shylie/llvm-project
1
0
Fork 0
Code Issues Pull Requests Actions 6 Packages Projects Releases Wiki Activity
llvm-project/mlir/test/Dialect/Linalg
History
Martin Erhart 6bf043e743
[mlir][bufferization] Remove allow-return-allocs and create-deallocs pass options, remove bufferization.escape attribute (#66619) 
This commit removes the deallocation capabilities of
one-shot-bufferization. One-shot-bufferization should never deallocate
any memrefs as this should be entirely handled by the
ownership-based-buffer-deallocation pass going forward. This means the
`allow-return-allocs` pass option will default to true now,
`create-deallocs` defaults to false and they, as well as the escape
attribute indicating whether a memref escapes the current region, will
be removed. A new `allow-return-allocs-from-loops` option is added as a
temporary workaround for some bufferization limitations.
2023-09-18 16:44:48 +02:00
..
affine.mlir
[mlir][Linalg] NFC - Retire LinalgToLLVM pass
2023-07-26 20:26:07 +02:00
bubble-up-extract-slice-op.mlir
[mlir] Refactor alias generation to support nested aliases
2022-10-23 23:59:55 -07:00
bufferize.mlir
[mlir][tensor][NFC] Rename linalg.init_tensor to tensor.empty
2022-10-04 17:25:35 +09:00
canonicalize.mlir
[mlir][Linalg] Move linalg.fill -> linalg.pack pattern into fill canonicalization patterns. (#66002)
2023-09-11 13:41:38 -07:00
collapse-dim.mlir
[mlir][Linalg] Avoid collapsing dimensions of linalg op that arent foldable.
2023-05-10 05:45:02 +00:00
conv-interface-invalid.mlir
Revert "Revert "[mlir][linalg] Replace "string" iterator_types attr with enums in LinalgInterface.""
2022-11-11 10:54:08 +01:00
convert-conv2d-to-img2col.mlir
[mlir] Add ComplexType conversion support for convertScalarToDtype
2023-07-17 14:00:58 -07:00
convert-elementwise-to-linalg.mlir
[mlir][Pass] Include anchor op in -pass-pipeline
2022-11-03 11:36:12 -04:00
data-layout-propagation.mlir
[mlir][Linalg] Move linalg.fill -> linalg.pack pattern into fill canonicalization patterns. (#66002)
2023-09-11 13:41:38 -07:00
decompose-ops.mlir
[mlir][NFC] Remove trailing whitespaces from *.td and *.mlir files.
2022-11-28 15:26:30 -08:00
detensorize_0d.mlir
[mlir][NFC] Remove trailing whitespaces from *.td and *.mlir files.
2022-11-28 15:26:30 -08:00
detensorize_br_operands.mlir
[mlir][NFC] Remove trailing whitespaces from *.td and *.mlir files.
2022-11-28 15:26:30 -08:00
detensorize_entry_block.mlir
[mlir] detensorize: don't accidentally convert function entry blocks
2023-04-25 03:33:54 +00:00
detensorize_if.mlir
[mlir][NFC] Remove trailing whitespaces from *.td and *.mlir files.
2022-11-28 15:26:30 -08:00
detensorize_trivial.mlir
[mlir][Pass] Include anchor op in -pass-pipeline
2022-11-03 11:36:12 -04:00
detensorize_while_impure_cf.mlir
[mlir][NFC] Remove trailing whitespaces from *.td and *.mlir files.
2022-11-28 15:26:30 -08:00
detensorize_while_pure_cf.mlir
[mlir] detensorize: don't accidentally convert function entry blocks
2023-04-25 03:33:54 +00:00
detensorize_while.mlir
[mlir][NFC] Remove trailing whitespaces from *.td and *.mlir files.
2022-11-28 15:26:30 -08:00
drop-unit-extent-dims.mlir
[mlir][sparse] Migrate more tests to new syntax (#66309)
2023-09-14 12:21:13 -04:00
erase-unused-operands-and-results.mlir
[mlir][linalg] Fold duplicate and unused inputs in linalg.generic
2022-12-09 15:18:26 +01:00
fill-interface-invalid.mlir
…
forward-vector-transfers.mlir
…
fuse-with-reshape-by-collapsing.mlir
[mlir][NFC] Remove trailing whitespaces from *.td and *.mlir files.
