| Age | Commit message (Collapse) | Author |
|---|
|
given tiled loop nest. (#167634)
The existing `scf::tileAndFuseConsumerOfSlices` takes a list of slices
(and loops they are part of), tries to find the consumer of these slices
(all slices are expected to be the same consumer), and then tiles the
consumer into the loop nest using the `TilingInterface`. A more natural
way of doing consumer fusion is to just start from the consumer, look
for operands that are produced by the loop nest passed in as `loops`
(presumably these loops are generated by tiling, but that is not a
requirement for consumer fusion). Using the consumer you can find the
slices of the operands that are accessed within the loop which you can
then use to tile and fuse the consumer (using `TilingInterface`). This
handles more naturally the case where multiple operands of the consumer
come from the loop nest.
The `scf::tileAndFuseConsumerOfSlices` was implemented as a mirror of
`scf::tileAndFuseProducerOfSlice`. For the latter, the slice has a
single producer for the source of the slice, which makes it a natural
way of specifying producer fusion. But for consumers, the result might
have multiple users, resulting in multiple candidates for fusion, as
well as a fusion candidate using multiple results from the tiled loop
nest. This means using slices
(`tensor.insert_slice`/`tensor.parallel_insert_slice`) as a hook for
consumer fusion turns out to be quite hard to navigate. The use of the
consumer directly avoids all those pain points. In time the
`scf::tileAndFuseConsumerOfSlices` should be deprecated in favor of
`scf::tileAndFuseConsumer`. There is a lot of tech-debt that has
accumulated in `scf::tileAndFuseConsumerOfSlices` that needs to be
cleanedup. So while that gets cleaned up, and required functionality is
moved to `scf::tileAndFuseConsumer`, the old path is still maintained.
The test for `scf::tileAndFuseConsumerUsingSlices` is copied to
`tile-and-fuse-consumer.mlir` to
`tile-and-fuse-consumer-using-slices.mlir`. All the tests that were
there in this file are now using the `tileAndFuseConsumer` method. The
test op `test.tile_and_fuse_consumer` is modified to call
`scf::tileAndFuseConsumer`, while a new op
`test.tile_and_fuse_consumer_of_slice` is used to keep the old path
tested while it is deprecated.
---------
Signed-off-by: MaheshRavishankar <mahesh.ravishankar@gmail.com>
|
|
CustomOp. (#161386)
This PR adds to the generateLoopTerminatorFn callback the loops
generated by GenerateLoopHeaderFn. This is needed to correctly set the
insertion point with scf.forall ops.
|
|
`mlir::tileUsingSCF`. (#159660)
This change adds an option to use a custom operation to generate the
inter-tile loops during tiling. When the loop type is set to
scf::SCFTilingOptions::LoopType::CustomOp, the method
mlir::tileUsingSCF provides two callback functions
First one to generate the header of the loop.
Second one to generate the terminator of the loop.
These methods receive the information needed to generate the
loops/terminator and expect to return information needed to generate
the code for the intra-tile computation. See comments for more
details.
Presently this is adds support only for tiling. Subsequent commits
will update this to add support for fusion as well.
The PR is split into two commits.
The first commit is an NFC that just refactors the code (and cleans up
some naming) to make it easier to add the support for custom loop
operations.
The second commit adds the support for using a custom loop operation, as
well as a test to exercise this path.
Note that this is duplicate of
https://github.com/llvm/llvm-project/pull/159506 that was accidently
committed and was reverted in
https://github.com/llvm/llvm-project/pull/159598 to wait for reviews.
