Summary:
Generalize broadcastable trait to variadic operands. Update the
documentation that still talked about element type as part of
broadcastable trait (that bug was already fixed). Also rename
Broadcastable to ResultBroadcastableShape to be more explicit that the
trait affects the result shape (it is possible for op to allow
broadcastable operands but not have result shape that is broadcast
compatible with operands).
Doing some intermediate work to have getBroadcastedType take an optional
elementType as input and use that if specified, instead of the common
element type of type1 and type2 in this function.
Differential Revision: https://reviews.llvm.org/D72559
Summary:
This is a simple extension to allow vectorization to work not only on GenericLinalgOp
but more generally across named ops too.
For now, this still only vectorizes matmul-like ops but is a step towards more
generic vectorization of Linalg ops.
Reviewers: ftynse
Subscribers: mehdi_amini, rriddle, jpienaar, burmako, shauheen, antiagainst, arpith-jacob, mgester, lucyrfox, aartbik, liufengdb, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D72942
Summary:
First step towards the consolidation
of a lot of vector related utilities
that are now all over the place
(or even duplicated).
Reviewers: nicolasvasilache, andydavis1
Reviewed By: nicolasvasilache, andydavis1
Subscribers: merge_guards_bot, mehdi_amini, rriddle, jpienaar, burmako, shauheen, antiagainst, arpith-jacob, mgester, lucyrfox, liufengdb, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D72955
Summary:
* Add shaped container type interface which allows infering the shape, element
type and attribute of shaped container type separately. Show usage by way of
tensor type inference trait which combines the shape & element type in
infering a tensor type;
- All components need not be specified;
- Attribute is added to allow for layout attribute that was previously
discussed;
* Expand the test driver to make it easier to test new creation instances
(adding new operands or ops with attributes or regions would trigger build
functions/type inference methods);
- The verification part will be moved out of the test and to verify method
instead of ops implementing the type inference interface in a follow up;
* Add MLIRContext as arg to possible to create type for ops without arguments,
region or location;
* Also move out the section in OpDefinitions doc to separate ShapeInference doc
where the shape function requirements can be captured;
- Part of this would move to the shape dialect and/or shape dialect ops be
included as subsection of this doc;
* Update ODS's variable usage to match camelBack format for builder,
state and arg variables;
- I could have split this out, but I had to make some changes around
these and the inconsistency bugged me :)
Differential Revision: https://reviews.llvm.org/D72432
Summary:
The visibility defines the structural reachability of the symbol within the IR. Symbols can define one of three visibilities:
* Public
The symbol \may be accessed from outside of the visible IR. We cannot assume that we can observe all of the uses of this symbol.
* Private
The symbol may only be referenced from within the operations in the current symbol table, via SymbolRefAttr.
* Nested
The symbol may be referenced by operations in symbol tables above the current symbol table, as long as each symbol table parent also defines a non-private symbol. This allows or referencing the symbol from outside of the defining symbol table, while retaining the ability for the compiler to see all uses.
These properties help to reason about the properties of a symbol, and will be used in a follow up to implement a dce pass on dead symbols.
A few examples of what this would look like in the IR are shown below:
module @public_module {
// This function can be accessed by 'live.user'
func @nested_function() attributes { sym_visibility = "nested" }
// This function cannot be accessed outside of 'public_module'
func @private_function() attributes { sym_visibility = "private" }
}
// This function can only be accessed from within this module.
func @private_function() attributes { sym_visibility = "private" }
// This function may be referenced externally.
func @public_function()
"live.user"() {uses = [@public_module::@nested_function,
@private_function,
@public_function]} : () -> ()
Depends On D72043
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D72044
Summary:
This enables tracking calls that cross symbol table boundaries. It also simplifies some of the implementation details of CallableOpInterface, i.e. there can only be one region within the callable operation.
Depends On D72042
Reviewed By: jpienaar
Differential Revision: https://reviews.llvm.org/D72043
Summary: This updates the use list algorithms to support querying from a specific symbol, allowing for the collection and detection of nested references. This works by walking the parent "symbol scopes" and applying the existing algorithm at each level.
