Matthias Springer 77f8297c6f [mlir][sparse] Improve sparse tensor type constraints (#112133) ...

Sparse tensors are always ranked tensors. Encodings cannot be attached
to unranked tensors. Change the type constraint to `RankedTensorOf`, so
that we generate `TypedValue<RankedTensorType>` instead of
`TypedValue<TensorType>`. This removes the need for type casting in some
cases.

Also improve the verifiers (missing `return` statements) and switch a
few other `AnyTensor` to `AnyRankedTensor`.

This commit is in preparation of a dialect conversion commit that
required fixes in the sparse dialect.

2024-10-13 21:12:38 +02:00

..

GPU

[mlir][sparse] use a consistent order between [dis]assembleOp and sto… (#84079)

2024-03-06 09:57:41 -08:00

binary_valued.mlir

[mlir][sparse] force a properly sized view on pos/crd/val under codegen (#91288)

2024-05-07 09:20:56 -07:00

buffer_rewriting.mlir

[mlir][sparse] renaming sparse_tensor.sort_coo to sparse_tensor.sort (#68161)

2023-10-03 16:28:25 -07:00

canonicalize.mlir

[mlir][sparse] add canonicalization patterns for IterateOp. (#95569)

2024-06-14 10:31:29 -07:00

codegen_buffer_initialization.mlir

[mlir][sparse] implement direct IR alloc/empty/new for non-permutations (#72585)

2023-11-16 17:17:41 -08:00

codegen_sparse_alloc.mlir

[mlir][sparse] Migrate tests to use new syntax (#66543)

2023-09-15 16:12:20 -04:00

codegen_to_llvm.mlir

[mlir][sparse] Migrate tests to use new syntax (#66146)

2023-09-13 11:41:25 -04:00

codegen.mlir

[mlir][sparse][test] Adjust tests for LowerSparseOpsToForeach (#110976)

2024-10-11 15:56:39 +08:00

constant_index_map.mlir

[mlir][sparse] reuse tensor.insert operation to insert elements into … (#84987)

2024-03-12 16:59:17 -07:00

conversion.mlir

[mlir][llvm] Implement ConstantLike for ZeroOp, UndefOp, PoisonOp (#93690)

2024-05-30 08:21:08 +02:00

convert_dense2sparse.mlir

[mlir][sparse] deallocate tmp coo buffer generated during stage-spars… (#82017)

2024-02-17 12:17:57 -08:00

convert_sparse2dense.mlir

[mlir][sparse] split post-sparsification-rewriting into two passes. (#70727)

2023-10-30 15:22:21 -07:00

convert_sparse2sparse_element.mlir

[mlir][sparse] Migrate tests to use new syntax (#66146)

2023-09-13 11:41:25 -04:00

convert_sparse2sparse.mlir

[mlir][sparse] implements tensor.insert on sparse tensors. (#70737)

2023-10-30 16:04:41 -07:00

dense.mlir

[mlir][sparse] setup SparseIterator to help generating code to traverse a sparse tensor level. (#78345)

2024-01-24 11:33:06 -08:00

external_direct.mlir

[mlir][sparse] allow for direct-out passing of sparse tensor buffers (#88327)

2024-04-11 10:07:24 -07:00

external.mlir

[mlir][sparse] use a consistent order between [dis]assembleOp and sto… (#84079)

2024-03-06 09:57:41 -08:00

fold.mlir

[mlir][sparse] Use variable instead of inlining sparse encoding (#72561)

2023-11-16 19:30:21 -05:00

fuse_sparse_concat_with_extract_slice.mlir

[mlir][sparse] fuse concat and extract_slice op if possible. (#89825)

2024-04-24 13:51:41 -07:00

fuse_sparse_convert_into_producer.mlir

[mlir][sparse] clone a empty sparse tensor when fuse convert into pro… (#92158)

2024-05-14 13:26:49 -07:00

fuse_sparse_pad_with_consumer.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

invalid_encoding.mlir

[mlir][sparse] Add verification for explicit/implicit value (#90111)

2024-05-07 20:28:39 -04:00

invalid.mlir

[mlir][sparse] Improve sparse tensor type constraints (#112133)

