Commits · e27197f3605450c372ddc71922d0e9982b30e115 · Roger Ferrer / llvm-epi · GitLab

Jan 22, 2021
- [mlir][spirv] Define spv.IsNan/spv.IsInf and add lowerings · e27197f3
  Lei Zhang authored Jan 22, 2021
  e27197f3
- [mlir][spirv] Fix script for availability autogen and refresh ops · 167fb9b4
  Lei Zhang authored Jan 22, 2021
  167fb9b4
- [mlir] Add coro intrinsics operations to LLVM dialect · cc77a2c7
  Eugene Zhulenev authored Jan 22, 2021
  cc77a2c7
- [mlir][StandardOps] Fix typos in the td file. · 1b535df1
  Hanhan Wang authored Jan 22, 2021
  1b535df1
- [MLIR] Add support for extracting an integer sample point (if one exists) from... · 14056dfb
  Arjun P authored Jan 22, 2021
  14056dfb
- [mlir][StandardToSPIRV] Add support for lowering uitofp to SPIR-V · 2cb130f7
  Hanhan Wang authored Jan 21, 2021
  2cb130f7
- [mlir][Linalg] Introduce linalg.pad_tensor op. · 16d4bbef
  Hanhan Wang authored Jan 21, 2021
  16d4bbef
- [mlir] Enable passing crash reproducer stream factory method · aee622fa
  Jacques Pienaar authored Jan 21, 2021
  aee622fa
- [mlir] Support FuncOpSignatureConversion for more FunctionLike ops. · 0a7a1ac7
  mikeurbach authored Jan 18, 2021
  0a7a1ac7
Jan 21, 2021
- Add Python bindings for the builtin dialect · 922b26cd
  Mehdi Amini authored Jan 20, 2021
  922b26cd
- Revert [mlir] Link mlir_runner_utils statically into cuda/rocm-runtime-wrappers (cf50f4f7 ) · bd3a387e
  Christian Sigg authored Jan 21, 2021
  bd3a387e
- Remove deprecated methods from OpState. · 8827e07a
  Christian Sigg authored Jan 21, 2021
  8827e07a
- [mlir]][SPIRV] Define OrderedOp and UnorderedOp and add lowerings from Standard. · 615167c9
  MaheshRavishankar authored Jan 20, 2021
  615167c9
- [mlir][SPIRV] Rename OpSpecConstantOperation -> OpSpecConstantOp · 4234292e
  MaheshRavishankar authored Jan 20, 2021
  4234292e
- Add log1p lowering from standard to ROCDL intrinsics · 4ef38f9c
  Frederik Gossen authored Jan 21, 2021
  4ef38f9c
- Add log1p lowering from standard to NVVM intrinsics · 294e2544
  Frederik Gossen authored Jan 21, 2021
  294e2544
- [mlir] Remove complex ops from Standard dialect. · fc58bfd0
  Alexander Belyaev authored Jan 20, 2021
  fc58bfd0
- [mlir][OpFormatGen] Fix incorrect kind used for RegionsDirective · 825c2b4a
  River Riddle authored Jan 20, 2021
  825c2b4a
- [mlir] Make MLIRContext::getOrLoadDialect(StringRef, TypeID, ...) public · 8a7ff730
  mfehr authored Jan 21, 2021
  8a7ff730
- [mlir] Add a new builtin `unrealized_conversion_cast` operation · c78219f6
  River Riddle authored Jan 20, 2021
  c78219f6
- [mlir] Add an interface for Cast-Like operations · 6ccf2d62
  River Riddle authored Jan 20, 2021
  6ccf2d62
Jan 20, 2021
- Revert "[mlir][Affine] Add support for multi-store producer fusion" · 735a07f0
  Diego Caballero authored Jan 21, 2021
  735a07f0
- [mlir][sparse] add asserts on reading in tensor data · 5959c28f
  Aart Bik authored Jan 20, 2021
  5959c28f
- [mlir] NFC - Fix unused variable in non-debug mode · 555a395f
  Nicolas Vasilache authored Jan 20, 2021
  
  555a395f
- [mlir:async] Fix data races in AsyncRuntime · a2223b09
  Eugene Zhulenev authored Jan 20, 2021
  a2223b09
- [mlir][Linalg] NFC - Fully compose map and operands when creating AffineMin in tiling. · 8dd58a50
  Nicolas Vasilache authored Jan 20, 2021
  8dd58a50
- [mlir] Add ComplexDialect to SCF->GPU pass. · b1e1bbae
  Alexander Belyaev authored Jan 20, 2021
  
  b1e1bbae
- [mlir] Fix SubTensorInsertOp semantics · 866cb260
  Nicolas Vasilache authored Jan 20, 2021
  866cb260
- [mlir][Linalg] NFC - Expose getSmallestBoundingIndex as an utility function · c0755726
  Nicolas Vasilache authored Jan 20, 2021
  
  c0755726
- [mlir][Linalg] NFC - getAssumedNonShapedOperands now returns OperandRange · f5d8eb08
  Nicolas Vasilache authored Jan 20, 2021
  f5d8eb08
- [MLIR][Standard] Add log1p operation to std · cc4244d5
  Frederik Gossen authored Jan 20, 2021
  cc4244d5
- [mlir] fix the rocm runtime wrapper to account for cuda / rocm api differences · 4c1eaf26
  Tobias Gysi authored Jan 20, 2021
  4c1eaf26
- Avoid unused variable warning in opt mode · cad16e4a
  Jacques Pienaar authored Jan 20, 2021
  
  cad16e4a
- [mlir][Affine] Add support for multi-store producer fusion · 7dd19885
  Diego Caballero authored Jan 20, 2021
  7dd19885
- Fix cuda-runner tests. · cba1ca90
  Christian Sigg authored Jan 20, 2021
  
  cba1ca90
- Added check if there are regions that do not implement the RegionBranchOpInterface. · 43f34f58
  Julian Gross authored Jan 13, 2021
  43f34f58
- [mlir] Link mlir_runner_utils statically into cuda/rocm-runtime-wrappers. · cf50f4f7
  Christian Sigg authored Jan 11, 2021
  cf50f4f7
- [mlir][sparse] add narrower choices for pointers/indices · b5c542d6
  Aart Bik authored Jan 19, 2021
  b5c542d6
- [mlir][python] Swap shape and element_type order for MemRefType. · b62c7e04
  Stella Laurenzo authored Jan 19, 2021
  b62c7e04
- Implement constant folding for DivFOp · 1bf2b166
  Jackson Fellows authored Jan 19, 2021
  1bf2b166