Signed-off-by: MaheshRavishankar
[mahesh.ravishankar@gmail.com](mailto:mahesh.ravishankar@gmail.com)
---------
Signed-off-by: MaheshRavishankar <mahesh.ravishankar@gmail.com>
|
|
`mlir::tileUsingSCF`." (#159598)
Reverts llvm/llvm-project#159506
It was committed by accident. Reverting it for reviews.
|
|
`mlir::tileUsingSCF`. (#159506)
This change adds an option to use a custom operation to generate the
inter-tile loops during tiling. When the loop type is set to
`scf::SCFTilingOptions::LoopType::CustomOp`, the method
`mlir::tileUsingSCF` provides two callback functions
1. First one to generate the header of the loop.
2. Second one to generate the terminator of the loop.
These methods receive the information needed to generate the
loops/terminator and expect to return information needed to generate
the code for the intra-tile computation. See comments for more
details.
Presently this is adds support only for tiling. Subsequent commits
will update this to add support for fusion as well.
The PR is split into two commits.
1) The first commit is an NFC that just refactors the code (and cleans
up some naming) to make it easier to add the support for custom loop
operations.
2) The second commit adds the support for using a custom loop operation,
as well as a test to exercise this path.
Signed-off-by: MaheshRavishankar <mahesh.ravishankar@gmail.com>
---------
Signed-off-by: MaheshRavishankar <mahesh.ravishankar@gmail.com>
|
|
`ReductionTilingStrategy::PartialReductionOuterParallelStrategy`. (#147593)
Since `scf::tileUsingSCF` is the core method used for tiling the root
operation within the `scf::tileConsumersAndFuseProducersUsingSCF`, the
latter can fuse into any tiled loop generated using `scf::tileUsingSCF`.
This patch adds a test for tiling a root operation using
`ReductionTilingStrategy::PartialReductionOuterParallelStrategy` and
fusing producers with it.
Since this strategy generates a rank-reducing extract slice
`tensor::replaceExtractSliceWithTiledProducer` which is the core method
used for the fusion was extended to handle the rank-reducing slices.
Also fix a small bug in the computation of the reduction induction
variable (which needs to use `floorDiv` instead of `ceilDiv`)
Signed-off-by: MaheshRavishankar <mahesh.ravishankar@gmail.com>
|
|
For consumer fusion cases of this form
```
%0:2 = scf.forall .. shared_outs(%arg0 = ..., %arg0 = ...) {
tensor.parallel_insert_slice ... into %arg0
tensor.parallel_insert_slice ... into %arg1
}
%1 = linalg.generic ... ins(%0#0, %0#1)
```
the current consumer fusion that handles one slice at a time cannot fuse
the consumer into the loop, since fusing along one slice will create and
SSA violation on the other use from the `scf.forall`. The solution is to
allow consumer fusion to allow considering multiple slices at once. This
PR changes the `TilingInterface` methods related to consumer fusion,
i.e.
- `getTiledImplementationFromOperandTile`
- `getIterationDomainFromOperandTile`
to allow fusion while considering multiple operands. It is upto the
`TilingInterface` implementation to return an error if a list of tiles
of the operands cannot result in a consistent implementation of the
tiled operation.
The Linalg operation implementation of `TilingInterface` has been
modified to account for these changes and allow cases where operand
tiles that can result in a consistent tiling implementation are handled.
---------
Signed-off-by: MaheshRavishankar <mahesh.ravishankar@gmail.com>
|
|
In #120115 the replacements for the tiled operations were wrapped within
the `MergeResult` object. That is a bit of an obfuscation and not
immediately obvious where to get the replacements post tiling. This
changes the `SCFTilingResult` to have `replacements` explicit (as it was
before that change).
`mergeOps` is added as a separate field of `SCFTilingResult`, which is
empty when the reduction type is `FullReduction`.
This is a API breaking change. All uses of `mergeResult.replacements`
should be replaced with `replacements`.
There was also an implicit assumption that
`PartialReductionTilingInterface` is derived from `TilingInterface`, so
all ops that implemented the `PartialReductionTilingInterface` were
expected to implement the `TilingInterface` as well. This pre-dated the
existence of derived inheritances. Make
`PartialReductionTilingInterface` derive from `TilingInterface`.
