Commits · 27fca57546c2828e2684c02b7aa677cbd6603bfd · Raul Torres / llvm-target-spread · GitLab

Jun 03, 2020
- [mlir][Linalg] Add support for fusion between indexed_generic ops and tensor_reshape ops · 27fca575
  Hanhan Wang authored Jun 03, 2020
  27fca575
- [mlir][Linalg] Add support for fusion between indexed_generic ops and generic ops on tensors. · cc11ceda
  Hanhan Wang authored Jun 03, 2020
  cc11ceda
- [mlir] [VectorOps] Use 'vector.flat_transpose' for 2-D 'vector.tranpose' · 6391da98
  aartbik authored Jun 03, 2020
  6391da98
- [mlir] Remove unneeded inference trait/fns · 5b454b98
  Jacques Pienaar authored Jun 03, 2020
  5b454b98
- [MLIR] Shape to standard dialect lowering · 3713314b
  Frederik Gossen authored Jun 03, 2020
  3713314b
- [mlir] DialectConversion: remove vtable from TypeConverter · 6163fa79
  Alex Zinenko authored Jun 02, 2020
  6163fa79
- [mlir][Linalg] NFC - Make markers use Identifier instead of StringRef · e349fb70
  Nicolas Vasilache authored Jun 02, 2020
  e349fb70
- Fix build: TableGen uses `is<T>` instead of `isa<T>` as predicate · 48c800cc
  Mehdi Amini authored Jun 03, 2020
  
  48c800cc
- Replace dyn_cast<>() with isa<>() when the result isn't used (NFC) · a09bb6d7
  Mehdi Amini authored Jun 03, 2020
  a09bb6d7
- [mlir][spirv] Fix coop matrix getExtension · 81dd3a47
  Thomas Raoux authored Jun 02, 2020
  81dd3a47
- [mlir][spirv] Fix encoding of cooperative matrix type to match SPIRV spec · bbe79e27
  Thomas Raoux authored Jun 02, 2020
  bbe79e27
- [mlir][Affine] Enable fusion of loops with vector loads/stores · 8a418e5f
  Diego Caballero authored Jun 03, 2020
  8a418e5f
- [mlir] NFC - Add AffineMinMaxOpBase getDim/SymbolOperands · 7de45878
  Nicolas Vasilache authored Jun 02, 2020
  
  7de45878
Jun 02, 2020
- [mlir][spirv] Add support for matrix type · 915e55c9
  HazemAbdelhafez authored Jun 02, 2020
  915e55c9
- [mlir] Provide defaults to make enabling dumping simpler · 635cde6e
  Jacques Pienaar authored Jun 02, 2020
  635cde6e
- [mlir] support materialization for 1-1 type conversions · 5c5dafc5
  Alex Zinenko authored Jun 02, 2020
  5c5dafc5
- [mlir] SCFToGPUPass: fix macros referring to LOOPS to use SCF instead · eb8edd85
  Alex Zinenko authored Jun 02, 2020
  eb8edd85
- [mlir] Toy tutorial: avoid erasing and then re-creating loop terminators · b596ecdd
  Alex Zinenko authored Jun 02, 2020
  b596ecdd
- [mlir] post-commit review fixes · 195d8571
  Alex Zinenko authored Jun 02, 2020
  195d8571
- [mlir] Introduce CallOp converter for buffer placement · 3f6a35e3
  Ehsan Toosi authored May 26, 2020
  3f6a35e3
- [mlir][SCFToGPU] Remove conversions from scf.for to gpu.launch. · 2bcd1927
  MaheshRavishankar authored Jun 01, 2020
  2bcd1927
May 31, 2020
- Change some extraneous /// comments to // comments inside methods. NFC. · 0cf5ef17
  Chris Lattner authored May 31, 2020
  
  0cf5ef17
May 30, 2020
- Use .empty() instead of .size() == 0 (NFC) · 21fee092
  Mehdi Amini authored May 30, 2020
  21fee092
- [mlir][spirv] Clean up coop matrix assembly declaration. · c652c306
  Thomas Raoux authored May 29, 2020
  c652c306
- [mlir] Fix Windows build · 7fc225f8
  Nicolas Vasilache authored May 29, 2020
  7fc225f8
- [mlir] NFC - Add debug information for Linalg transformations. · 91beb517
  Nicolas Vasilache authored May 29, 2020
  91beb517
May 29, 2020
- [mlir][Affine] Minor clean-up of D79829 · e75325cf
  Diego Caballero authored May 29, 2020
  e75325cf
- [mlir][Linalg] Make contraction vectorization use vector transfers · 9534192c
  Nicolas Vasilache authored May 29, 2020
  9534192c
- [MLIR][OpenMP] Defined master operation in OpenMP Dialect · dbb5979d
  Anchu Rajendran authored May 28, 2020
  dbb5979d
- [mlir][Linalg][Vector] Add forwarding patterns between linalg.copy and vector.transfer · 1ee11432
  Nicolas Vasilache authored May 29, 2020
  1ee11432
- [mlir][SCF] Add utility to clone an scf.ForOp while appending new yield values. · aa93659c
  Nicolas Vasilache authored May 29, 2020
  aa93659c
- Unbreak the build of mlir-cuda-runner · b9bb3ad3
  Benjamin Kramer authored May 29, 2020
  
  b9bb3ad3
- [MLIR][BufferPlacement] Support functions that return Memref typed results · 7a3a2535
  Ehsan Toosi authored May 20, 2020
  7a3a2535
- [mlir] Add test to check if standalone dialect is registered · b0b25077
  Marius Brehler authored May 29, 2020
  b0b25077
May 28, 2020
- [mlir][Vector] Fix vector.transfer alignment calculation · 5f9e0466
  Nicolas Vasilache authored May 28, 2020
  5f9e0466
- [SVE] Eliminate calls to default-false VectorType::get() from mlir · a9c7b498
  Christopher Tetreault authored May 28, 2020
  a9c7b498
- [mlir] Extend standalone example by standalone-translate · 3bff62d4
  Marius Brehler authored May 28, 2020
  3bff62d4
- [MLIR] Fix build when NVPTX is not enabled · 8b315582
  Stephen Neuendorffer authored May 28, 2020
  8b315582
- [mlir][Linalg] Add pass to remove unit-extent dims from tensor · 2b0c8546
  MaheshRavishankar authored May 28, 2020
  2b0c8546
- [mlir][GPU] Link relevant LLVM components in GPUCommon instead of test · 72ede60b
  Alex Zinenko authored May 28, 2020
  72ede60b