**Description**
The documentation of `transform.structured.tile_using_forall` says:
_"It is the user’s responsibility to ensure that num_threads/tile_sizes
is a valid tiling specification (i.e. that only tiles parallel
dimensions, e.g. in the Linalg case)."_
In other words, tiling a non-parallel dimension would generate code with
data races which is not safe to parallelize. For example, consider this
example (included in the tests in this PR):
```
func.func @tile_thread_safety2(%arg0: tensor<100x300x8xf32>, %arg1: tensor<300x8xf32>) -> tensor<300x8xf32> {
%0 = scf.forall (%arg2) in (8) shared_outs(%arg3 = %arg1) -> (tensor<300x8xf32>) {
%1 = affine.min #map(%arg2)
%2 = affine.max #map1(%1)
%3 = affine.apply #map2(%arg2)
%extracted_slice = tensor.extract_slice %arg0[%3, 0, 0] [%2, 300, 8] [1, 1, 1] : tensor<100x300x8xf32> to tensor<?x300x8xf32>
%4 = linalg.generic {indexing_maps = [#map3, #map4], iterator_types = ["reduction", "parallel", "parallel"]} ins(%extracted_slice : tensor<?x300x8xf32>) outs(%arg3 : tensor<300x8xf32>) {
^bb0(%in: f32, %out: f32):
%5 = arith.addf %in, %out : f32
linalg.yield %5 : f32
} -> tensor<300x8xf32>
scf.forall.in_parallel {
tensor.parallel_insert_slice %4 into %arg3[0, 0] [300, 8] [1, 1] : tensor<300x8xf32> into tensor<300x8xf32>
}
}
return %0 : tensor<300x8xf32>
}
```
We can easily see that this is not safe to parallelize because all
threads would be writing to the same position in `%arg3` (in the
`scf.forall.in_parallel`.
This PR detects wether it's safe to `tile_using_forall` and emits a
warning in the case it is not.
**Brief explanation**
It first generates a vector of affine expressions representing the tile
values and stores it in `dimExprs`. These affine expressions are
compared with the affine expressions coming from the results of the
affine map of each output in the linalg op. So going back to the
previous example, the original transform is:
```
#map = affine_map<(d0, d1, d2) -> (d0, d1, d2)>
#map1 = affine_map<(d0, d1, d2) -> (d1, d2)>
func.func @tile_thread_safety2(%arg0: tensor<100x300x8xf32>, %arg1: tensor<300x8xf32>) -> tensor<300x8xf32> {
// expected-warning@+1 {{tiling is not thread safe at axis #0}}
%0 = linalg.generic {indexing_maps = [#map, #map1], iterator_types = ["reduction", "parallel", "parallel"]} ins(%arg0 : tensor<100x300x8xf32>) outs(%arg1 : tensor<300x8xf32>) {
^bb0(%in: f32, %out: f32):
%1 = arith.addf %in, %out : f32
linalg.yield %1 : f32
} -> tensor<300x8xf32>
return %0 : tensor<300x8xf32>
}
module attributes {transform.with_named_sequence} {
transform.named_sequence @__transform_main(%arg0: !transform.any_op {transform.readonly}) {
%0 = transform.structured.match ops{["linalg.generic"]} in %arg0 : (!transform.any_op) -> !transform.any_op
%forall, %tiled_generic = transform.structured.tile_using_forall %0 num_threads [8]
: (!transform.any_op) -> (!transform.any_op, !transform.any_op)
transform.yield
}
}
```
The `num_threads` attribute would be represented as `(d0)`. Because the
linalg op has only one output (`arg1`) it would only check against the
results of `#map1`, which are `(d1, d2)`. The idea is to check that all
affine expressions in `dimExprs` are present in the output affine map.
In this example, `d0` is not in `(d1, d2)`, so tiling that axis is
considered not thread safe.