Signed-off-by: MaheshRavishankar <mahesh.ravishankar@gmail.com>
|
|
loops as an argument. (#132082)
This gets the consumer fusion method in sync with the corresponding
producer fusion method `tileAndFuseProducerOfSlice`. Not taking this as
input required use of complicated analysis to retrieve the surrounding
loops which are very fragile. Just like the producer fusion method, the
loops need to be taken in as an argument, with typically the loops being
created by the tiling methods.
Some utilities are added to check that the loops passed in are perfectly
nested (in the case of an `scf.for` loop nest.
This is change 1 of N to simplify the implementation of tile and fuse
consumers.
---------
Signed-off-by: MaheshRavishankar <mahesh.ravishankar@gmail.com>
|
|
(#120115)
This patch unifies the tiling implementation for tileUsingFor and
tileReductionUsingFor. This is done by passing an addition option to
SCFTilingOptions, allowing it to set how reduction dimensions should be
tiled. Currently, there are 3 different options for reduction tiling:
FullReduction (old tileUsingFor), PartialReductionOuterReduction (old
tileReductionUsingFor) and PartialReductionOuterParallel
(linalg::tileReductionUsingForall, this isn't implemented in this
patch).
The patch makes tileReductionUsingFor use the tileUsingFor
implementation with the new reduction tiling options.
There are no test changes because the implementation was doing almost
the exactly same thing. This was also tested in IREE (which uses both
these APIs heavily) and there were no test changes.
|
|
Before, consumer fusion expects single usage(or others are terminator
op). This patch supports multiple tilable consumers fusion.
E.g.
```
%0 = scf.for {
...
%p = tiledProducer
...
}
%1 = tilableConsumer1 ins(%0 : ...)
%2 = tilableConsumer2 ins(%0 : ...)
```
===>
```
%0:3 = scf.for {
...
%p = tiledProducer
%1 = tiledConsumer1 ins(%p : ...)
%2 = tiledConsumer2 ins(%p : ...)
...
}
```
The key process is ensuring that the first user of loop
should not dominate any define of consumer operand(s).
|
|
continue tile + fuse. (#107882)
Current implementation of `scf::tileConsumerAndFuseProducerUsingSCF`
looks at operands of tiled/tiled+fused operations to see if they are
produced by `extract_slice` operations to populate the worklist used to
continue fusion. This implicit assumption does not always work. Instead
make the implementations of `getTiledImplementation` return the slices
to use to continue fusion.
This is a breaking change
- To continue to get the same behavior of
`scf::tileConsumerAndFuseProducerUsingSCF`, change all out-of-tree
implementation of `TilingInterface::getTiledImplementation` to return
the slices to continue fusion on. All in-tree implementations have been
adapted to this.
- This change touches parts that required a simplification to the
`ControlFn` in `scf::SCFTileAndFuseOptions`. It now returns a
`std::optional<scf::SCFTileAndFuseOptions::ControlFnResult>` object that
should be `std::nullopt` if fusion is not to be performed.
Signed-off-by: MaheshRavishankar <mahesh.revishankar@gmail.com>
|
|
`PassManager::run` loads the dependent dialects for each pass into the
current context prior to invoking the individual passes. If the
dependent dialect is already loaded into the context, this should be a
no-op. However, if there are extensions registered in the
`DialectRegistry`, the dependent dialects are unconditionally registered
into the context.
This poses a problem for dynamic pass pipelines, however, because they
will likely be executing while the context is in an immutable state
(because of the parent pass pipeline being run).
To solve this, we'll update the extension registration API on
`DialectRegistry` to require a type ID for each extension that is
registered. Then, instead of unconditionally registered dialects into a
context if extensions are present, we'll check against the extension
type IDs already present in the context's internal `DialectRegistry`.
The context will only be marked as dirty if there are net-new extension
types present in the `DialectRegistry` populated by
`PassManager::getDependentDialects`.
