Commits · e2398a4d7cfc0415c63cc13792bda80045c7c803 · Roger Ferrer / llvm-epi · GitLab

Sep 24, 2022
- [mlir][spirv] Switch to kEmitAccessorPrefix_Predixed · 90a1632d
  Jakub Kuderski authored Sep 24, 2022
  90a1632d
- [mlir][nvgpu] Use TableGen TypeDef for NVGPU dialect types · afb0b21f
  Christopher Bate authored Sep 23, 2022
  afb0b21f
- [mlir][spirv] Switch to kEmitAccessorPrefix_Both · 250f9e09
  Jakub Kuderski authored Sep 23, 2022
  250f9e09
Sep 23, 2022
- [mlir] NFC: move mergeOffsetsSizesAndStrides into Affine/Utils · 465ec4e0
  Lei Zhang authored Sep 23, 2022
  465ec4e0
- [mlir] Plumb missing paramter to gpu transform op · e99f4371
  Thomas Raoux authored Sep 23, 2022
  e99f4371
- [mlir] Add map_nested_foreach_thread_to_gpu_blocks op to transform dialect · f7907bc5
  Guray Ozen authored Sep 23, 2022
  f7907bc5
- [mlir][memref] fix overflow in realloc · 8b587113
  Peiming Liu authored Sep 23, 2022
  8b587113
- [mlir] Add `parseEllipsis` · ac74c51c
  Jeff Niu authored Sep 16, 2022
  ac74c51c
- [mlir][sparse] temporarily disable push back for asan · 560bcc3d
  Aart Bik authored Sep 22, 2022
  560bcc3d
- [mlir][sparse] Introducing a new sparse_tensor.foreach operator. · e08865a1
  Peiming Liu authored Sep 22, 2022
  e08865a1
- [mlir][sparse] Mark sparse_tensor.concatenate as NoSideEffect · 1ab2b007
  Peiming Liu authored Sep 22, 2022
  1ab2b007
- [mlir][Linalg] Expose the implementation of the tiling to scf.foreach_thread. · acc2a12c
  Mahesh Ravishankar authored Sep 22, 2022
  acc2a12c
Sep 22, 2022
- Reland "[mlir][tensor] Support more cases in MergeConsecutiveExtractSlice" · bd81524e
  Lei Zhang authored Sep 22, 2022
  bd81524e
- [mlir][MemRef] Simplify extract_strided_metadata(expand_shape) · d0aeb74e
  Quentin Colombet authored Sep 10, 2022
  d0aeb74e
- [mlir] Modify LinalgStructuredInterface to allow the computation block to have... · 91563c41
  Oleg Shyshkov authored Sep 22, 2022
  91563c41
- [mlir][vector] Fix test for vector-warp-distribute · cb3d0260
  Christopher Bate authored Sep 22, 2022
  cb3d0260
- [mlir][sparse] Add codegen rule for the push_back operator. · 4132bce9
  bixia1 authored Sep 21, 2022
  4132bce9
- [mlir] Fix a cast that should be a dyn_cast. · eaf20c4f
  Johannes Reifferscheid authored Sep 22, 2022
  eaf20c4f
- [MLIR] Revert default NVIDIA GPU version · e01c7f09
  Ivan Radanov Ivanov authored Sep 22, 2022
  e01c7f09
- [mlir] Flip PDL to use Both accessors · 72fddfb5
  River Riddle authored Sep 21, 2022
  72fddfb5
- [mlir] Flip Async/GPU/OpenACC/OpenMP to use Both accessors · 986b5c56
  River Riddle authored Sep 21, 2022
  986b5c56
- [MLIR] Set default NVIDIA GPU version · f9211330
  Ivan Radanov Ivanov authored Sep 21, 2022
  
  f9211330
- [MLIR] Add a create function for mlir::SerializeToCubinPass · 2f7a774e
  Ivan Radanov Ivanov authored Sep 21, 2022
  2f7a774e
Sep 21, 2022
- [mlir][LLVMIR] Do not create pseudo debug file name using llvm::Instruction · 006a752a
  Min Hsu authored Sep 19, 2022
  006a752a
- [mlir][tosa] Revert added support for dynamic height/weight for pooling in tosa-to-linalg · 3e41bec0
  natashaknk authored Sep 21, 2022
  3e41bec0
- [mlir][llvm] Support pointer entries in data layout translation · aa00e3e6
  rkayaith authored Sep 07, 2022
  aa00e3e6
- [mlir][memref] Add realloc op. · 9f13b934
  bixia1 authored Sep 21, 2022
  9f13b934
- [mlir] Move getDimsOfType to StructuredOpsUtils.h. · 9926ad17
  Oleg Shyshkov authored Sep 21, 2022
  9926ad17
- [mlir] Allow negative strides and offset in StridedLayoutAttr · 54d81e49
  Ivan Butygin authored Sep 18, 2022
  54d81e49
- [mlir][tensor][bufferize] Implement getBufferType for Expand/CollapseShapeOp · 04ff6009
  Matthias Springer authored Sep 21, 2022
  04ff6009
- [mlir] Update tests to use kEmitAccessorPrefix_Prefixed · 9879261d
  River Riddle authored Sep 20, 2022
  9879261d
- [mlir][NFC] Tidy up the doc comment on CastInfo for Types · 019e2075
  River Riddle authored Sep 20, 2022
  019e2075
- [mlir] Update Attributes to use the new casting infra · ecb17da8
  River Riddle authored Sep 20, 2022
  ecb17da8
- Revert "[mlir][tensor] Support more cases in MergeConsecutiveExtractSlice" · e0a6df53
  Mehdi Amini authored Sep 21, 2022
  e0a6df53
- [mlir][SCF] Fix loop pipelining unable to handle ops with regions · f5fe92f6
  Christopher Bate authored Sep 20, 2022
  f5fe92f6
- [mlir][tensor] NFC: name various Transforms/ files consistently · 2d3b54fe
  Lei Zhang authored Sep 20, 2022
  2d3b54fe
- [mlir][tensor] Support more cases in MergeConsecutiveExtractSlice · 5d4603a0
  Lei Zhang authored Sep 20, 2022
  5d4603a0
- [mlir][tensor] Merge consecutive insert_slice/extract_slice ops · bb4c53b7
  Lei Zhang authored Sep 20, 2022
  bb4c53b7
- [mlir][arith] Fix constant naming in integration tests. NFC. · 41f3602d
  Jakub Kuderski authored Sep 20, 2022
  41f3602d
- [mlir][arith] Add integration tests for addi emulation · 7d299333
  Jakub Kuderski authored Sep 20, 2022
  7d299333