Changes in bufferize.mlir

b3ebe3be - [mlir][bufferize] Bufferize after TensorCopyInsertion

Fri, 17 Jun 2022 11:23:25 +0000

[mlir][bufferize] Bufferize after TensorCopyInsertionThis change changes the bufferization so that it utilizes the new TensorCopyInsertion pass. One-Shot Bufferize no longer calls the One-Shot Analysis. Instead, it relies on the TensorCopyInsertion pass to make the entire IR fully inplacable. The `bufferize` implementations of all ops are simplified; they no longer have to account for out-of-place bufferization decisions. These were already materialized in the IR in the form of `bufferization.alloc_tensor` ops during the TensorCopyInsertion pass.Differential Revision: https://reviews.llvm.org/D127652 List of files: /llvm-project-15.0.7/mlir/test/Dialect/Linalg/bufferize.mlir

2fe40c34 - [mlir][bufferize] Fix op filter

Thu, 12 May 2022 07:27:21 +0000

[mlir][bufferize] Fix op filterBufferization has an optional filter to exclude certain ops from analysis+bufferization. There were a few remaining places in the codebase where the filter was not checked.Differential Revision: https://reviews.llvm.org/D125356 List of files: /llvm-project-15.0.7/mlir/test/Dialect/Linalg/bufferize.mlir

92d38adb - [mlir][NFC] Update textual references of `func` to `func.func` in Linalg tests

Wed, 20 Apr 2022 23:16:23 +0000

[mlir][NFC] Update textual references of `func` to `func.func` in Linalg testsThe special case parsing of `func` operations is being removed. List of files: /llvm-project-15.0.7/mlir/test/Dialect/Linalg/bufferize.mlir

39ec46bd - [mlir][bufferize] Extract buffer hoisting into separate function

Tue, 15 Mar 2022 12:22:21 +0000

[mlir][bufferize] Extract buffer hoisting into separate functionThis improves the modularity of the bufferization.From now on, all ops that do not implement BufferizableOpInterface are considered hoisting barriers. Previously, all ops that do not implement the interface were not considered barriers and such ops had to be marked as barriers explicitly. This was unsafe because we could've hoisted across unknown ops where it was not safe to hoist.As a side effect, this allows for cleaning up AffineBufferizableOpInterfaceImpl. This build unit no longer needed and can be deleted.Differential Revision: https://reviews.llvm.org/D121519 List of files: /llvm-project-15.0.7/mlir/test/Dialect/Linalg/bufferize.mlir

7294be2b - [mlir][linalg] Replace linalg.fill by OpDSL variant.

Mon, 14 Mar 2022 10:45:04 +0000

[mlir][linalg] Replace linalg.fill by OpDSL variant.The revision removes the linalg.fill operation and renames the OpDSL generated linalg.fill_tensor operation to replace it. After the change, all named structured operations are defined via OpDSL and there are no handwritten operations left.A side-effect of the change is that the pretty printed form changes from:```%1 = linalg.fill(%cst, %0) : f32, tensor -> tensor```changes to```%1 = linalg.fill ins(%cst : f32) outs(%0 : tensor) -> tensor```Additionally, the builder signature now takes input and output value ranges as it is the case for all other OpDSL operations:```rewriter.create(loc, val, output)```changes to```rewriter.create(loc, ValueRange{val}, ValueRange{output})```All other changes remain minimal. In particular, the canonicalization patterns are the same and the `value()`, `output()`, and `result()` methods are now implemented by the FillOpInterface.Depends On D120726Reviewed By: nicolasvasilacheDifferential Revision: https://reviews.llvm.org/D120728 List of files: /llvm-project-15.0.7/mlir/test/Dialect/Linalg/bufferize.mlir

16cbe883 - [mlir][linalg][bufferize] Migrate --linalg-bufferize to BufferizableOpInterface-based bufferization

