This aligns the SCF dialect file layout with the majority of the dialects.
Reviewed By: jpienaar
Differential Revision: https://reviews.llvm.org/D128049
With the recent refactorings, this class is no longer needed. We can use BufferizationOptions in all places were BufferizationState was used.
Differential Revision: https://reviews.llvm.org/D127653
This 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
contraction op can have mixed type, add support for this case to the pattern
lowering contraction op to outerproduct.
Differential Revision: https://reviews.llvm.org/D127926
Make the reduction distribution pattern more generic and remove layering
problem. The new pattern to distribute reduction is now independent of
GPU and takes a lamdba to decide how the distributed reduction should be
generated.
Differential Revision: https://reviews.llvm.org/D127867
Add a pattern to do ad hoc lowering of vector.reduction to a sequence of
warp shuffles. This allow distributing reduction on a warp for GPU targets.
Also add an execution test for warp reduction.
co-authored with @springerm
Differential Revision: https://reviews.llvm.org/D127176
Add patterns to propagate vector distribution and remove dead
arguments. This handles propagation for several vector operations.
recommit after minor bug fix.
Differential Revision: https://reviews.llvm.org/D127167
Add patterns to propagate vector distribution and remove dead
arguments. This handles propagation for several vector operations.
Differential Revision: https://reviews.llvm.org/D127167
Add pattern to hoist scalar code outside of warp distribute region as
those cannot be distributed and we would want to execute them on all
the lanes.
Add patterns to distribute transfer_write ops. Those operations can be
distributed in different ways and it is control by user.
Differential Revision: https://reviews.llvm.org/D127152
Fixed issue with vector.contract default unroll permutation.
Adds support for vector unroll transformations to unroll in different
orders. For example, the vector.contract can be unrolled into a
smaller set of contractions. There is a choice of how to unroll the
decomposition based on the traversal order of (dim0, dim1, dim2).
The choice of traversal order can now be specified by a callback which
given by the caller of the transform. For now, only the
vector.contract, vector.transfer_read/transfer_write operations
support the callback.
Differential Revision: https://reviews.llvm.org/D127004
Reverts commit 1469ebf8382107e0344173f362b690d19e24029d (original commit)
Reverts commit a392a39f75af586e3d3cd046a8361939277e067f (build fix for above commit)
The commit broke tests in out-of-tree projects, indicating that some logical
error was made in the previous change but not covered by current tests.
Adds supprot for vector unroll transformations to unroll in different
orders. For example, the `vector.contract` can be unrolled into a
smaller set of contractions. There is a choice of how to unroll the
decomposition based on the traversal order of (dim0, dim1, dim2).
The choice of traversal order can now be specified by a callback which
given by the caller of the transform. For now, only the
`vector.contract`, `vector.transfer_read/transfer_write` operations
support the callback.
Differential Revision: https://reviews.llvm.org/D127004
Add lowering for cases where the reduction dimension is fully unrolled.
It is common to unroll the reduction dimension, therefore we would want
to lower the contractions to an elementwise vector op in this case.
Differential Revision: https://reviews.llvm.org/D126120
Add lowering of the vector.warp_execute_on_lane_0 into scf.if plus memory
transfer for the operands and yield values.
This also add an integration test running on GPU warp. The same tests can be
later re-used with different comment lines to tests distribution
transformations.
This is mostly from @springerm contribution.
Differential Revision: https://reviews.llvm.org/D125430
After https://reviews.llvm.org/D119743 added the `AutomaticAllocationScope`
trait to loop-like constructs, the vector transfer full/partial splitting pass
started inserting allocations for temporaries within the closest loop rather
than the closest function (or other allocation scope such as `async.execute`).
While this is correct as long as the lowered code takes care of automatic
deallocation at the end of each iteration of the loop, this interferes with
downstream optimizations that expect `alloca`s to be at the function level.
Step over loops when looking for the closest allocation scope in vector
transfer full/partial splitting pass thus restoring the original behavior.
Reviewed By: hanchung
Differential Revision: https://reviews.llvm.org/D124366
Now that dialect constructors are generated in the .cpp file, we can
drop all of the dependent dialect includes from the .h file.
Differential Revision: https://reviews.llvm.org/D124298
Similar to the existing pattern for reodering cast(transpose),
this makes transpose following transpose and increases the chance
of embedding the transposition inside contraction op. Actually
cast ops are just special instances of elementwise ops.
Reviewed By: ThomasRaoux
Differential Revision: https://reviews.llvm.org/D123596
Support unrolling for vector.transpose following the same interface as
other vector unrolling ops.
Differential Revision: https://reviews.llvm.org/D123688
Right now `populateVectorInsertExtractStridedSliceTransforms` contains
two categories of patterns, one for decomposing high-D insert/extract
strided slices, the other for lowering them to shuffle ops.
They are at different levels---the former is in the middle, while
the latter is a step of final lowering. Split them to give users
more control of which pattern to pick.
This means break down the previous `VectorExtractStridedSliceOpRewritePattern`,
which is doing two things together.
Also renamed those patterns to be clearer.
Reviewed By: ThomasRaoux
Differential Revision: https://reviews.llvm.org/D123137
We are using "enable-index-optimizations" and "indexOptimizations" as
names for an optimization that consists of using i32 for indices within
a vector. For instance, when building a vector comparison for mask
generation. The name is confusing and suggests a scope beyond these
vector indices. This change makes the function of the option explicit
in its name.