Reviewed By: jpienaar
Differential Revision: https://reviews.llvm.org/D72042
Summary: Introduce m_Constant() which allows matching a constant operation without forcing the user also to capture the attribute value.
Differential Revision: https://reviews.llvm.org/D72397
Introduce a set of function that promote a memref argument of a `gpu.func` to
workgroup memory using memory attribution. The promotion boils down to
additional loops performing the copy from the original argument to the
attributed memory in the beginning of the function, and back at the end of the
function using all available threads. The loop bounds are specified so as to
adapt to any size of the workgroup. These utilities are intended to compose
with other existing utilities (loop coalescing and tiling) in cases where the
distribution of work across threads is uneven, e.g. copying a 2D memref with
only the threads along the "x" dimension. Similarly, specialization of the
kernel to specific launch sizes should be implemented as a separate pass
combining constant propagation and canonicalization.
Introduce a simple attribute-driven pass to test the promotion transformation
since we don't have a heuristic at the moment.
Differential revision: https://reviews.llvm.org/D71904
Summary:
This changes the implementation of OpResult to have some of the results be represented inline in Value, via a pointer int pair of Operation*+result number, and the rest being trailing objects on the main operation. The full details of the new representation is detailed in the proposal here:
https://groups.google.com/a/tensorflow.org/g/mlir/c/XXzzKhqqF_0/m/v6bKb08WCgAJ
The only difference between here and the above proposal is that we only steal 2-bits for the Value kind instead of 3. This means that we can only fit 2-results inline instead of 6. This allows for other users to steal the final bit for PointerUnion/etc. If necessary, we can always steal this bit back in the future to save more space if 3-6 results are common enough.
Reviewed By: jpienaar
Differential Revision: https://reviews.llvm.org/D72020
Summary:
This diff adds support to allow `linalg.generic` and
`linalg.indexed_generic` to take tensor input and output
arguments.
The subset of output tensor operand types must appear
verbatim in the result types after an arrow. The parser,
printer and verifier are extended to accomodate this
behavior.
The Linalg operations now support variadic ranked tensor
return values. This extension exhibited issues with the
current handling of NativeCall in RewriterGen.cpp. As a
consequence, an explicit cast to `SmallVector<Value, 4>`
is added in the proper place to support the new behavior
(better suggestions are welcome).
Relevant cleanups and name uniformization are applied.
Relevant invalid and roundtrip test are added.
Reviewers: mehdi_amini, rriddle, jpienaar, antiagainst, ftynse
Subscribers: burmako, shauheen, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D72022
Summary:
This is part of an ongoing cleanup and uniformization work.
This diff performs 3 types of cleanups:
1. Uniformize transformation names.
2. Replace all pattern operands that need not be captured by `$_`
3. Replace all usage of pattern captured op by the normalized `op` name (instead of positional parameters such as `$0`)
Reviewers: ftynse
Subscribers: mehdi_amini, rriddle, jpienaar, burmako, shauheen, antiagainst, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D72081
This change refactors pass options to be more similar to how statistics are modeled. More specifically, the options are specified directly on the pass instead of in a separate options class. (Note that the behavior and specification for pass pipelines remains the same.) This brings about several benefits:
* The specification of options is much simpler
* The round-trip format of a pass can be generated automatically
* This gives a somewhat deeper integration with "configuring" a pass, which we could potentially expose to users in the future.
PiperOrigin-RevId: 286953824
This is an initial step to refactoring the representation of OpResult as proposed in: https://groups.google.com/a/tensorflow.org/g/mlir/c/XXzzKhqqF_0/m/v6bKb08WCgAJ
This change will make it much simpler to incrementally transition all of the existing code to use value-typed semantics.
PiperOrigin-RevId: 286844725
This is the block argument equivalent of the existing `getAsmResultNames` hook.