Using `LoopLikeOpInterface` as the basis for the implementation unifies
all the tiling logic for both `scf.for` and `scf.forall`. The only
difference is the actual loop generation. This is a follow up to
https://github.com/llvm/llvm-project/pull/72178
Instead of many entry points for each loop type, the loop type is now
passed as part of the options passed to the tiling method.
This is a breaking change with the following changes
1) The `scf::tileUsingSCFForOp` is renamed to `scf::tileUsingSCF`
2) The `scf::tileUsingSCFForallOp` is deprecated. The same
functionality is obtained by using `scf::tileUsingSCF` and setting
the loop type in `scf::SCFTilingOptions` passed into this method to
`scf::SCFTilingOptions::LoopType::ForallOp` (using the
`setLoopType` method).
3) The `scf::tileConsumerAndFusedProducerGreedilyUsingSCFForOp` is
renamed to `scf::tileConsumerAndFuseProducerUsingSCF`. The use of
the `controlFn` in `scf::SCFTileAndFuseOptions` allows implementing
any strategy with the default callback implemeting the greedy fusion.
4) The `scf::SCFTilingResult` and `scf::SCFTileAndFuseResult` now use
`SmallVector<LoopLikeOpInterface>`.
5) To make `scf::ForallOp` implement the parts of
`LoopLikeOpInterface` needed, the `getOutputBlockArguments()`
method is replaced with `getRegionIterArgs()`
These changes now bring the tiling and fusion capabilities using
`scf.forall` on par with what was already supported by `scf.for`
This commit renames 4 pattern rewriter API functions:
* `updateRootInPlace` -> `modifyOpInPlace`
* `startRootUpdate` -> `startOpModification`
* `finalizeRootUpdate` -> `finalizeOpModification`
* `cancelRootUpdate` -> `cancelOpModification`
The term "root" is a misnomer. The root is the op that a rewrite pattern
matches against
(https://mlir.llvm.org/docs/PatternRewriter/#root-operation-name-optional).
A rewriter must be notified of all in-place op modifications, not just
in-place modifications of the root
(https://mlir.llvm.org/docs/PatternRewriter/#pattern-rewriter). The old
function names were confusing and have contributed to various broken
rewrite patterns.
Note: The new function names use the term "modify" instead of "update"
for consistency with the `RewriterBase::Listener` terminology
(`notifyOperationModified`).
* "init" operands are specified with `MutableOperandRange` (which gives
access to the underlying `OpOperand *`). No more magic numbers.
* Remove most interface methods and make them helper functions. Only
`getInitsMutable` should be implemented.
* Provide separate helper functions for accessing mutable/immutable
operands (`OpOperand`/`Value`, in line with #66515): `getInitsMutable`
and `getInits` (same naming convention as auto-generated op accessors).
`getInputOperands` was not renamed because this function cannot return a
`MutableOperandRange` (because the operands are not necessarily
consecutive). `OpOperandVector` is no longer needed.
* The new `getDpsInits`/`getDpsInitsMutable` is more efficient than the
old `getDpsInitOperands` because no `SmallVector` is created. The new
functions return a range of operands.
* Fix a bug in `getDpsInputOperands`: out-of-bounds operands were
potentially returned.
`getConstantIntValue` extracts constant values from all constant-like ops, not just `arith::ConstantIndexOp`.
Differential Revision: https://reviews.llvm.org/D154356
The old code used to materialize constants as ops, immediately folded them into the resulting affine map and then deleted the constant ops again. Instead, directly fold the attributes into the affine map. Furthermore, all helpers accept `OpFoldResult` instead of `Value` now. This makes the code at call sites more efficient, because it is no longer necessary to materialize a `Value`, just to be able to use these helper functions.
Note: The API has changed (accepts OpFoldResult instead of Value), otherwise this change is NFC.