2022-11-28 15:26:30 -08:00
fusion-elementwise-ops.mlir
[mlir][arith] Rename operations: maxf → maximumf, minf → minimumf (#65800)
2023-09-11 22:02:19 -07:00
fusion-elementwise-options.mlir
[mlir][Linalg] Avoid unnecessary propagating producer result to fused op result.
2022-11-22 07:08:17 +00:00
fusion-elementwise.mlir
[mlir][Linalg] Avoid unnecessary propagating producer result to fused op result.
2022-11-22 07:08:17 +00:00
fusion-multiuse-producer.mlir
[mlir][Linalg] Make Elementwise op fusion return a map from existing values to values in the fused op.
2023-01-31 20:33:44 +00:00
fusion-push-reshape.mlir
[mlir][NFC] Remove trailing whitespaces from *.td and *.mlir files.
2022-11-28 15:26:30 -08:00
generalize-named-ops.mlir
[mlir][arith] Rename operations: maxf → maximumf, minf → minimumf (#65800)
2023-09-11 22:02:19 -07:00
generalize-named-polymorphic-ops.mlir
[mlir][arith] Rename operations: maxf → maximumf, minf → minimumf (#65800)
2023-09-11 22:02:19 -07:00
generalize-pad-tensor.mlir
[MLIR][Linalg] (NFC) Improve RUN command in generalize-pad-tensor.mlir
2023-05-31 18:28:17 +02:00
generalize-tensor-pack-tile.mlir
[MLIR][Linalg] Retire tile_to_scf_for (#65633)
2023-09-07 16:13:23 -04:00
generalize-tensor-pack.mlir
[mlir][tensor] More static padding sizes for PadOp
2023-06-30 14:04:33 +02:00
generalize-tensor-unpack-tile.mlir
[MLIR][Linalg] Retire tile_to_scf_for (#65633)
2023-09-07 16:13:23 -04:00
generalize-tensor-unpack.mlir
[mlir][linalg] Allow outer dims perm and untiled dims in pack/unpack generalization
2023-05-02 12:26:45 -04:00
hoisting.mlir
[mlir][vector] Refine vector.transfer_read hoisting/forwarding (#65770)
2023-09-12 10:33:58 +01:00
inline-scalar-operands.mlir
[mlir][NFC] Remove trailing whitespaces from *.td and *.mlir files.
2022-11-28 15:26:30 -08:00
inlining.mlir
…
int-narrowing.mlir
[mlir][arith] Add narrowing patterns for index casts
2023-05-03 13:55:02 -04:00
invalid.mlir
[mlir][Linalg] Clarify error message in YieldOp verification NFC
2023-08-07 22:12:05 -07:00
library-calls.mlir
[MLIR][Linalg] Generate unique LibraryCallName for LinalgOps.
2023-03-20 13:50:22 -04:00
loops.mlir
[mlir][Linalg] NFC - Retire LinalgToLLVM pass
2023-07-26 20:26:07 +02:00
match-ops-interpreter.mlir
[mlir][Transform] Add classify_convolution_dims op
2023-07-31 16:15:25 -04:00
match-ops-invalid.mlir
[mlir] generalize matchers to support batch matmul
2023-07-07 14:44:44 +00:00
match-ops.mlir
[mlir] add structured (Linalg) transform op matchers
2023-04-13 12:37:51 +00:00
matmul-shared-memory-padding.mlir
[mlir][linalg] Optional dealloc insertion for bufferize_to_allocation (#65610)
2023-09-07 17:49:48 +02:00
multisize-tiling-full.mlir
[mlir] update types in remaining Linalg TransformOps test
2023-05-16 08:16:59 +00:00
named-ops-fail.mlir
[MLIR][Linalg] Add max named op to linalg
2023-07-07 13:39:12 +01:00
named-ops.mlir
Finish renaming getOperandSegmentSizeAttr() from operand_segment_sizes to operandSegmentSizes
2023-08-09 19:37:01 -07:00
namedop_conversion.mlir
[mlir][linalg] Propagate attributes when doing named ops conversion.
