Commits · 9932561b4892b6e9bbb0c2369272dfff2305fdb9 · Lorenzo Albano / LLVM bpEVL · GitLab

Sep 18, 2020
- [mlir][shape] Extend shape.cstr_require with a message. · 7c446513
  Sean Silva authored Sep 17, 2020
  7c446513
- [mlir][openacc] Add missing operands for acc.data operation · 88a1d402
  Valentin Clement authored Sep 18, 2020
  88a1d402
- [mlir][openacc] Support Index and AnyInteger in loop op · 22dde1f9
  Valentin Clement authored Sep 18, 2020
  22dde1f9
- [mlir][Linalg] Evolve named ops to use assembly form and support linalg on tensors. · 93fd30ba
  Nicolas Vasilache authored Sep 18, 2020
  93fd30ba
- [mlir][shape] Add `shape.cstr_require %bool` · bae63742
  Sean Silva authored Sep 17, 2020
  bae63742
Sep 17, 2020
- [MLIR] Support for return values in Affine.For yield · 296e97ae
  Abhishek Varma authored Sep 17, 2020
  296e97ae
- [mlir][Vector] Add a folder for vector.broadcast · f16abe5f
  Hanhan Wang authored Sep 17, 2020
  f16abe5f
- [mlir][openacc] Change operand type from index to AnyInteger in parallel op · 6d3cabd9
  Valentin Clement authored Sep 17, 2020
  6d3cabd9
- [mlir][Standard] Canonicalize chains of tensor_cast operations · 5e0ded26
  Stephan Herhut authored Sep 16, 2020
  5e0ded26
- [mlir][Linalg] Convolution tiling added to ConvOp vectorization pass · 347d59b1
  Jakub Lichman authored Sep 17, 2020
  347d59b1
- [mlir][spirv] Add GroupNonUniformBroadcastOp · 4ce84b0e
  Artur Bialas authored Sep 16, 2020
  4ce84b0e
Sep 16, 2020
- [mlir][openacc] Add missing operands for acc.parallel operation · d9953d15
  Valentin Clement authored Sep 16, 2020
  d9953d15
Sep 15, 2020
- [mlir][openacc] Add loop op verifier · 01f5fcd8
  Valentin Clement authored Sep 15, 2020
  01f5fcd8
- [mlir][Standard] Add canonicalizer for dynamic_tensor_from_elements · c897a7fb
  Stephan Herhut authored Sep 14, 2020
  c897a7fb
Sep 14, 2020
- Make namespace handling uniform across dialect backends. · 7d1ed69c
  Federico Lebrón authored Sep 14, 2020
  7d1ed69c
- Add support for casting elements in vectors for certain Std dialect type conversion operations. · ef7a255c
  Lubomir Litchev authored Sep 09, 2020
  ef7a255c
Sep 11, 2020
- [mlir] Fix some edge cases around 0-element TensorFromElementsOp · 84a6da67
  Sean Silva authored Sep 10, 2020
  84a6da67
- [mlir][spirv] OpConvertSToF support operands with different bitwidth. · aeb43143
  Xin Wang authored Sep 11, 2020
  aeb43143
- [NFC] Follow up on D87111 - Add an option for unrolling loops up to a factor - CR issues addressed. · 32062478
  Lubomir Litchev authored Sep 08, 2020
  32062478
- [MLIR][Linalg] Undo spurious parameter name change · 1851bab1
  Nicolas Vasilache authored Sep 11, 2020
  
  1851bab1
- [mlir][Linalg] Refactor StructuredOpInterface - NFC · e6f2f17f
  Nicolas Vasilache authored Sep 11, 2020
  e6f2f17f
- [mlir][Linalg] Print both types for linalg.transpose · a0e0d30a
  Benjamin Kramer authored Sep 10, 2020
  a0e0d30a
- [mlir] Fix generation of AVX512 dialect documentation · a68673cc
  Marius Brehler authored Sep 10, 2020
  a68673cc
- [mlir][Linalg] Make LinalgBaseTilingPattern not delete the original operation. · d380b582
  MaheshRavishankar authored Sep 10, 2020
  d380b582
Sep 10, 2020
- Introduce linalg.vecmat · 5638df19
  Eugene Burmako authored Sep 10, 2020
  5638df19
- [MLIR][Standard] Simplify `tensor_from_elements` · 018f6936
  Frederik Gossen authored Sep 10, 2020
  018f6936
- [mlir][Linalg] Small refactoring of ConvOpVectorization · fea175b5
  Jakub Lichman authored Sep 10, 2020
  fea175b5
- [mlir][Linalg] Add Utility method to get loop ranges for a LinalgOp. · a7b2977a
  MaheshRavishankar authored Sep 09, 2020
  a7b2977a
Sep 09, 2020
- [mlir][Linalg] Reduction dimensions specified in TC definition of ConvOps. · 53ffeea6
  Jakub Lichman authored Sep 08, 2020
  53ffeea6
- [MLIR][Shape] Lower `shape_of` to `dynamic_tensor_from_elements` · 5106a8b8
  Frederik Gossen authored Sep 09, 2020
  5106a8b8
- [MLIR][Standard] Update `tensor_from_elements` assembly format · 133322d2
  Frederik Gossen authored Sep 09, 2020
  133322d2
Sep 08, 2020
- Add an option for unrolling loops up to a factor. · e2394245
  Lubomir Litchev authored Sep 03, 2020
  e2394245
- [mlir][VectorOps] Clean up outdated comments. NFCI. · 307dc7b2
  Benjamin Kramer authored Sep 08, 2020
  307dc7b2
- [mlir] Conv ops vectorization pass · 67b37f57
  Jakub Lichman authored Sep 08, 2020
  67b37f57
Sep 07, 2020
- [MLIR][Standard] Add `dynamic_tensor_from_elements` operation · 136eb79a
  Frederik Gossen authored Sep 07, 2020
  136eb79a
- [mlir][Vector] Revisit VectorToSCF. · 8d64df9f
  Nicolas Vasilache authored Sep 04, 2020
  8d64df9f
Sep 04, 2020
- [spirv] Add more target and resource limit fields · 7d53fecb
  Lei Zhang authored Sep 03, 2020
  7d53fecb
- [mlir] introduce type constraints for operands of LLVM dialect operations · aec9e20a
  Alex Zinenko authored Sep 04, 2020
  aec9e20a
Sep 03, 2020
- [spirv][nfc] Simplify resource limit with default values · 8d420fb3
  Lei Zhang authored Sep 03, 2020
  8d420fb3
- [spirv] Add SPIR-V GLSL extended Round op · 1e21ca4d
  Zhibin Li authored Sep 03, 2020
  1e21ca4d