Differential Revision: https://reviews.llvm.org/D153324
The MLIR classes Type/Attribute/Operation/Op/Value support
cast/dyn_cast/isa/dyn_cast_or_null functionality through llvm's doCast
functionality in addition to defining methods with the same name.
This change begins the migration of uses of the method to the
corresponding function call as has been decided as more consistent.
Note that there still exist classes that only define methods directly,
such as AffineExpr, and this does not include work currently to support
a functional cast/isa call.
Context:
- https://mlir.llvm.org/deprecation/ at "Use the free function variants
for dyn_cast/cast/isa/…"
- Original discussion at https://discourse.llvm.org/t/preferred-casting-style-going-forward/68443
Implementation:
This patch updates all remaining uses of the deprecated functionality in
mlir/. This was done with clang-tidy as described below and further
modifications to GPUBase.td and OpenMPOpsInterfaces.td.
Steps are described per line, as comments are removed by git:
0. Retrieve the change from the following to build clang-tidy with an
additional check:
main...tpopp:llvm-project:tidy-cast-check
1. Build clang-tidy
2. Run clang-tidy over your entire codebase while disabling all checks
and enabling the one relevant one. Run on all header files also.
3. Delete .inc files that were also modified, so the next build rebuilds
them to a pure state.
```
ninja -C $BUILD_DIR clang-tidy
run-clang-tidy -clang-tidy-binary=$BUILD_DIR/bin/clang-tidy -checks='-*,misc-cast-functions'\
-header-filter=mlir/ mlir/* -fix
rm -rf $BUILD_DIR/tools/mlir/**/*.inc
```
Differential Revision: https://reviews.llvm.org/D151542
The MLIR classes Type/Attribute/Operation/Op/Value support
cast/dyn_cast/isa/dyn_cast_or_null functionality through llvm's doCast
functionality in addition to defining methods with the same name.
This change begins the migration of uses of the method to the
corresponding function call as has been decided as more consistent.
Note that there still exist classes that only define methods directly,
such as AffineExpr, and this does not include work currently to support
a functional cast/isa call.
Caveats include:
- This clang-tidy script probably has more problems.
- This only touches C++ code, so nothing that is being generated.
Context:
- https://mlir.llvm.org/deprecation/ at "Use the free function variants
for dyn_cast/cast/isa/…"
- Original discussion at https://discourse.llvm.org/t/preferred-casting-style-going-forward/68443
Implementation:
This first patch was created with the following steps. The intention is
to only do automated changes at first, so I waste less time if it's
reverted, and so the first mass change is more clear as an example to
other teams that will need to follow similar steps.
Steps are described per line, as comments are removed by git:
0. Retrieve the change from the following to build clang-tidy with an
additional check:
https://github.com/llvm/llvm-project/compare/main...tpopp:llvm-project:tidy-cast-check
1. Build clang-tidy
2. Run clang-tidy over your entire codebase while disabling all checks
and enabling the one relevant one. Run on all header files also.
3. Delete .inc files that were also modified, so the next build rebuilds
them to a pure state.
4. Some changes have been deleted for the following reasons:
- Some files had a variable also named cast
- Some files had not included a header file that defines the cast
functions
- Some files are definitions of the classes that have the casting
methods, so the code still refers to the method instead of the
function without adding a prefix or removing the method declaration
at the same time.
```
ninja -C $BUILD_DIR clang-tidy
run-clang-tidy -clang-tidy-binary=$BUILD_DIR/bin/clang-tidy -checks='-*,misc-cast-functions'\
-header-filter=mlir/ mlir/* -fix
rm -rf $BUILD_DIR/tools/mlir/**/*.inc
git restore mlir/lib/IR mlir/lib/Dialect/DLTI/DLTI.cpp\
mlir/lib/Dialect/Complex/IR/ComplexDialect.cpp\
mlir/lib/**/IR/\
mlir/lib/Dialect/SparseTensor/Transforms/SparseVectorization.cpp\
mlir/lib/Dialect/Vector/Transforms/LowerVectorMultiReduction.cpp\
mlir/test/lib/Dialect/Test/TestTypes.cpp\
mlir/test/lib/Dialect/Transform/TestTransformDialectExtension.cpp\
mlir/test/lib/Dialect/Test/TestAttributes.cpp\
mlir/unittests/TableGen/EnumsGenTest.cpp\
mlir/test/python/lib/PythonTestCAPI.cpp\
mlir/include/mlir/IR/
```
Differential Revision: https://reviews.llvm.org/D150123
Currently the `getTiledImplementation` and `generateResultTileValue`
return just `SmallVector<Operation *>` and `FailureOr<Value>`.
