Erick Ochoa Lopez 9d19250610

[mlir][vector] Add vector.to_elements unrolling (#157142) ...

This PR adds support for unrolling `vector.to_element`'s source operand.

It transforms

```mlir
%0:8 = vector.to_elements %v : vector<2x2x2xf32>
```

to

```mlir
%v0 = vector.extract %v[0] : vector<2x2xf32> from vector<2x2x2xf32>
%v1 = vector.extract %v[1] : vector<2x2xf32> from vector<2x2x2xf32>
%0:4 = vector.to_elements %v0 : vector<2x2xf32>
%1:4 = vector.to_elements %v1 : vector<2x2xf32>
// %0:8 = %0:4 - %1:4
```

This pattern will be applied until there are only 1-D vectors left.

---------

Signed-off-by: hanhanW <hanhan0912@gmail.com>
Co-authored-by: hanhanW <hanhan0912@gmail.com>
Co-authored-by: Jakub Kuderski <kubakuderski@gmail.com>

2025-09-11 13:56:57 -04:00

..

gpu

[mlir][target][nvvm] Perf by stage and store into properties (#126178)

2025-02-11 12:58:58 +01:00

linalg

[MLIR][Linalg] Remove matmul_transpose variants (#147961)

2025-08-08 22:20:27 +01:00

sparse_tensor

[mlir python] Port in-tree dialects to nanobind. (#119924)

2024-12-20 20:32:32 -08:00

affine.py

[mlir][Affine] Let affine.[de]linearize_index omit outer bounds (#116103)

2024-11-18 15:41:54 -06:00

amdgpu.py

[MLIR][AMDGPU] Use Attr for resetOffset + boundsCheck in RawBufferCastOp (#149939)

2025-07-21 16:43:57 -07:00

arith_dialect.py

…

arith_llvm.py

…

async_dialect.py

…

builtin.py

Reapply "[mlir][python] allow DenseIntElementsAttr for index type (#118947)" (#124804)

2025-01-29 09:14:37 +01:00

cf.py

…

complex_dialect.py

…

emitc_dialect.py

[mlir][emitc] Add support for C-API/python binding to EmitC dialect (#119476)

2024-12-11 10:07:21 -08:00

func.py

[mlir,python] Fix case when FuncOp.arg_attrs is not set (#117188)

2024-12-02 08:55:51 -08:00

index_dialect.py

…

llvm.py

…

math_dialect.py

…

memref.py

…

ml_program.py

…

nvgpu.py

…

nvvm.py

[MLIR][NVVM] Improve inline_ptx, add readwrite support (#154358)

2025-08-21 17:42:18 +02:00

ods_helpers.py

…

openmp_ops.py

…

pdl_ops.py

…

pdl_types.py

…

python_test.py

[MLIR][Python] Add optional results parameter for building op with inferable result types (#156818)

2025-09-04 17:57:39 -07:00

quant.py

Sub-channel quantized type implementation (#120172)

2025-03-23 07:37:55 -05:00

rocdl.py

[mlir][python] fix value-builder generation for snake_case ops (#135302)

2025-04-11 08:55:38 -04:00

scf.py

[MLIR][SCF] Add dedicated Python bindings for ForallOp (#149416)

2025-07-18 19:53:11 -04:00

shape.py

…

smt.py

[mlir][SMT] add python bindings (#135674)

2025-04-16 18:17:09 -04:00

spirv_dialect.py

…

tensor.py

…

tosa.py

…

transform_bufferization_ext.py

…

transform_debug_ext.py

[MLIR][Transform] expose transform.debug extension in Python (#145550)

2025-06-25 16:39:01 +01:00

transform_extras.py

…

transform_gpu_ext.py

…

transform_interpreter.py

…

transform_loop_ext.py

…

transform_memref_ext.py

…

transform_nvgpu_ext.py

…

transform_sparse_tensor_ext.py

…

transform_structured_ext.py

[MLIR][Python] Add structured.fuseop to python interpreter (#120601)

2025-01-03 11:21:59 +00:00

transform_tensor_ext.py

…

transform_tune_ext.py

[MLIR][Transform] Introduce transform.tune.knob op (#146732)

2025-07-08 11:00:34 +01:00

transform_vector_ext.py

[mlir][vector] Add vector.to_elements unrolling (#157142)

2025-09-11 13:56:57 -04:00

transform.py

[MLIR][Transform] apply_registered_pass: support ListOptions (#144026)

2025-06-16 12:40:50 +01:00

ub.py

[mlir][python] UB dialect python bindings (#157127)

2025-09-05 18:52:01 +03:00

vector.py

…