2022-09-15 11:44:52 -07:00
one-shot-bufferize-analysis-2fill-extract-matmul-all-perms.mlir
[mlir][linalg][NFC] Cleanup: Drop linalg.inplaceable attribute
2022-07-14 15:50:03 +02:00
one-shot-bufferize-analysis.mlir
[mlir][bufferization] Remove allow-return-allocs and create-deallocs pass options, remove bufferization.escape attribute (#66619)
2023-09-18 16:44:48 +02:00
one-shot-bufferize-empty-tensor-elimination.mlir
[mlir][linalg] LinalgOp-anchored empty tensor elimination
2023-07-03 09:17:48 +02:00
one-shot-bufferize.mlir
[mlir][bufferization] Remove allow-return-allocs and create-deallocs pass options, remove bufferization.escape attribute (#66619)
2023-09-18 16:44:48 +02:00
pad_fusion.mlir
[mlir][Affine] Delete duplicate code: applyMapToValues
2023-06-30 14:01:13 +02:00
pad-to-specific-memory-space.mlir
[mlir][linalg] Optional dealloc insertion for bufferize_to_allocation (#65610)
2023-09-07 17:49:48 +02:00
parallel-loops.mlir
[mlir][NFC] Remove trailing whitespaces from *.td and *.mlir files.
2022-11-28 15:26:30 -08:00
promote.mlir
[mlir][Linalg]: Add memory space to linalg transform::PromoteOp
2023-09-07 17:35:32 +03:00
promotion_options.mlir
[mlir] make structured transform ops use types
2023-05-16 08:16:56 +00:00
reshape_control_fusion.mlir
[mlir][tensor][NFC] Rename linalg.init_tensor to tensor.empty
2022-10-04 17:25:35 +09:00
reshape_fusion.mlir
[mlir][NFC] Remove trailing whitespaces from *.td and *.mlir files.
2022-11-28 15:26:30 -08:00
resolve-shaped-type-result-dims.mlir
[mlir][Affine] Delete duplicate code: applyMapToValues
2023-06-30 14:01:13 +02:00
roundtrip.mlir
[mlir][Linalg] NFC - Retire LinalgToLLVM pass
2023-07-26 20:26:07 +02:00
standard.mlir
[mlir][Linalg] Fix crash in LinalgToStandard
2023-01-20 00:17:55 -08:00
subtensor-of-padtensor.mlir
…
swap-extract-slice-with-fill.mlir
[mlir][linalg] Swap tensor.extract_slice(linalg.fill)
2022-09-20 17:31:22 -04:00
test-drop-unit-dims.mlir
[mlir][Linalg] Cleanup the drop unit dims pass in Linalg.
2023-07-19 17:47:18 +00:00
tile-and-fuse-tensors.mlir
[MLIR][Tensor] Add Destination style RewritePattern for DimOp. (#65780)
2023-09-09 06:01:56 -04:00
tile-conv.mlir
[mlir] update types in remaining Linalg TransformOps test
2023-05-16 08:16:59 +00:00
tile-indexed.mlir
[mlir] update types in remaining Linalg TransformOps test
2023-05-16 08:16:59 +00:00
tile-softmax.mlir
[mlir][Linalg] Implement the tiling interface for softmax
2023-07-03 10:49:16 +02:00
tile-tensors.mlir
[mlir] update types in remaining Linalg TransformOps test
2023-05-16 08:16:59 +00:00
tile-to-forall.mlir
[MLIR][Linalg] Rename tile-to-foreach-thread.mlir (NFC)
2023-06-21 18:44:24 +02:00
transform-lower-pack.mlir
[mlir][linalg] Fix lowering of tensor.pack operations
2023-09-05 21:08:13 +02:00
transform-op-bufferize-to-allocation.mlir
[mlir][linalg] Optional dealloc insertion for bufferize_to_allocation (#65610)
2023-09-07 17:49:48 +02:00
transform-op-compose-masked-vectorize-and-cleanups.mlir
[mlir][linalg][nfc] Fix typo in a file name
2023-09-15 15:31:36 +00:00
transform-op-decompose.mlir
[mlir][arith] Rename operations: maxf → maximumf, minf → minimumf (#65800)
2023-09-11 22:02:19 -07:00
transform-op-fuse-into-containing.mlir
[mlir][arith] Rename operations: maxf → maximumf, minf → minimumf (#65800)
2023-09-11 22:02:19 -07:00
transform-op-fuse.mlir
[mlir][tensor] Add producer fusion for tensor.pack op.