- For `getTiledImplementation` returning empty implies tiling wasnt
done. There is also an implicit assumption that the tiled operation
results correspond to the tiled values of the result of the original
operation. This cannot handle cases where the tiled implementation
might use multiple operations to compute the tiled value for the
results of the untiled operation. Sometimes, the tiled operation
might not directly give the tiled values, and might require casts,
etc to get a replacement.
- For `generateResultTileValue`, it is assumed that the op defining
the returned `Value` is the operation that represents the tiled
computation. Again presence of casts, etc violate this.
Instead make these methods return
```
struct TilingResult {
SmallVector<Operation *> tiledOps;
SmallVector<Value> tiledValues;
};
```
The `tiledOps` represent the operations generated that are relevant
for subsequent transformations. The `tiledValues` represent the tiled
values for the results of the original operation. This better
transmits the state of the transformed IR.
As a consequence the following methods also return `FailureOr<TilingResult>`
- `tensor::replaceExtractSliceWithTiledProducer`
- `tensor::bubbleUpPadSlice`
Differential Revision: https://reviews.llvm.org/D145133
Replace references to enumerate results with either result_pairs
(reference wrapper type) or structured bindings. I did not use
structured bindings everywhere as it wasn't clear to me it would
improve readability.
This is in preparation to the switch to zip semantics which won't
support non-const lvalue reference to elements:
https://reviews.llvm.org/D144503.
I chose to use values instead of const lvalue-refs because MLIR is
biased towards avoiding `const` local variables. This won't degrade
performance because currently `result_pair` is cheap to copy (size_t
+ iterator), and in the future, the enumerator iterator dereference
will return temporaries anyway.
Reviewed By: dblaikie
Differential Revision: https://reviews.llvm.org/D146006
Rewrite and document multi-buffering properly:
1. Use IndexingUtils / StaticValueUtils instead of duplicating functionality
2. Properly plumb RewriterBase through.
3. Add support
4. Better debug messages.
This revision is otherwise almost NFC, if it weren't for the extra DeallocOp
support that would previoulsy make multi-buffering fail.
Depends on: D145036
Differential Revision: https://reviews.llvm.org/D145055
Our tiling implementation assumes a "tensor semantic" for the operation to
be tiled.
Prior to this patch, if we provide a tilable op with "buffer semantic", we
will assert instead of gracefully reject the input.
This patch turns the assert in a proper error.
Differential Revision: https://reviews.llvm.org/D143558
Use the recently introduced transform dialect parameter mechanism to
perform controllable multi-size tiling with sizes computed at the
transformation time rather than at runtime.
This requires to generalize tile and split structured transform
operations to work with any transform dialect handle types, which is
desirable in itself to avoid unchecked overuse of PDL OperationType.
Reviewed By: shabalin
Differential Revision: https://reviews.llvm.org/D140980
The patch adds operations to `BlockAndValueMapping` and renames it to `IRMapping`. When operations are cloned, old operations are mapped to the cloned operations. This allows mapping from an operation to a cloned operation. Example:
```
Operation *opWithRegion = ...
