Commits · fe0bf7d4694e3e265434fa0099b6b1cad16fa2a8 · Roger Ferrer / llvm-epi · GitLab

Feb 10, 2022
- [mlir][vector][NFC] Use CombiningKindAttr instead of StringAttr · fe0bf7d4
  Matthias Springer authored Feb 10, 2022
  fe0bf7d4
- [MLIR][NFC] Move expandAffineMap/Expr out to Affine utils · 8d12bf4a
  Uday Bondhugula authored Feb 10, 2022
  8d12bf4a
Feb 08, 2022
- [MLIR][GPU] Update GPUToROCDL to account for ControlFlow dialect · 24a1869d
  Krzysztof Drewniak authored Feb 07, 2022
  24a1869d
- [mlir][NFC] Remove a few op builders that simply swap parameter order · 3c69bc4d
  River Riddle authored Feb 06, 2022
  3c69bc4d
Feb 07, 2022
- [mlir] Use SmallBitVector instead of SmallDenseSet for AffineMap::compressSymbols · 6635c12a
  Benjamin Kramer authored Feb 06, 2022
  6635c12a
- [mlir] Fix GpuToLLVM conversion pass after ControlFlow operations were split from Standard · 330838eb
  River Riddle authored Feb 06, 2022
  
  330838eb
Feb 06, 2022
- [mlir] Split out a new ControlFlow dialect from Standard · ace01605
  River Riddle authored Feb 03, 2022
  ace01605
Feb 04, 2022
- [mlir][spirv] Add support for converting vector.shuffle · 36e68c11
  Lei Zhang authored Feb 04, 2022
  36e68c11
Feb 02, 2022
- [mlir:Standard] Remove support for creating a `unit` ConstantOp · 8e123ca6
  River Riddle authored Jan 31, 2022
  8e123ca6
- [mlir] Move SelectOp from Standard to Arithmetic · dec8af70
  River Riddle authored Jan 31, 2022
  dec8af70
- [mlir] Split std.splat into tensor.splat and vector.splat · 6a8ba318
  River Riddle authored Jan 25, 2022
  6a8ba318
- [mlir][NFC] Update AMX/LLVM/NVVM/X86 vector operations to use `hasVerifier` instead of `verifier` · 38abdddf
  River Riddle authored Feb 02, 2022
  38abdddf
- [mlir][vector] Avoid hoisting alloca'ed temporary buffers across AutomaticAllocationScope · 3c3810e7
  Nicolas Vasilache authored Feb 02, 2022
  3c3810e7
Feb 01, 2022
- Revert "Revert "[mlir] Purge `linalg.copy` and use `memref.copy` instead."" · ebc81537
  Alexander Belyaev authored Feb 01, 2022
  ebc81537
Jan 31, 2022
- [async] Get the number of worker threads from the runtime. · 149311b4
  bakhtiyar authored Jan 31, 2022
  149311b4
- Revert "[mlir] Purge `linalg.copy` and use `memref.copy` instead." · 25bf6a2a
  Alexander Belyaev authored Jan 31, 2022
  25bf6a2a
- [mlir] Purge `linalg.copy` and use `memref.copy` instead. · 016956b6
  Alexander Belyaev authored Jan 31, 2022
  016956b6
- [mlir][vector][NFC] Split into IR, Transforms and Utils · 99ef9eeb
  Matthias Springer authored Jan 31, 2022
  99ef9eeb
Jan 30, 2022
- [mlir] Silence warnings when building with MSVC · dc3b9365
  Alexandre Ganea authored Jan 30, 2022
  dc3b9365
- Apply clang-tidy fixes for readability-identifier-naming in AsyncToLLVM.cpp (NFC) · f65994c9
  Mehdi Amini authored Jan 30, 2022
  
  f65994c9
Jan 27, 2022
- [VectorToGPU] Fix horizontal stride calculation for N-D memref · a57ccad5
  Thomas Raoux authored Jan 27, 2022
  a57ccad5
- [tosa][mlir] Add dynamic shape support for remaining ops · 024a1fab
  natashaknk authored Jan 27, 2022
  024a1fab
- [mlir][complex] Lower complex.constant to LLVM · 608cc6b1
  Benjamin Kramer authored Jan 27, 2022
  608cc6b1
- [mlir] Finish replacing OwningRewritePatternList with RewritePatternSet · 9f85c198
  River Riddle authored Jan 26, 2022
  9f85c198
Jan 26, 2022
- [mlir] Move std.generic_atomic_rmw to the memref dialect · 632a4f88
  River Riddle authored Jan 25, 2022
  632a4f88
Jan 25, 2022
- [MLIR] [AsyncToLLVM] Use llvm.coro.align intrinsic · dbbe0109
  Chuanqi Xu authored Jan 25, 2022
  dbbe0109
- [mlir] Remove unnecessary dependency on Tensor from MemRef · 881ff4e4
  River Riddle authored Jan 24, 2022
  
  881ff4e4
- [mlir] Add more missing dependencies after D118062 · 633f5bad
  River Riddle authored Jan 24, 2022
  633f5bad
- [mlir] Add more missing dependencies after D118062 · e697b971
  River Riddle authored Jan 24, 2022
  e697b971
- [mlir] Add missing dependencies after D118062 · b827b634
  River Riddle authored Jan 24, 2022
  b827b634
- [mlir:Transforms] Move out the remaining non-dialect independent transforms and utilities · a70aa7bb
  River Riddle authored Jan 20, 2022
  a70aa7bb
- Fix bugs in GPUToNVVM lowering · e01e4c91
  harsh authored Jan 25, 2022
  e01e4c91
- [mlir] Add support for ExpM1 to GLSL/OpenCL SPIRV Backends · 3e746c6d
  Rob Suderman authored Jan 24, 2022
  3e746c6d
Jan 21, 2022
- [mlir] Move linalg::PadTensorOp to tensor::PadOp. · fd0c6f53
  Alexander Belyaev authored Jan 21, 2022
  fd0c6f53
- [mlir][spirv] Support size-1 vector inserts during conversion · 47107508
  Lei Zhang authored Jan 07, 2022
  47107508
Jan 20, 2022
- [mlir] Make locations required when adding/creating block arguments · e084679f
  River Riddle authored Jan 18, 2022
  e084679f
Jan 19, 2022
- [tosa][mlir] Lowering for dynamic shapes in the reduce_x ops in tosa-to-linalg · b9b10c0e
  natashaknk authored Jan 19, 2022
  b9b10c0e
- [mlir][spirv] Add math.fma lowering to spirv · d9edc1a5
  Thomas Raoux authored Jan 19, 2022
  d9edc1a5
- [mlir] Convert OpTrait::FunctionLike to FunctionOpInterface · 7ceffae1
  River Riddle authored Jan 13, 2022
  7ceffae1
- [mlir][Pass] Deprecate FunctionPass in favor of OperationPass<FuncOp> · 41574554
  River Riddle authored Jan 04, 2022
  41574554