_tensor.unpack op (#66287) · 25059049
  Aart Bik authored Sep 13, 2023
  25059049
- [mlir] Make `extraClassOf` compile with attribute and type interfaces (#66292) · be59265b
  Markus Böck authored Sep 13, 2023
  be59265b
- [mlir][spirv][gpu] Clean up wmma to coop matrix NV conversion. NFC. (#66278) · d6d4a526
  Jakub Kuderski authored Sep 13, 2023
  d6d4a526
- [sparse] allow unpack op to return 0-ranked tensor type. (#66269) · 098f46dc
  Peiming Liu authored Sep 13, 2023
  098f46dc
- Revert `Add host-supports-nvptx requirement to lit tests` (#66102 and #66129) (#66225) · 1cddbf8c
  frgossen authored Sep 13, 2023
  
  1cddbf8c
- [mlir] Split up VectorToLLVM pass · df852599
  Krzysztof Drewniak authored Jul 18, 2023
  df852599
- [mlir][AMDGPU] Add gfx941 to buffer atomics emulation · ba6d7a0f
  Krzysztof Drewniak authored May 18, 2023
  ba6d7a0f
- [mlir][sparse] Migrate tests to use new syntax (#66146) · dbe1be9a
  Yinying Li authored Sep 13, 2023
  dbe1be9a
- [MLIR][Linalg] Bail out if the tiles provided are more than the number (#66007) · d65885ae
  lorenzo chelini authored Sep 13, 2023
  d65885ae
- Revert "[mlir][bufferization] Remove allow-return-allocs and create-deallocs... · c199f7dc
  Martin Erhart authored Sep 13, 2023
  c199f7dc
- Revert "[mlir][bufferization] Update linalg integration tests to lower ops... · 94c8faea
  Martin Erhart authored Sep 13, 2023
  94c8faea
- Revert "[mlir][bufferization] Improve buffer deallocation pass" · 520407a7
  Martin Erhart authored Sep 13, 2023
  520407a7
- Revert "[mlir][bufferization][NFC] Introduce BufferDeallocationOpInterface" · 792caac0
  Martin Erhart authored Sep 13, 2023
  792caac0
- Revert "[mlir][bufferization] BufferDeallocationOpInterface: support custom ownership update logic" · 9782232e
  Martin Erhart authored Sep 13, 2023
  9782232e
- Revert "[mlir][bufferization] Implement BufferDeallocationopInterface for scf.forall.in_parallel" · ccb16acd
  Martin Erhart authored Sep 13, 2023
  ccb16acd
- Revert "[mlir][bufferization] Define a pipeline for buffer deallocation" · 7995a470
  Martin Erhart authored Sep 13, 2023
  7995a470
- Revert "[mlir][bufferization] Remove unused function" · 99cd1e70
  Martin Erhart authored Sep 13, 2023
  99cd1e70
- [MLIR] Fix the tma_load test (#66208) · ba81cd10
  Guray Ozen authored Sep 13, 2023
  ba81cd10
- [MLIR] SM_90 integration test of TMA with 128b Swizzling (#65953) · 38faecc6
  Guray Ozen authored Sep 13, 2023
  38faecc6
- [MLIR][NVVM] Fix the register number of predicate (#65970) · 6360d095
  Guray Ozen authored Sep 13, 2023
  6360d095
- [mlir] Add pass to add comdat to all linkonce functions (#65270) · a6857156
  David Truby authored Sep 13, 2023
  a6857156
- [mlir][IR] SingleBlockImplicitTerminator derives from TraitBase (#66189) · 076e238c
  Matthias Springer authored Sep 13, 2023
  076e238c
- [mlir][bufferization] Generalize tensor slice rules to subset ops (#65619) · 8143307b
  Matthias Springer authored Sep 13, 2023
  8143307b
- [mlir][bufferization] Remove unused function · cb5fe6ce
  Martin Erhart authored Sep 13, 2023
  
  cb5fe6ce
- [mlir][bufferization] Define a pipeline for buffer deallocation · f0c46639
  Martin Erhart authored Sep 12, 2023
  f0c46639
- [mlir][bufferization] Implement BufferDeallocationopInterface for scf.forall.in_parallel · 1356e853
  Martin Erhart authored Sep 12, 2023
  1356e853
- [mlir][bufferization] BufferDeallocationOpInterface: support custom ownership update logic · 89117f18
  Martin Erhart authored Sep 12, 2023
  89117f18
- [mlir][bufferization][NFC] Introduce BufferDeallocationOpInterface · 29d86175
  Martin Erhart authored Sep 12, 2023
  29d86175
- [mlir][bufferization] Improve buffer deallocation pass · 1bebb60a
  Martin Erhart authored Sep 12, 2023
  1bebb60a
- [mlir][bufferization] Update linalg integration tests to lower ops created by... · 6f35401f
  Martin Erhart authored Sep 12, 2023
  6f35401f
- [mlir][bufferization] Remove allow-return-allocs and create-deallocs pass... · 6a91dfed
  Martin Erhart authored Sep 12, 2023
  6a91dfed
- [sparse] allow unpack op to return any integer type. (#66161) · 64df1c08
  Peiming Liu authored Sep 12, 2023
  
  64df1c08
Sep 12, 2023
- [mlir][spirv] Fix remaining coop matrix verification corner cases (#66137) · 4c4bdf0c
  Jakub Kuderski authored Sep 12, 2023
  4c4bdf0c
- [mlir][gpu][sparse] gracefully accept zero size allocation (#66127) · 3635c743
  Aart Bik authored Sep 12, 2023
  3635c743