2023-08-16 11:02:59 -07:00
transform-op-generalize.mlir
[mlir] make structured transform ops use types
2023-05-16 08:16:56 +00:00
transform-op-gpu-map-copy-to-threads.mlir
[mlir][transforms] Revamp the implementation of mapping loops to GPUs
2023-07-26 00:09:08 +02:00
transform-op-hoist-pad-build-packing-loop-nest.mlir
[MLIR][Linalg] Retire tile_to_scf_for (#65633)
2023-09-07 16:13:23 -04:00
transform-op-hoist-pad.mlir
[MLIR][Linalg] Retire tile_to_scf_for (#65633)
2023-09-07 16:13:23 -04:00
transform-op-insert-slice-to-copy.mlir
[mlir][Linalg] Support tensor.parallel_insert_slice in transform.insert_slice_to_copy
2023-04-14 06:11:29 -07:00
transform-op-interchange.mlir
[mlir] make structured transform ops use types
2023-05-16 08:16:56 +00:00
transform-op-match.mlir
[mlir] transform.structured.match fix tilingIface condition (#65337)
2023-09-05 18:02:33 +02:00
transform-op-matmul-to-outerproduct.mlir
[mlir][transform] Improve transform.get_closest_isolated_parent
2023-07-04 16:23:08 +02:00
transform-op-multitile-sizes.mlir
[mlir] update types in remaining Linalg TransformOps test
2023-05-16 08:16:59 +00:00
transform-op-pack.mlir
[MLIR][Linalg] Avoid padding attribute in pack when possible
2023-07-11 11:32:51 +02:00
transform-op-pad.mlir
[mlir][bufferization][NFC] Rename copy_tensor op to materialize_in_destination (#65467)
2023-09-12 15:20:41 +02:00
transform-op-replace.mlir
[mlir] make structured transform ops use types
2023-05-16 08:16:56 +00:00
transform-op-rewrite-in-destination-passing-style.mlir
[mlir] update types in remaining Linalg TransformOps test
2023-05-16 08:16:59 +00:00
transform-op-scalarize.mlir
[mlir] make structured transform ops use types
2023-05-16 08:16:56 +00:00
transform-op-split-reduction-by-scaling.mlir
[mlir] make structured transform ops use types
2023-05-16 08:16:56 +00:00
transform-op-split-reduction.mlir
[mlir][arith] Rename operations: maxf → maximumf, minf → minimumf (#65800)
2023-09-11 22:02:19 -07:00
transform-op-split.mlir
[mlir] update types in remaining Linalg TransformOps test
2023-05-16 08:16:59 +00:00
transform-op-tile.mlir
[MLIR][Linalg] Bail out if the tiles provided are more than the number (#66007)
2023-09-13 10:41:03 -04:00
transform-op-vectorize.mlir
[mlir][transform] Improve transform.get_closest_isolated_parent
2023-07-04 16:23:08 +02:00
transform-ops-invalid.mlir
[MLIR][Linalg] Retire tile_to_scf_for (#65633)
2023-09-07 16:13:23 -04:00
transform-ops.mlir
[mlir][linalg][transform] Return copy_back op from PadOp.
2023-08-29 14:55:33 +00:00
transform-pack-greedily.mlir
[MLIR][Linalg] Avoid padding attribute in pack when possible
2023-07-11 11:32:51 +02:00
transform-patterns.mlir
[mlir] make structured transform ops use types
2023-05-16 08:16:56 +00:00
transform-promotion.mlir
[mlir] make structured transform ops use types
2023-05-16 08:16:56 +00:00
transform-tile-and-fuse.mlir
[mlir][arith] Rename operations: maxf → maximumf, minf → minimumf (#65800)
2023-09-11 22:02:19 -07:00
transform-tile-reduction.mlir
[MLIR][Tensor] Add Destination style RewritePattern for DimOp. (#65780)
2023-09-09 06:01:56 -04:00
value-bounds-op-interface-impl.mlir
[mlir][linalg] ValueBoundsOpInterface: Add support for linalg.index
2023-04-19 15:51:37 +09:00
vectorization-masked.mlir
[mlir][linalg] Extend scalable vectorisation support
2023-08-22 09:40:34 +00:00
vectorization-scalable.mlir
[mlir][Vector] Add basic scalable vectorization support to Linalg vectorizer
2023-06-13 23:55:15 +00:00
vectorization.mlir
[mlir][vector] Rename vector reductions: maxf → maximumf, minf → minimumf
2023-09-13 22:49:07 +00:00
vectorize-convolution.mlir
[mlir][arith] Rename operations: maxf → maximumf, minf → minimumf (#65800)
2023-09-11 22:02:19 -07:00
vectorize-tensor-extract-masked.mlir
[mlir][nfc] Make vectorize_nd_extract optional
2023-09-04 18:37:36 +01:00
vectorize-tensor-extract.mlir
[mlir][transform] Extend MaskedVectorizeOp to work for regular vectorization too
2023-09-01 16:19:55 +01:00