Operation *opInsideRegion = &opWithRegion->front().front();
IRMapping map
Operation *newOpWithRegion = opWithRegion->clone(map);
Operation *newOpInsideRegion = map.lookupOrNull(opInsideRegion);
```
Migration instructions:
All includes to `mlir/IR/BlockAndValueMapping.h` should be replaced with `mlir/IR/IRMapping.h`. All uses of `BlockAndValueMapping` need to be renamed to `IRMapping`.
Reviewed By: rriddle, mehdi_amini
Differential Revision: https://reviews.llvm.org/D139665
A transformation tiling a reduction dimension of a Linalg op needs a
tile size for said dimension. When an insufficient number of dimensions
was provided, it would segfault due to out-of-bounds access to a vector.
Also fix incorrect error reporting in the structured transform op
exercising this functionality.
Reviewed By: springerm, ThomasRaoux
Differential Revision: https://reviews.llvm.org/D141046
std::optional::value() has undesired exception checking semantics and is
unavailable in older Xcode (see _LIBCPP_AVAILABILITY_BAD_OPTIONAL_ACCESS). The
call sites block std::optional migration.
This is part of an effort to migrate from llvm::Optional to
std::optional. This patch changes the way mlir-tblgen generates .inc
files, and modifies tests and documentation appropriately. It is a "no
compromises" patch, and doesn't leave the user with an unpleasant mix of
llvm::Optional and std::optional.
A non-trivial change has been made to ControlFlowInterfaces to split one
constructor into two, relating to a build failure on Windows.
See also: https://discourse.llvm.org/t/deprecating-llvm-optional-x-hasvalue-getvalue-getvalueor/63716
Signed-off-by: Ramkumar Ramachandra <r@artagnon.com>
Differential Revision: https://reviews.llvm.org/D138934
In the process, numerous insertion point issues were found and fixed.
RAII on insertion points is now used more dilligently.
Differential Revision: https://reviews.llvm.org/D139714
This adds a tile_size parameter, when it is used the tiles are
cyclically distributed onto the threads of the scf.foreach_thread op.
Differential Revision: https://reviews.llvm.org/D139474
This patch mechanically replaces None with std::nullopt where the
compiler would warn if None were deprecated. The intent is to reduce
the amount of manual work required in migrating from Optional to
std::optional.
This is part of an effort to migrate from llvm::Optional to
std::optional:
https://discourse.llvm.org/t/deprecating-llvm-optional-x-hasvalue-getvalue-getvalueor/63716
This fixes the case where scf::LoopNest::loops is empty.
Change LoopVector and ValueVector to SmallVector.
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D136926
This adds a transformation to tile reduction operations to partial
reduction using scf.foreachthread. This uses
PartialReductionOpInterface to create a merge operation of the partial
tiles.
Differential Revision: https://reviews.llvm.org/D137912
`scf.foreach_thread` defines mapping its loops to processors via an integer array, see an example below. A lowering can use this mapping. However, expressing mapping as an integer array is very confusing, especially when there are multiple levels of parallelism. In addition, the op does not verify the integer array. This change introduces device mapping attribute to make mapping descriptive and verifiable. Then it makes GPU transform dialect use it.
```
scf.foreach_thread (%i, %j) in (%c1, %c2) {
scf.foreach_thread (%i2, %j2) in (%c1, %c2)
{...} { thread_dim_mapping = [0, 1]}
} { thread_dim_mapping = [0, 1]}
```
It first introduces a `DeviceMappingInterface` which is an attribute interface. `scf.foreach_thread` defines its mapping via this interface. A lowering must define its attributes and implement this interface as well. This way gives us a clear validation.
The change also introduces two new attributes (`#gpu.thread<x/y/z>` and `#gpu.block<x,y,z>` ). After this change, the above code prints as below, as seen here, this way clarifies the loop mappings. The change also implements consuming of these two new attribute by the transform dialect. Transform dialect binds the outermost loops to the thread blocks and innermost loops to threads.