2024-10-13 21:12:38 +02:00

minipipeline_parallel.mlir

[mlir][sparse] add parallelization options to mini pipeline (#104233)

2024-08-30 17:47:37 -04:00

minipipeline_vector.mlir

[mlir][sparse] expose optimization flags to mini pipeline (#95158)

2024-06-11 14:20:58 -07:00

no_lowering.mlir

[mlir][sparse][test] Adjust tests for LowerSparseOpsToForeach (#110976)

2024-10-11 15:56:39 +08:00

one_shot_bufferize_invalid.mlir

Changed the phrase sparse-compiler to sparsifier in comments (#71578)

2023-11-07 20:55:00 +00:00

one_shot_bufferize_tensor_copy_insertion.mlir

[mlir][linalg][bufferize] Fix element-wise access optimization for sparse tensors (#87305)

2024-04-03 09:57:25 +09:00

one_trip.mlir

[mlir][sparse] Use variable instead of inlining sparse encoding (#72561)

2023-11-16 19:30:21 -05:00

pack_copy.mlir

[mlir][sparse] use a consistent order between [dis]assembleOp and sto… (#84079)

2024-03-06 09:57:41 -08:00

pre_rewriting.mlir

[mlir][sparse] Use variable instead of inlining sparse encoding (#72561)

2023-11-16 19:30:21 -05:00

rejected.mlir

[mlir][sparse] Use variable instead of inlining sparse encoding (#72561)

2023-11-16 19:30:21 -05:00

rewriting_for_codegen.mlir

[mlir][sparse] Use variable instead of inlining sparse encoding (#72561)

2023-11-16 19:30:21 -05:00

roundtrip_encoding.mlir

[mlir][sparse] allow multiple COO segments in sparse encodings. (#91786)

2024-05-10 11:36:01 -07:00

roundtrip.mlir

[mlir][sparse] introduce sparse_tensor.coiterate operation. (#101100)

2024-07-31 15:14:27 -07:00

scf_1_N_conversion.mlir

[mlir][sparse] Migrate tests to use new syntax (#66146)

2023-09-13 11:41:25 -04:00

semi_ring.mlir

[mlir][sparse] Use variable instead of inlining sparse encoding (#72561)

2023-11-16 19:30:21 -05:00

sorted_coo.mlir

[mlir][sparse] setup SparseIterator to help generating code to traverse a sparse tensor level. (#78345)

2024-01-24 11:33:06 -08:00

sparse_1d.mlir

[mlir][Transforms] GreedyPatternRewriteDriver: Do not CSE constants during iterations (#75897)

2024-01-05 09:22:18 +01:00

sparse_2d.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_3d.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_affine.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_batch.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_broadcast.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_concat.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_conv_2d_slice_based.mlir

[mlir][sparse] reuse tensor.insert operation to insert elements into … (#84987)

2024-03-12 16:59:17 -07:00

sparse_expand.mlir

[mlir][sparse] schedule sparse kernels in a separate pass from sparsification. (#72423)

2023-11-15 12:16:05 -08:00

sparse_extract_slice.mlir

[mlir][sparse] Use variable instead of inlining sparse encoding (#72561)

2023-11-16 19:30:21 -05:00

sparse_fill_zero.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_foreach.mlir

[mlir][sparse] setup SparseIterator to help generating code to traverse a sparse tensor level. (#78345)

2024-01-24 11:33:06 -08:00

sparse_fp_ops.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_fusion.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_index.mlir

[mlir][sparse] reuse tensor.insert operation to insert elements into … (#84987)

2024-03-12 16:59:17 -07:00

sparse_int_ops.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_iteration_to_scf.mlir

[mlir][sparse] expose emit strategy option to mini pipeline (#96238)

2024-06-20 14:23:36 -07:00

sparse_itertion_licm.mlir

[mlir][sparse] introduce sparse_tensor.iterate operation (#88955)

2024-06-10 10:20:24 -07:00

sparse_kernels_to_iterator.mlir

[mlir][sparse] partially support lowering sparse coiteration loops to scf.while/for. (#105565)

2024-08-23 10:47:44 -07:00

sparse_kernels.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_lower_col.mlir