Closestensorflow/mlir#329
COPYBARA_INTEGRATE_REVIEW=https://github.com/tensorflow/mlir/pull/329 from plaidml:flaub-region-arg-names fc7876f2d1335024e441083cd25263fd6247eb7d
PiperOrigin-RevId: 286523299
This enables providing a default implementation of an interface method. This method is defined on the Trait that is attached to the operation, and thus has all of the same constraints and properties as any other interface method. This allows for interface authors to provide a conservative default implementation for certain methods, without requiring that all users explicitly define it. The default implementation can be specified via the argument directly after the interface method body:
StaticInterfaceMethod<
/*desc=*/"Returns whether two array of types are compatible result types for an op.",
/*retTy=*/"bool",
/*methodName=*/"isCompatibleReturnTypes",
/*args=*/(ins "ArrayRef<Type>":$lhs, "ArrayRef<Type>":$rhs),
/*methodBody=*/[{
return ConcreteOp::isCompatibleReturnTypes(lhs, rhs);
}],
/*defaultImplementation=*/[{
/// Returns whether two arrays are equal as strongest check for
/// compatibility by default.
return lhs == rhs;
}]
PiperOrigin-RevId: 286226054
This PR targest issue tensorflow/mlir#295. It exposes the already existing
subiew promotion pass as a declarative pattern
Change-Id: If901ebef9fb53fcd0b12ecc536f6b174ce320b92
Closestensorflow/mlir#315
COPYBARA_INTEGRATE_REVIEW=https://github.com/tensorflow/mlir/pull/315 from tetuante:issue295 8e5f268b6d85f31015c33505329dbd7a4db97ac5
PiperOrigin-RevId: 285801463
This change allows for DialectConversion to attempt folding as a mechanism to legalize illegal operations. This also expands folding support in OpBuilder::createOrFold to generate new constants when folding, and also enables it to work in the context of a PatternRewriter.
PiperOrigin-RevId: 285448440
This CL adds more common information to StructuredOpsUtils.h
The n_view attribute is retired in favor of args_in + args_out but the CL is otherwise NFC.
PiperOrigin-RevId: 285000621
This patch closes issue tensorflow/mlir#272
We add a standalone iterator permutation transformation to Linalg.
This transformation composes a permutation map with the maps in the
"indexing_maps" attribute. It also permutes "iterator_types"
accordingly.
Change-Id: I7c1e693b8203aeecc595a7c012e738ca1100c857
Closestensorflow/mlir#307
COPYBARA_INTEGRATE_REVIEW=https://github.com/tensorflow/mlir/pull/307 from tetuante:issue272 f7908d58792f4111119721885e247045104f1131
PiperOrigin-RevId: 284824102
This reorganizes the vector transformations to be more easily testable as patterns and more easily composable into fused passes in the future.
PiperOrigin-RevId: 284817474
This CL uses the newly expanded matcher support to easily detect when a linalg.generic has a multiply-accumulate body. A linalg.generic with such a body is rewritten as a vector contraction.
This CL additionally limits the rewrite to the case of matrix multiplication on contiguous and statically shaped memrefs for now.
Before expanding further, we should harden the infrastructure for expressing custom ops with the structured ops abstraction.
PiperOrigin-RevId: 284566659
Follows ValueRange in representing a generic abstraction over the different
ways to represent a range of Regions. This wrapper is not as ValueRange and only
considers the current cases of interest: MutableArrayRef<Region> and
ArrayRef<std::unique_ptr<Region>> as occurs during op construction vs op region
querying.
Note: ArrayRef<std::unique_ptr<Region>> allows for unset regions, so this range
returns a pointer to a Region instead of a Region.
PiperOrigin-RevId: 284563229
This CL adds support for building matchers recursively.
The following matchers are provided:
1. `m_any()` can match any value
2. `m_val(Value *)` binds to a value and must match it
3. `RecursivePatternMatcher<OpType, Matchers...>` n-arity pattern that matches `OpType` and whose operands must be matched exactly by `Matchers...`.
This allows building expression templates for patterns, declaratively, in a very natural fashion.
For example pattern `p9` defined as follows:
```
auto mul_of_muladd = m_Op<MulFOp>(m_Op<MulFOp>(), m_Op<AddFOp>());
auto mul_of_anyadd = m_Op<MulFOp>(m_any(), m_Op<AddFOp>());
auto p9 = m_Op<MulFOp>(m_Op<MulFOp>(
mul_of_muladd, m_Op<MulFOp>()),
m_Op<MulFOp>(mul_of_anyadd, mul_of_anyadd));
```
Successfully matches `%6` in:
```
%0 = addf %a, %b: f32
%1 = addf %a, %c: f32 // matched
%2 = addf %c, %b: f32
%3 = mulf %a, %2: f32 // matched
%4 = mulf %3, %1: f32 // matched
%5 = mulf %4, %4: f32 // matched
%6 = mulf %5, %5: f32 // matched
```
Note that 0-ary matchers can be used as leaves in place of n-ary matchers. This alleviates from passing explicit `m_any()` leaves.