```
scf.foreach_thread (%i, %j) in (%c1, %c2) {
scf.foreach_thread (%i2, %j2) in (%c1, %c2)
{...} { thread_dim_mapping = [#gpu.thread<x>, #gpu.thread<y>]}
} { thread_dim_mapping = [#gpu.block<x>, #gpu.block<y>]}
```
Reviewed By: ftynse, nicolasvasilache
Differential Revision: https://reviews.llvm.org/D137413
[RFC: EnumAttr for iterator types in Linalg](https://discourse.llvm.org/t/rfc-enumattr-for-iterator-types-in-linalg/64535)
This affect touches and probably breaks most of the code that creates `linalg.generic`. A fix would be to replace calls to `getParallelIteratorTypeName/getReductionIteratorTypeName` with `mlir::utils::IteratorType::parallel/reduction` and types from `StringRef` to `mlir::utils::IteratorType`.
Due to limitations of tablegen, shared C++ definition of IteratorType enum lives in StructuredOpsUtils.td, but each dialect should have it's own EnumAttr wrapper. To avoid conflict, all enums in a dialect are put into a separate file with a separate tablegen rule.
Test dialect td files are refactored a bit.
Printed format of `linalg.generic` temporarily remains unchanged to avoid breaking code and tests in the same change.
Differential Revision: https://reviews.llvm.org/D137658
`getDestinationOperands` was almost a duplicate of `DestinationStyleOpInterface::getOutputOperands`. Now that the interface has been moved to mlir/Interfaces, it is no longer needed.
Differential Revision: https://reviews.llvm.org/D136240
Context: https://discourse.llvm.org/t/psa-retire-linalg-filter-based-patterns/63785
Uses of `LinalgTilingPattern::returningMatchAndRewrite` are replaced by a top-level `tileWithLinalgTilingOptions` function that is marked obsolete and serves
as a temporary means to transition away from `LinalgTilingOptions`-based tiling.
LinalgTilingOptions supports too many options that have been orthogonalized with the use of the transform dialect.
Additionally, the revision introduces a `transform.structured.tile_to_scf_for` structured transform operation that is needed to properly tile `tensor.pad`
via the TilingInterface. Uses of `transform.structured.tile` will be deprecated and replaced by this new op.
This will achieve the deprecation of `linalg::tileLinalgOp`.
Context: https://discourse.llvm.org/t/psa-retire-tileandfuselinalgops-method/63850
In the process of transitioning, tests that were performing tile and distribute on tensors are retired: transformations should be orthogonalized better in the future.
In particular, tiling to specific loop types and tileAndDistribute behavior are not available via the transform ops.
The behavior is still available as part of the `tileWithLinalgTilingOptions` method to allow downstream clients to transition without breakages but is meant to be retired soon.
As more tests are ported to the transform dialect, it became necessary to introduce a test-transform-dialect-erase-schedule-pass to discard the transform specification
once applied so that e2e lowering and execution is possible.
Lastly, a number of redundant tests that were testing composition of patterns are retired as they are available with a better mechanism via the transform dialect.
Differential Revision: https://reviews.llvm.org/D135573
tensor.empty/linalg.init_tensor produces an uninititalized tensor that can be used as a destination operand for destination-style ops (ops that implement `DestinationStyleOpInterface`).
This change makes it possible to implement `TilingInterface` for non-destination-style ops without depending on the Linalg dialect.
RFC: https://discourse.llvm.org/t/rfc-add-tensor-from-shape-operation/65101
Differential Revision: https://reviews.llvm.org/D135129
The tiling interface generates the order of parallel_insert_slice incorrectly when there are multiple destionation operands. This revision fixes that and adds a test for it. It depends on D132937
Reviewed By: nicolasvasilache
Differential Revision: https://reviews.llvm.org/D133204
This revision relaxes constraint of tiling when there are multiple destination operands. It also adds a test.
Reviewed By: nicolasvasilache
Differential Revision: https://reviews.llvm.org/D132937
