Commits · 660f281b5e755663b2d8ad061075eb5f5abdf9fb · Raul Torres / llvm-target-spread · GitLab

Sep 15, 2021
- [mlir][Linalg] Make codegen strategy late transformations opt-in · 660f281b
  Nicolas Vasilache authored Sep 15, 2021
  660f281b
- [mlir][Linalg] Replace DenseSet by UnionFind in ComprehensiveBufferize - NFC · e3889b30
  Nicolas Vasilache authored Sep 15, 2021
  e3889b30
- [mlir][linalg] ComprehensiveBufferize: Do not copy InitTensorOp results · 934e2f69
  Matthias Springer authored Sep 15, 2021
  934e2f69
- [mlir] Update docs on conversion and translation to LLVM · b10940ed
  Alex Zinenko authored Sep 10, 2021
  b10940ed
- [mlir][linalg] Update OpDSL doc (NFC). · a543abc5
  Tobias Gysi authored Sep 15, 2021
  a543abc5
- Make the --mlir-disable-threading command line option overrides the C++ API usage · a32300a6
  Mehdi Amini authored Sep 15, 2021
  a32300a6
- [MLIR] Use memref.copy ops in BufferResultsToOutParams pass. · 500d4c45
  cwz920716 authored Sep 15, 2021
  500d4c45
- [mlir][linalg] PadTensorOp vectorization: Avoid redundant FillOps · 9adc0114
  Matthias Springer authored Sep 15, 2021
  9adc0114
- Remove unused llvm/Support/Parallel.h from MLIR (NFC) · 1a406cd5
  Mehdi Amini authored Sep 14, 2021
  1a406cd5
Sep 14, 2021
- [mlir] Apply py::module_local() to a few more classes. · 8dca953d
  Sean Silva authored Sep 14, 2021
  8dca953d
- [mli][linalg] Reuse getValueOrCreateConstantIndexOp method (NFC). · 60918736
  Tobias Gysi authored Sep 14, 2021
  60918736
- [mlir][linalg] Fold ExtractSliceOps during tiling. · 44a88977
  Tobias Gysi authored Sep 14, 2021
  44a88977
- [MLIR] Improve op parse error message for AtLeastNOperands trait · a91cfd19
  Uday Bondhugula authored Sep 14, 2021
  a91cfd19
- [mlir][linalg] makeTiledShape: No affine.min if tile size == 1 · 62883459
  Matthias Springer authored Sep 14, 2021
  62883459
- [mlir][linalg] New tiling option: Scalarize dynamic dims · fb1def9c
  Matthias Springer authored Sep 14, 2021
  fb1def9c
- [mlir][linalg] Add scf.for loop peeling to codegen strategy · 8faf35c0
  Matthias Springer authored Sep 14, 2021
  8faf35c0
- [mlir][linalg] TiledLoopOp peeling: Do not peel partial iterations · a4a654d3
  Matthias Springer authored Sep 14, 2021
  a4a654d3
Sep 13, 2021
- [mlir][doc] fix typos. · ce6d5120
  Jian Cai authored Sep 13, 2021
  ce6d5120
- Reorder mmt4d shapes: · 340314c4
  Benoit Jacob authored Sep 13, 2021
  340314c4
- [mlir][Linalg] Insert static buffers as high as possible during ComprehensiveBufferization. · 181d18ef
  Nicolas Vasilache authored Sep 13, 2021
  181d18ef
- [mlir][emitc] Print signed integers properly · ec92f788
  Simon Camphausen authored Sep 13, 2021
  ec92f788
- [MLIR] Mark test case XFAIL on SystemZ for now. · 5f781ddf
  Jonas Paulsson authored Sep 13, 2021
  5f781ddf
- [mlir][linalg] ComprehensiveBufferize: Do not copy InitTensorOps · 7c9b6a33
  Matthias Springer authored Sep 13, 2021
  7c9b6a33
- [mlir][Linalg] Use reify for padded op shape derivation. · b01d223f
  Nicolas Vasilache authored Sep 10, 2021
  b01d223f
- [mlir] Add missing namespace to createInlinerPass · 57bf8560
  Valentin Clement authored Sep 13, 2021
  57bf8560
- [mlir] Allows to query traits from types and attributes · 802bf02a
  Mathieu Fehr authored Sep 13, 2021
  802bf02a
- Add sanity check in MLIR ODS to catch case where an... · 7fb2394a
  Mehdi Amini authored Sep 09, 2021
  7fb2394a
Sep 12, 2021
- [OpenMP][MLIR] Add a conversion pattern for the master op · 187d9f8c
  Kiran Chandramohan authored Sep 12, 2021
  187d9f8c
Sep 10, 2021
- [mlir][tosa] Add shape inference for tosa.while · b0532286
  Rob Suderman authored Aug 26, 2021
  b0532286
- [mlir] spelling and style changes in ReconcileUnrealizedCasts.cpp. NFC. · 61bc6aa5
  Alex Zinenko authored Sep 10, 2021
  
  61bc6aa5
- [mlir][linalg] Fix bufferize pattern to allow unknown operations in body of generic · 5e6c170b
  Stephan Herhut authored Sep 10, 2021
  5e6c170b
- [mlir][scf] Loop peeling: Use scf.for for partial iteration · 0f3544d1
  Matthias Springer authored Sep 10, 2021
  0f3544d1
- [mlir][linalg] Pass all operands to tile to the tile loop region builder (NFC). · 16488dc3
  Tobias Gysi authored Sep 10, 2021
  16488dc3
- [mlir][Linalg] Properly order extract_slice traversal in comprehensive bufferization · 5f1a1af4
  Nicolas Vasilache authored Sep 09, 2021
  5f1a1af4
- [mlir] Replace `include_directories` · 6593cd3f
  Marius Brehler authored Sep 09, 2021
  6593cd3f
- [mlir][tosa] Add lowering for tosa.clz using scf::whileOp · d4d50e47
  natashaknk authored Sep 09, 2021
  d4d50e47
- [mlir][sparse] add folding to sparse_tensor.convert · 066d786c
  Aart Bik authored Sep 09, 2021
  066d786c
Sep 09, 2021
- [mlir][sparse] Mark convert op as noSideEffect · 2a69790b
  thomasraoux authored Sep 09, 2021
  2a69790b
- Adding min(f/s/u) and max(f/s/u) cases for vector reduction · 89837a0e
  Alexander Slepko authored Sep 09, 2021
  89837a0e
- [APInt] Normalize naming on keep constructors / predicate methods. · 735f4671
  Chris Lattner authored Sep 08, 2021
  735f4671