[mlir] Remove dialect specific bufferization passes (Reland) (#93535)

2024-05-28 20:04:27 +01:00

sparse_lower_inplace.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_lower.mlir

[mlir] Remove dialect specific bufferization passes (Reland) (#93535)

2024-05-28 20:04:27 +01:00

sparse_matmul_codegen.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_matmul_one.mlir

[mlir][sparse] Support explicit/implicit value for complex type (#90771)

2024-05-02 12:28:34 -04:00

sparse_nd.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_out.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_outbuf.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_pack.mlir

[mlir][sparse] use a consistent order between [dis]assembleOp and sto… (#84079)

2024-03-06 09:57:41 -08:00

sparse_parallel_reduce.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_parallel.mlir

[mlir][sparse] schedule sparse kernels in a separate pass from sparsification. (#72423)

2023-11-15 12:16:05 -08:00

sparse_perm_lower.mlir

[mlir][sparse] setup SparseIterator to help generating code to traverse a sparse tensor level. (#78345)

2024-01-24 11:33:06 -08:00

sparse_perm.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_reinterpret_map.mlir

[mlir][sparse] reuse tensor.insert operation to insert elements into … (#84987)

2024-03-12 16:59:17 -07:00

sparse_relu.mlir

[mlir][sparse] recognize ReLu operation during sparsification (#92016)

2024-05-13 14:02:29 -07:00

sparse_reshape.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_scalars.mlir

[mlir][sparse] Use variable instead of inlining sparse encoding (#72561)

2023-11-16 19:30:21 -05:00

sparse_sddmm_org.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_sddmm.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_space_collapse.mlir

[mlir][sparse] implement lowering rules for ExtractIterSpaceOp. (#89143)

2024-06-12 10:49:12 -07:00

sparse_storage.mlir

[mlir][Transforms] GreedyPatternRewriteDriver: Do not CSE constants during iterations (#75897)

2024-01-05 09:22:18 +01:00

sparse_tensor_reshape.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_transpose.mlir

[mlir][sparse] reuse tensor.insert operation to insert elements into … (#84987)

2024-03-12 16:59:17 -07:00

sparse_vector_chain.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_vector_concat.mlir

Changed all code and comments that used the phrase "sparse compiler" to instead use "sparsifier" (#71875)

2023-11-15 20:12:35 +00:00

sparse_vector_index.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

sparse_vector_mv.mlir

[mlir][sparse] made sparse vectorizer more robust on position of invariants (#80766)

2024-02-05 16:12:47 -08:00

sparse_vector_ops.mlir

[mlir][sparse] schedule sparse kernels in a separate pass from sparsification. (#72423)

2023-11-15 12:16:05 -08:00

sparse_vector_peeled.mlir

[mlir][sparse] schedule sparse kernels in a separate pass from sparsification. (#72423)

2023-11-15 12:16:05 -08:00

sparse_vector.mlir

[mlir][sparse] schedule sparse kernels in a separate pass from sparsification. (#72423)

2023-11-15 12:16:05 -08:00

specifier_to_llvm.mlir

[mlir][llvm] Implement ConstantLike for ZeroOp, UndefOp, PoisonOp (#93690)

2024-05-30 08:21:08 +02:00

spy_sddmm_bsr.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00

spy_sddmm.mlir

[mlir][sparse] Use variable instead of inlining sparse encoding (#72561)

2023-11-16 19:30:21 -05:00

torch_linalg.mlir

[mlir][sparse] ensure [dis]assembler wrapper methods properly inline (#81907)

2024-02-15 11:39:32 -08:00

transform-ops.mlir

[mlir] introduce debug transform dialect extension (#77595)

2024-01-12 13:24:02 +01:00

unsparsifiable_dense_op.mlir

[mlir][sparse] schedule sparse kernels in a separate pass from sparsification. (#72423)

2023-11-15 12:16:05 -08:00

unused-tensor.mlir

[mlir][sparse] Use variable instead of inlining sparse encoding (#72561)

2023-11-16 19:30:21 -05:00

vectorize_reduction.mlir

[mlir][sparse] support sparsifying sparse kernels to sparse-iterator-based loop (#95858)

2024-06-17 16:50:12 -07:00