Note: this PR removes the `addExtension` overload that utilizes
`std::function` as the parameter. This is because `std::function` is
copyable and potentially allocates memory for the contained function so
we can't use the function pointer as the unique type ID for the
extension.
Downstream changes required:
- Existing `DialectExtension` subclasses will need a type ID to be
registered for each subclass. More details on how to register a type ID
can be found here:
https://github.com/llvm/llvm-project/blob/8b68e06731e0033ed3f8d6fe6292ae671611cfa1/mlir/include/mlir/Support/TypeID.h#L30
- Existing uses of the `std::function` overload of `addExtension` will
need to be refactored into dedicated `DialectExtension` classes with
associated type IDs. The attached `std::function` can either be inlined
into or called directly from `DialectExtension::apply`.
---------
Co-authored-by: Mehdi Amini <joker.eph@gmail.com>
|
|
`linalg::tileToForallOpUsingTileSizes` (#91878)
The implementation of these methods are legacy and they are removed in
favor of using the `scf::tileUsingSCF` methods as replacements. To get
the latter on par with requirements of the deprecated methods, the
tiling allows one to specify the maximum number of tiles to use instead
of specifying the tile sizes. When tiling to `scf.forall` this
specification is used to generate the `num_threads` version of the
operation.
A slight deviation from previous implementation is that the deprecated
method always generated the `num_threads` variant of the `scf.forall`
operation. Instead now this is driven by the tiling options specified.
This reduces the indexing math generated when the tile sizes are
specified.
**Moving from `linalg::tileToForallOp` to `scf::tileUsingSCF`**
```
OpBuilder b;
TilingInterface op;
ArrayRef<OpFoldResult> numThreads;
ArrayAttr mapping;
FailureOr<ForallTilingResult> result =linalg::tileToForallOp(b, op, numThreads, mapping);
```
can be replaced by
```
scf::SCFTilingOptions options;
options.setNumThreads(numThreads);
options.setLoopType(scf::SCFTilingOptions::LoopType::ForallOp);
options.setMapping(mapping.getValue()); /*note the difference that setMapping takes an ArrayRef<Attribute> */
FailureOr<scf::SCFTilingResult> result = scf::tileUsingSCF(b, op, options);
```
This generates the `numThreads` version of the `scf.forall` for the
inter-tile loops, i.e.
```
... = scf.forall (%arg0, %arg1) in (%nt0, %nt1) shared_outs(...)
```
**Moving from `linalg::tileToForallOpUsingTileSizes` to
`scf::tileUsingSCF`**
```
OpBuilder b;
TilingInterface op;
ArrayRef<OpFoldResult> tileSizes;
ArrayAttr mapping;
FailureOr<ForallTilingResult> result =linalg::tileToForallOpUsingTileSizes(b, op, tileSizes, mapping);
```
can be replaced by
```
scf::SCFTilingOptions options;
options.setTileSizes(tileSizes);
options.setLoopType(scf::SCFTilingOptions::LoopType::ForallOp);
options.setMapping(mapping.getValue()); /*note the difference that setMapping takes an ArrayRef<Attribute> */
FailureOr<scf::SCFTilingResult> result = scf::tileUsingSCF(b, op, options);
```
Also note that `linalg::tileToForallOpUsingTileSizes` would effectively
call the `linalg::tileToForallOp` by computing the `numThreads` from the
`op` and `tileSizes` and generate the `numThreads` version of the
`scf.forall`. That is not the case anymore. Instead this will directly
generate the `tileSizes` version of the `scf.forall` op
```
... = scf.forall(%arg0, %arg1) = (%lb0, %lb1) to (%ub0, %ub1) step(%step0, %step1) shared_outs(...)