In the future, we may add extra patterns to specify that operands may be matched in any order.
PiperOrigin-RevId: 284469446
This allows for users to provide operand_range and result_range in builder.create<> calls, instead of requiring an explicit copy into a separate data structure like SmallVector/std::vector.
PiperOrigin-RevId: 284360710
Previously the error case was using a sentinel in the error case which was bad. Also make the one `build` invoke the other `build` to reuse verification there.
And follow up on suggestion to use formatv which I missed during previous review.
PiperOrigin-RevId: 284265762
For ops with infer type op interface defined, generate version that calls the inferal method on build. This is intermediate step to removing special casing of SameOperandsAndResultType & FirstAttrDereivedResultType. After that would be generating the inference code, with the initial focus on shaped container types. In between I plan to refactor these a bit to reuse generated paths. The intention would not be to add the type inference trait in multiple places, but rather to take advantage of the current modelling in ODS where possible to emit it instead.
Switch the `inferReturnTypes` method to be static.
Skipping ops with regions here as I don't like the Region vs unique_ptr<Region> difference at the moment, and I want the infer return type trait to be useful for verification too. So instead, just skip it for now to avoid churn.
PiperOrigin-RevId: 284217913
This patch closes issue tensorflow/mlir#271.
It adds an optional permutation map to declarative tiling transformations.
The map is expressed as a list of integers.
Closestensorflow/mlir#288
COPYBARA_INTEGRATE_REVIEW=https://github.com/tensorflow/mlir/pull/288 from tetuante:issue271 2df2938d6a1f01b3bc404ded08dea2dd1e10b588
PiperOrigin-RevId: 284064151
Statistics are a way to keep track of what the compiler is doing and how effective various optimizations are. It is useful to see what optimizations are contributing to making a particular program run faster. Pass-instance specific statistics take this even further as you can see the effect of placing a particular pass at specific places within the pass pipeline, e.g. they could help answer questions like "what happens if I run CSE again here".
Statistics can be added to a pass by simply adding members of type 'Pass::Statistics'. This class takes as a constructor arguments: the parent pass pointer, a name, and a description. Statistics can be dumped by the pass manager in a similar manner to how pass timing information is dumped, i.e. via PassManager::enableStatistics programmatically; or -pass-statistics and -pass-statistics-display via the command line pass manager options.
Below is an example:
struct MyPass : public OperationPass<MyPass> {
Statistic testStat{this, "testStat", "A test statistic"};
void runOnOperation() {
...
++testStat;
...
}
};
$ mlir-opt -pass-pipeline='func(my-pass,my-pass)' foo.mlir -pass-statistics
Pipeline Display:
===-------------------------------------------------------------------------===
... Pass statistics report ...
===-------------------------------------------------------------------------===
'func' Pipeline
MyPass
(S) 15 testStat - A test statistic
MyPass
(S) 6 testStat - A test statistic
List Display:
===-------------------------------------------------------------------------===
... Pass statistics report ...
===-------------------------------------------------------------------------===
MyPass
(S) 21 testStat - A test statistic
PiperOrigin-RevId: 284022014
This CL refactors some of the MLIR vector dependencies to allow decoupling VectorOps, vector analysis, vector transformations and vector conversions from each other.
This makes the system more modular and allows extracting VectorToVector into VectorTransforms that do not depend on vector conversions.
This refactoring exhibited a bunch of cyclic library dependencies that have been cleaned up.
PiperOrigin-RevId: 283660308
Right now op argument matching in DRR is position-based, meaning we need to
specify N arguments for an op with N ODS-declared argument. This can be annoying
when we don't want to capture all the arguments. `$_` is to remedy the situation.
PiperOrigin-RevId: 283339992
Jacques Pienaar