Commits · 1a2b3536efef20f12c44201c2834a383b7c5c4c2 · Roger Ferrer / llvm-epi · GitLab

Feb 19, 2021
- [mlir][sparse] assert fail on mismatch between rank and annotations array · 2556d622
  Aart Bik authored Feb 18, 2021
  2556d622
- [mlir] Load dynamic libraries in JitRunner from absolute paths so that GDB can... · c86c96a7
  Christian Sigg authored Feb 18, 2021
  c86c96a7
- Reland "[MLIR] Make structured op tests permutation invariant" · db011775
  Geoffrey Martin-Noble authored Feb 18, 2021
  db011775
- Add DivOp to the Shape dialect · d690cbf8
  Jing Pu authored Feb 18, 2021
  d690cbf8
- Revert "[MLIR] Make structured op tests permutation invariant" · cd8cc00b
  Mehdi Amini authored Feb 19, 2021
  cd8cc00b
Feb 18, 2021
- [MLIR] Make structured op tests permutation invariant · b9ff6709
  Geoffrey Martin-Noble authored Feb 18, 2021
  b9ff6709
- [mlir] Fold trivial subtensor / subtensor_insert ops. · b006902b
  Nicolas Vasilache authored Feb 18, 2021
  b006902b
- [mlir][Vector] Fold tensor_cast + vector.transfer_read · 8e01e2ec
  Nicolas Vasilache authored Feb 18, 2021
  8e01e2ec
- Perform memory accesses in the same addrspace as the corresponding memref. · 08c681f6
  Andrew Pritchard authored Feb 18, 2021
  08c681f6
- [MLIR][TOSA] Add lowering from TOSA to Linalg for math-based and elementwise ops · 25b4a6a7
  natashaknk authored Feb 18, 2021
  25b4a6a7
- [mlir] Add `linalg.tiled_loop` op. · 624fccba
  Alexander Belyaev authored Feb 18, 2021
  624fccba
- Revert "[MLIR] Create memref dialect and move several dialect-specific ops from std." · a89035d7
  Alexander Belyaev authored Feb 18, 2021
  a89035d7
- [MLIR] Create memref dialect and move several dialect-specific ops from std. · 8aa6c376
  Julian Gross authored Feb 10, 2021
  8aa6c376
- [mlir] generate enum translation functions with unused attribute · 12875ed9
  Alex Zinenko authored Feb 17, 2021
  12875ed9
- [mlir][sparse] generalize sparse storage format to many more types · ff6c84b8
  Aart Bik authored Feb 17, 2021
  ff6c84b8
Feb 17, 2021
- [mlir] silence unused-function warnings in table-generated code · 4a3473ff
  Alex Zinenko authored Feb 15, 2021
  4a3473ff
- [mlir] Add fma operation to std dialect · 519f5917
  Eugene Zhulenev authored Feb 17, 2021
  519f5917
- [mlir][StandardToSPIRV] Add support for lowering trunci to SPIR-V to i1 types. · c80484e1
  Hanhan Wang authored Feb 17, 2021
  c80484e1
- [mlir][spirv] Add spv.GLSL.FrexpStruct · 77426206
  Weiwei Li authored Feb 17, 2021
  77426206
- [mlir][Shape] Generalize cstr_broadcastable folding for n-ary broadcasts · 63a35f35
  Benjamin Kramer authored Feb 16, 2021
  63a35f35
- [mlir][python] Add python binding for AffineMapAttribute. · 4c3f1be8
  Stella Laurenzo authored Feb 16, 2021
  4c3f1be8
Feb 16, 2021
- [mlir][Linalg] Add utility method to reshape ops to express output shape in terms of input shape. · 81264dfb
  MaheshRavishankar authored Feb 16, 2021
  81264dfb
- separate AffineMapAccessInterface from AffineRead/WriteOpInterface · 99c0458f
  Adam Straw authored Feb 16, 2021
  99c0458f
- [mlir] Fix memref_cast + subview folder when reducing rank · adfd3c70
  Thomas Raoux authored Feb 16, 2021
  adfd3c70
- [mlir] Simplify ModuleTranslation for LLVM IR · ce8f10d6
  Alex Zinenko authored Feb 16, 2021
  ce8f10d6
- [mlir] tighten LLVM dialect verifiers to generate valid LLVM IR · 2ab57c50
  Alex Zinenko authored Feb 16, 2021
  2ab57c50
- [mlir] add verifiers for NVVM and ROCDL kernel attributes · 9cd47a26
  Alex Zinenko authored Feb 16, 2021
  9cd47a26
- [mlir][vector] Add missing support for contract of integer lowering. · 397336dc
  Thomas Raoux authored Feb 10, 2021
  397336dc
- [mlir] Add canonicalization for tensor_cast + tensor_to_memref · 807e5467
  Thomas Raoux authored Feb 15, 2021
  807e5467
- [mlir][vector] Move splitting transfer ops into a separate entry point · cb1a4235
  Lei Zhang authored Feb 16, 2021
  cb1a4235
- [mlir][vector] Add support for unrolling vector.fma · d8c7f442
  Lei Zhang authored Feb 16, 2021
  d8c7f442
- [mlir] Don't return nullptrs from scf::IfOp::getSuccessorRegions · 787d771d
  Tres Popp authored Feb 16, 2021
  787d771d
- [mlir][Linalg] Generalize vector::transfer hoisting on tensors. · 21debeae
  Nicolas Vasilache authored Feb 16, 2021
  21debeae
- Lower math.expm1 to intrinsics in the GPUToNVVM and GPUToROCDL conversions. · 07cc7718
  Adrian Kuegel authored Feb 16, 2021
  07cc7718
- [mlir] Drop reliance of SliceAnalysis on specific ops. · d01ea0ed
  Nicolas Vasilache authored Feb 15, 2021
  d01ea0ed
Feb 15, 2021
- [mlir][Vector] Add a canonicalization pattern for vector.contract + add · 02d053ed
  Nicolas Vasilache authored Feb 15, 2021
  02d053ed
- [mlir] Add clone method to ShapedType · 381a65fa
  Jacques Pienaar authored Feb 15, 2021
  381a65fa
- Make shape.is_broadcastable/shape.cstr_broadcastable nary · 3842d4b6
  Tres Popp authored Feb 10, 2021
  3842d4b6
- [mlir] Use the interface-based translation for LLVM "intrinsic" dialects · 176379e0
  Alex Zinenko authored Feb 12, 2021
  176379e0
- [mlir] Add error message on shape.broadcast verification failure · 89d900b2
  Tres Popp authored Feb 15, 2021
  
  89d900b2