Differential Revision: https://reviews.llvm.org/D122415
This has been on _Both for a couple of weeks. Flip usages in core with
intention to flip flag to _Prefixed in follow up. Needed to add a couple
of helper methods in AffineOps and Linalg to facilitate a pure flag flip
in follow up as some of these classes are used in templates and so
sensitive to Vector dialect changes.
Differential Revision: https://reviews.llvm.org/D122151
The way vector.create_mask is currently lowered is
vector-length-dependent, and therefore incompatible with scalable vector
types. This patch adds an alternative lowering path for create_mask
operations that return a scalable vector mask.
Differential Revision: https://reviews.llvm.org/D118248
This provides a way to create an operation without manipulating
OperationState directly. This is useful for creating unregistered ops.
Reviewed By: rriddle, mehdi_amini
Differential Revision: https://reviews.llvm.org/D120787
The current dialect registry allows for attaching delayed interfaces, that are added to attrs/dialects/ops/etc.
when the owning dialect gets loaded. This is clunky for quite a few reasons, e.g. each interface type has a
separate tracking structure, and is also quite limiting. This commit refactors this delayed mutation of
dialect constructs into a more general DialectExtension mechanism. This mechanism is essentially a registration
callback that is invoked when a set of dialects have been loaded. This allows for attaching interfaces directly
on the loaded constructs, and also allows for loading new dependent dialects. The latter of which is
extremely useful as it will now enable dependent dialects to only apply in the contexts in which they
are necessary. For example, a dialect dependency can now be conditional on if a user actually needs the
interface that relies on it.
Differential Revision: https://reviews.llvm.org/D120367
This commit moves FuncOp out of the builtin dialect, and into the Func
dialect. This move has been planned in some capacity from the moment
we made FuncOp an operation (years ago). This commit handles the
functional aspects of the move, but various aspects are left untouched
to ease migration: func::FuncOp is re-exported into mlir to reduce
the actual API churn, the assembly format still accepts the unqualified
`func`. These temporary measures will remain for a little while to
simplify migration before being removed.
Differential Revision: https://reviews.llvm.org/D121266
Implement the vectorLoopUnroll interface for MultiDimReduceOp and add a
pattern to do the unrolling following the same interface other vector
unroll patterns.
Differential Revision: https://reviews.llvm.org/D121263
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<?x?xf32> -> tensor<?x?xf32>
```
changes to
```
%1 = linalg.fill ins(%cst : f32) outs(%0 : tensor<?x?xf32>) -> tensor<?x?xf32>
```
Additionally, the builder signature now takes input and output value ranges as it is the case for all other OpDSL operations:
```
rewriter.create<linalg::FillOp>(loc, val, output)
```
changes to
```
rewriter.create<linalg::FillOp>(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 D120726
Reviewed By: nicolasvasilache
Differential Revision: https://reviews.llvm.org/D120728
This patch removes an old recursive implementation to lower vector.transpose to extract/insert operations
and replaces it with a iterative approach that leverages newer linearization/delinearization utilities.
The patch should be NFC except by the order in which the extract/insert ops are generated.
Reviewed By: nicolasvasilache
Differential Revision: https://reviews.llvm.org/D121321
FuncOp isn't really important to hardcode here, it is only used to act
as a root operation for the transformation.
Differential Revision: https://reviews.llvm.org/D121195
Currently, the transfer mask is materialized by generating the vector
comparison: [offset + 0, .., offset + length - 1] < [dim, .., dim]
A better alternative is to materialize the transfer mask by using the
operation: `vector.create_mask (dim - offset)`, which will generate
simpler code and compose better with scalable vectors.
Differential Revision: https://reviews.llvm.org/D120487
In quantized comutation, there are casting ops around computation ops.
Reorder the ops to make reduce-to-contract actually work.
Reviewed By: ThomasRaoux
Differential Revision: https://reviews.llvm.org/D120760
The default lowering of vector transpose operations generates a large sequence of
scalar extract/insert operations, one pair for each scalar element in the input tensor.
In other words, the vector transpose is scalarized. However, there are transpose
patterns where one or more adjacent high-order dimensions are not transposed (for
example, in the transpose pattern [1, 0, 2, 3], dimensions 2 and 3 are not transposed).
This patch improves the lowering of those cases by not scalarizing them and extracting/
inserting a full n-D vector, where 'n' is the number of adjacent high-order dimensions
not being transposed. By doing so, we prevent the scalarization of the code and generate a
more performant vector version.
Paradoxically, this patch shouldn't improve the performance of transpose operations if
we are using LLVM. The LLVM pipeline is able to optimize away some of the extract/insert
operations and the SLP vectorizer is converting the scalar operations back to its vector
form. However, scalarizing a vector version of the code in MLIR and relying on the SLP
vectorizer to reconstruct the vector code again is highly undesirable for several reasons.
Reviewed By: nicolasvasilache, ThomasRaoux
Differential Revision: https://reviews.llvm.org/D120601
If the result operand has a unit leading dim it is removed from all operands.
Reviewed By: ThomasRaoux
Differential Revision: https://reviews.llvm.org/D119206
