aartbik
Follow
Popular repositories
-
-
llvm-project Public
Forked from llvm/llvm-project
The LLVM Project is a collection of modular and reusable compiler and toolchain technologies. Note: the repository does not accept github pull requests at this moment. Please submit your patches at…
-
stablehlo Public
Forked from openxla/stablehlo
Backward compatible ML compute opset inspired by HLO/MHLO
MLIR
-
-
torch-mlir Public
Forked from llvm/torch-mlir
The Torch-MLIR project aims to provide first class support from the PyTorch ecosystem to the MLIR ecosystem.
C++
-
iree Public
Forked from openxla/iree
A retargetable MLIR-based machine learning compiler and runtime toolkit.
C++ 1
529 contributions in the last year
Contribution activity
September 2023
Created 22 commits in 2 repositories
Created 1 repository
- aartbik/iree C++
Created a pull request in openxla/iree that received 9 comments
[iree][sparse] allow sparse tensors in input IR
Just a "hello sparse world" example that allows sparse tensors in the input IR. Nothing else is being done with the type yet other than proper roun…
Opened 21 other pull requests in 2 repositories
llvm/llvm-project
20
merged
- [mlir][sparse][gpu] protect BSR method with cuda 12.1
- [mlir][sparse][gpu] add CSC to libgen GPU sparsification using cuSparse
- [mlir][sparse][gpu] add CSC and BSR format to cuSparse GPU ops
- [mlir][sparse] update BSR specification
- [mlir][sparse] make high and 2:4 a level format, not a level property
- [mlir][sparse] add lvlToDim field to sparse tensor encoding
- [mlir][sparse] replace "sparse compiler" with "sparsifier" in doc
- [mlir][sparse] update doc for new surface syntax encoding
- [mlir][sparse] remove most bufferization.alloc_tensor ops from sparse
- [mlir][sparse][gpu] free all buffers allocated for spGEMM
- [mlir][sparse] More allocate -> empty tensor migration
- [mlir][sparse] fix merge conflict in sddmm test
- [mlir][sparse] refine sparse fusion with empty tensors materialization
- [mlir][sparse][tensor] replace bufferization with empty tensor
- [mlir][sparse][gpu] refine type of workspace size variables
- [mlir][sparse] deprecate the convert{To,From}MLIRSparseTensor methods
- [mlir][sparse] remove the MLIR PyTACO tests
- [mlir][sparse] remove sparse output python example
- [mlir][sparse] fix typo in sparse_tensor.unpack op
- [mlir][gpu][sparse] gracefully accept zero size allocation
openxla/stablehlo
1
merged
Reviewed 25 pull requests in 4 repositories
llvm/llvm-project
22 pull requests
- [mlir][sparse] rename sparse_tensor.(un)pack to sparse_tensor.(dis)as…
- [bazel] fix bazel for (D152457)
- [mlir][sparse] Change tests to use new syntax for ELL and slice
- [bazel] fix bazel
- [mlir][sparse] Generalize sparse encoding in check tests
- [bazel] fix bazel
- [mlir][sparse] Treat high and 2OutOf4 as level formats
- [mlir][sparse] update doc for new surface syntax encoding
- [mlir][sparse] Add more error messages and avoid crashing in new parser
- [mlir][sparse] Generates python bindings for SparseTensorTransformOps.
- [mlir][sparse] remove most bufferization.alloc_tensor ops from sparse
- [mlir][sparse] unifies sparse_tensor.sort_coo/sort into one operation.
- [mlir][sparse] fix logical error when generating sort_coo.
- [mlir][sparse] Migrate tests to use new syntax
- [mlir][sparse] Migrate more tests to use new syntax
- [mlir][gpu][NVPTX] Enable NVIDIA GPU JIT compilation path
- [mlir][sparse] Migrate more tests to new syntax
- [mlir][gpu] Deprecate gpu::Serialization* passes.
- [sparse] allow unpack op to return 0-ranked tensor type.
- [sparse] allow unpack op to return any integer type.
- [mlir][sparse] Migrate tests to use new syntax
- [mlir][sparse] Fix bug in new syntax parser
openxla/stablehlo
1 pull request
openai/triton
1 pull request
aartbik/iree
1 pull request
Started 1 discussion in 1 repository
openxla/xla
-
[RFC] Add 2:4 structured sparsity support to XLA:GPU
This contribution was made on Sep 25