```
If you actually want to use the `numThreads` version, it is upto the
caller to compute the `numThreads` and set `options.setNumThreads`
instead of `options.setTileSizes`. Note that there is a slight
difference in the num threads version and tile size version. The former
requires an additional `affine.max` on the tile size to ensure
non-negative tile sizes. When lowering to `numThreads` version this
`affine.max` is not needed since by construction the tile sizes are
non-negative. In previous implementations, the `numThreads` version
generated when using the `linalg::tileToForallOpUsingTileSizes` method
would avoid generating the `affine.max` operation. To get the same
state, downstream users will have to additionally normalize the
`scf.forall` operation.
**Changes to `transform.structured.tile_using_forall`**
The transform dialect op that called into `linalg::tileToForallOp` and
`linalg::tileToForallOpUsingTileSizes` have been modified to call
`scf::tileUsingSCF`. The transform dialect op always generates the
`numThreads` version of the `scf.forall` op. So when `tile_sizes` are
specified for the transform dialect op, first the `tile_sizes` version
of the `scf.forall` is generated by the `scf::tileUsingSCF` method which
is then further normalized to get back to the same state. So there is no
functional change to `transform.structured.tile_using_forall`. It always
generates the `numThreads` version of the `scf.forall` op (as it did
before this change).
---------
Signed-off-by: MaheshRavishankar <mahesh.ravishankar@gmail.com>
|
|
This patch adds more precise side effects to the current ops with memory
effects, allowing us to determine which OpOperand/OpResult/BlockArgument
the
operation reads or writes, rather than just recording the reading and
writing
of values. This allows for convenient use of precise side effects to
achieve
analysis and optimization.
Related discussions:
https://discourse.llvm.org/t/rfc-add-operandindex-to-sideeffect-instance/79243
|
|
This commit adds an API (`tileAndFuseConsumerOfSlice`) to fuse consumer to a producer within
scf.for/scf.forall loop.
To support this two new methods are added to the `TilingInterface`
- `getIterationDomainTileFromOperandTile`
- `getTiledImplementationFromOperandTile`.
Consumer operations that implement this method can be used to be fused with tiled producer operands in a manner similar to (but essentially the inverse of) the fusion of an untiled producer with a tiled consumer.
Note that this only does one `tiled producer` -> `consumer` fusion. This could be called repeatedly for fusing multiple consumers. The current implementation also is conservative in when this kicks in (like single use of the value returned by the inter-tile loops that surround the tiled producer, etc.) These can be relaxed over time.
Signed-off-by: Abhishek Varma <abhvarma@amd.com>
---------
Signed-off-by: Abhishek Varma <abhvarma@amd.com>
Signed-off-by: Abhishek Varma <avarma094@gmail.com>
Co-authored-by: cxy <chenxunyu1993@gmail.com>
|
|
Transform interfaces are implemented, direction or via extensions, in
libraries belonging to multiple other dialects. Those dialects don't
need to depend on the non-interface part of the transform dialect, which
includes the growing number of ops and transitive dependency footprint.
Split out the interfaces into a separate library. This in turn requires
flipping the dependency from the interface on the dialect that has crept
in because both co-existed in one library. The interface shouldn't
depend on the transform dialect either.
As a consequence of splitting, the capability of the interpreter to
automatically walk the payload IR to identify payload ops of a certain
kind based on the type used for the entry point symbol argument is
disabled. This is a good move by itself as it simplifies the interpreter
logic. This functionality can be trivially replaced by a
`transform.structured.match` operation.
|
|
unify tiling with `scf.for` and `scf.forall`. (#77874)
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`
|
|
ops. (#77204)
In the process a couple of test transform dialect ops are added just
for testing. These operations are not intended to use as full flushed
out of transformation ops, but are rather operations added for testing.
A separate operation is added to `LinalgTransformOps.td` to convert a
`TilingInterface` operation to loops using the
`generateScalarImplementation` method implemented by the
operation. Eventually this and other operations related to tiling
using the `TilingInterface` need to move to a better place (i.e. out
of `Linalg` dialect)
|