Thu, 03 Mar 2022 10:50:32 +0000

[mlir][linalg][bufferize] Migrate --linalg-bufferize to BufferizableOpInterface-based bufferizationThis commit deletes the old dialect conversion-based bufferization patterns, which are now obsolete.Differential Revision: https://reviews.llvm.org/D120883 List of files: /llvm-project-15.0.7/mlir/test/Dialect/Linalg/bufferize.mlir

4086b3be - [mlir][bufferize][NFC] Remove obsolete tensor bufferization patterns from Linalg/Bufferize.cpp

Fri, 18 Feb 2022 09:46:20 +0000

[mlir][bufferize][NFC] Remove obsolete tensor bufferization patterns from Linalg/Bufferize.cppDifferential Revision: https://reviews.llvm.org/D119824 List of files: /llvm-project-15.0.7/mlir/test/Dialect/Linalg/bufferize.mlir

73e880fb - [mlir][bufferize] Add vector-bufferize pass and remove obsolete patterns from Linalg Bufferize

Tue, 15 Feb 2022 12:16:50 +0000

[mlir][bufferize] Add vector-bufferize pass and remove obsolete patterns from Linalg BufferizeDifferential Revision: https://reviews.llvm.org/D119444 List of files: /llvm-project-15.0.7/mlir/test/Dialect/Linalg/bufferize.mlir

ebc81537 - Revert "Revert "[mlir] Purge `linalg.copy` and use `memref.copy` instead.""

Tue, 01 Feb 2022 17:07:33 +0000

Revert "Revert "[mlir] Purge `linalg.copy` and use `memref.copy` instead.""This reverts commit 25bf6a2a9bc6ecb3792199490c70c4ce50a94aea. List of files: /llvm-project-15.0.7/mlir/test/Dialect/Linalg/bufferize.mlir

25bf6a2a - Revert "[mlir] Purge `linalg.copy` and use `memref.copy` instead."

Mon, 31 Jan 2022 17:51:39 +0000

Revert "[mlir] Purge `linalg.copy` and use `memref.copy` instead."This reverts commit 016956b68081705ffee511c334e31e414fa1ddbf.Reverting it to fix NVidia build without being in a hurry. List of files: /llvm-project-15.0.7/mlir/test/Dialect/Linalg/bufferize.mlir

016956b6 - [mlir] Purge `linalg.copy` and use `memref.copy` instead.

Mon, 31 Jan 2022 17:18:28 +0000

[mlir] Purge `linalg.copy` and use `memref.copy` instead.Differential Revision: https://reviews.llvm.org/D118028 List of files: /llvm-project-15.0.7/mlir/test/Dialect/Linalg/bufferize.mlir

fd0c6f53 - [mlir] Move linalg::PadTensorOp to tensor::PadOp.

Fri, 21 Jan 2022 18:29:08 +0000

[mlir] Move linalg::PadTensorOp to tensor::PadOp.RFC: https://llvm.discourse.group/t/rfc-move-linalg-padtensorop-to-tensor-padop/5785Differential Revision: https://reviews.llvm.org/D117892 List of files: /llvm-project-15.0.7/mlir/test/Dialect/Linalg/bufferize.mlir

d75c3e83 - [mlir] Don't print `// no predecessors` on entry blocks

Wed, 19 Jan 2022 20:18:30 +0000

[mlir] Don't print `// no predecessors` on entry blocksEntry blocks can never have predecessors, so this is unnecessary.Fixes #53287Differential Revision: https://reviews.llvm.org/D117713 List of files: /llvm-project-15.0.7/mlir/test/Dialect/Linalg/bufferize.mlir

b618880e - [mlir] Move `linalg.tensor_expand/collapse_shape` to TensorDialect.

Fri, 10 Dec 2021 11:03:47 +0000

[mlir] Move `linalg.tensor_expand/collapse_shape` to TensorDialect.RFC: https://llvm.discourse.group/t/rfc-reshape-ops-restructuring/3310linalg.fill gets a canonicalizer, because `FoldFillWithTensorReshape` cannot be moved to tensorops (it uses linalg::FillOp inside). Before it was listed as a canonicalization pattern for the reshape operations, now it became a canonicalization for FillOp.Differential Revision: https://reviews.llvm.org/D115502 List of files: /llvm-project-15.0.7/mlir/test/Dialect/Linalg/bufferize.mlir

57470abc - [mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect.

Thu, 25 Nov 2021 10:42:16 +0000

[mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect.https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712Differential Revision: https://reviews.llvm.org/D114552 List of files: /llvm-project-15.0.7/mlir/test/Dialect/Linalg/bufferize.mlir

015192c6 - [mlir:DialectConversion] Restructure how argument/target materializations get invoked

Wed, 27 Oct 2021 02:00:10 +0000

[mlir:DialectConversion] Restructure how argument/target materializations get invokedThe current implementation invokes materializationswhenever an input operand does not have a mapping for thedesired type, i.e. it requires materialization at the earliest possiblepoint. This conflicts with goal of dialect conversion (and also thecurrent documentation) which states that a materialization is onlyrequired if the materialization is supposed to persist after theconversion process has finished.This revision refactors this such that whenever a targetmaterialization "might" be necessary, we insert anunrealized_conversion_cast to act as a temporary materialization.This allows for deferring the invocation of the usermaterialization hooks until the end of the conversion process,where we actually have a better sense if it's actuallynecessary. This has several benefits:* In some cases a target materialization hook is no longer necessaryWhen performing a full conversion, there are some situationswhere a temporary materialization is necessary. Moving forward,these users won't need to provide any target materializations,as the temporary materializations do not require the user toprovide materialization hooks.* getRemappedValue can now handle values that haven't been converted yetBefore this commit, it wasn't well supported to get the remappedvalue of a value that hadn't been converted yet (making itdifficult/impossible to convert multiple operations in manysituations). This commit updates getRemappedValue to properlyhandle this case by inserting temporary materializations whennecessary.Another code-health related benefit is that with this change wecan move a majority of the complexity related to materializationsto the end of the conversion process, instead of handling adhocwhile conversion is happening.Differential Revision: https://reviews.llvm.org/D111620 List of files: /llvm-project-15.0.7/mlir/test/Dialect/Linalg/bufferize.mlir

a54f4eae - [MLIR] Replace std ops with arith dialect ops

Tue, 12 Oct 2021 23:14:57 +0000

[MLIR] Replace std ops with arith dialect opsPrecursor: https://reviews.llvm.org/D110200Removed redundant ops from the standard dialect that were moved to the`arith` or `math` dialects.Renamed all instances of operations in the codebase and in tests.Reviewed By: rriddle, jpienaarDifferential Revision: https://reviews.llvm.org/D110797 List of files: /llvm-project-15.0.7/mlir/test/Dialect/Linalg/bufferize.mlir

5e6c170b - [mlir][linalg] Fix bufferize pattern to allow unknown operations in body of generic

Fri, 10 Sep 2021 10:58:18 +0000

[mlir][linalg] Fix bufferize pattern to allow unknown operations in body of genericThe original version of the bufferization pattern for linalg.generic wouldmanually clone operations within the region to the bufferized clone of theoperation. This triggers legality requirements on those operations in theconversion infra. Instead, this now uses the rewriter to inline the regioninstead, avoiding those legality requirements.Differential Revision: https://reviews.llvm.org/D109581 List of files: /llvm-project-15.0.7/mlir/test/Dialect/Linalg/bufferize.mlir

42431b82 - [tests] Make testsuite more resilient to "order of constant" changes. NFC.

Wed, 08 Sep 2021 17:09:42 +0000

[tests] Make testsuite more resilient to "order of constant" changes. NFC. List of files: /llvm-project-15.0.7/mlir/test/Dialect/Linalg/bufferize.mlir

1e733a8c - Revert "Bufferization for tiled loop."

Wed, 11 Aug 2021 08:04:12 +0000

Revert "Bufferization for tiled loop."This reverts commit edaffebcb2a62b0195e23fe7d4ead005822865c3. List of files: /llvm-project-15.0.7/mlir/test/Dialect/Linalg/bufferize.mlir