Added support to the Std dialect cast operations to do casts in vector types when feasible.
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D87410
Type converter may fail and return nullptr on unconvertible types. The function
conversion did not include a check and was attempting to use a nullptr type to
construct an LLVM function, leading to a crash. Add a check and return early.
The rest of the call stack propagates errors properly.
Fixes PR47403.
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D87075
This introduces a builder for the more general case that supports zero
elements (where the element type can't be inferred from the ValueRange,
since it might be empty).
Also, fix up some cases in ShapeToStandard lowering that hit this. It
happens very easily when dealing with shapes of 0-D tensors.
The SameOperandsAndResultElementType is redundant with the new
TypesMatchWith and prevented having zero elements.
Differential Revision: https://reviews.llvm.org/D87492
Addressed some CR issues pointed out in D87111. Formatting and other nits.
The original Diff D87111 - Add an option for unrolling loops up to a factor.
Reviewed By: bondhugula
Differential Revision: https://reviews.llvm.org/D87313
This revision refactors and cleans up a bunch of things to simplify StructuredOpInterface
before work can proceed on Linalg on tensors:
- break out pieces of the StructuredOps trait that are part of the StructuredOpInterface,
- drop referenceIterators and referenceIndexingMaps that end up being more confusing than useful,
- drop NamedStructuredOpTrait
Previously only the input type was printed, and the parser applied it to
both input and output, creating an invalid transpose. Print and parse
both types, and verify that they match.
Differential Revision: https://reviews.llvm.org/D87462
This patch adds a new named structured op to accompany linalg.matmul and
linalg.matvec. We needed it for our codegen, so I figured it would be useful
to add it to Linalg.
Reviewed By: nicolasvasilache, mravishankar
Differential Revision: https://reviews.llvm.org/D87292
Rationale:
After some discussion we decided that it is safe to assume 32-bit
indices for all subscripting in the vector dialect (it is unlikely
the dialect will be used; or even work; for such long vectors).
So rather than detecting specific situations that can exploit
32-bit indices with higher parallel SIMD, we just optimize it
by default, and let users that don't want it opt-out.
Reviewed By: nicolasvasilache, bkramer
Differential Revision: https://reviews.llvm.org/D87404
I was having a lot of trouble parsing the messages. In particular, the
messages like:
```
<stdin>:3:8: error: 'scf.if' op along control flow edge from Region #0 to scf.if source #1 type '!npcomprt.tensor' should match input #1 type 'tensor<?xindex>'
```
In particular, one thing that kept catching me was parsing the "to scf.if
source #1 type" as one thing, but really it is
"to parent results: source type #1".
Differential Revision: https://reviews.llvm.org/D87334
This commit specifies reduction dimensions for ConvOps. This prevents
running reduction loops in parallel and enables easier detection of kernel dimensions
which we will need later on.
Differential Revision: https://reviews.llvm.org/D87288
The current BufferPlacement transformation cannot handle loops properly. Buffers
passed via backedges will not be freed automatically introducing memory leaks.
This CL adds support for loops to overcome these limitations.
Differential Revision: https://reviews.llvm.org/D85513
Take advantage of the new `dynamic_tensor_from_elements` operation in `std`.
Instead of stack-allocated memory, we can now lower directly to a single `std`
operation.
Differential Revision: https://reviews.llvm.org/D86935
Currently, there is no option to allow for unrolling a loop up to a specific factor (specified by the user).
The code for doing that is there and there are benefits when unrolling is done to smaller loops (smaller than the factor specified).
Reviewed By: bondhugula
Differential Revision: https://reviews.llvm.org/D87111
This replaces the select chain for edge-padding with an scf.if that
performs the memory operation when the index is in bounds and uses the
pad value when it's not. For transfer_write the same mechanism is used,
skipping the store when the index is out of bounds.
The integration test has a bunch of cases of how I believe this should
work.
Differential Revision: https://reviews.llvm.org/D87241
In this commit a new way of convolution ops lowering is introduced.
The conv op vectorization pass lowers linalg convolution ops
into vector contractions. This lowering is possible when conv op
is first tiled by 1 along specific dimensions which transforms
it into dot product between input and kernel subview memory buffers.
This pass converts such conv op into vector contraction and does
all necessary vector transfers that make it work.
Differential Revision: https://reviews.llvm.org/D86619
With `dynamic_tensor_from_elements` tensor values of dynamic size can be
created. The body of the operation essentially maps the index space to tensor
elements.
Declare SCF operations in the `scf` namespace to avoid name clash with the new
`std.yield` operation. Resolve ambiguities between `linalg/shape/std/scf.yield`
operations.
Differential Revision: https://reviews.llvm.org/D86276
Vector to SCF conversion still had issues due to the interaction with the natural alignment derived by the LLVM data layout. One traditional workaround is to allocate aligned. However, this does not always work for vector sizes that are non-powers of 2.
This revision implements a more portable mechanism where the intermediate allocation is always a memref of elemental vector type. AllocOp is extended to use the natural LLVM DataLayout alignment for non-scalar types, when the alignment is not specified in the first place.
An integration test is added that exercises the transfer to scf.for + scalar lowering with a 5x5 transposition.
Differential Revision: https://reviews.llvm.org/D87150
* Resolves todos from D87091.
* Also modifies PyConcreteAttribute to follow suite (should be useful for ElementsAttr and friends).
* Adds a test to ensure that the ShapedType base class functions as expected.
Differential Revision: https://reviews.llvm.org/D87208
Based on the PyType and PyConcreteType classes, this patch implements the bindings of Shaped Type, Tensor Type and MemRef Type subclasses.
The Tensor Type and MemRef Type are bound as ranked and unranked separately.
This patch adds the ***GetChecked C API to make sure the python side can get a valid type or a nullptr.
Shaped type is not a kind of standard types, it is the base class for vectors, memrefs and tensors, this patch binds the PyShapedType class as the base class of Vector Type, Tensor Type and MemRef Type subclasses.
Reviewed By: stellaraccident
Differential Revision: https://reviews.llvm.org/D87091
Historically, the operations in the MLIR's LLVM dialect only checked that the
operand are of LLVM dialect type without more detailed constraints. This was
due to LLVM dialect types wrapping LLVM IR types and having clunky verification
methods. With the new first-class modeling, it is possible to define type
constraints similarly to other dialects and use them to enforce some
correctness rules in verifiers instead of having LLVM assert during translation
to LLVM IR. This hardening discovered several issues where MLIR was producing
LLVM dialect operations that cannot exist in LLVM IR.
Depends On D85900
Reviewed By: rriddle
Differential Revision: https://reviews.llvm.org/D85901
When allowed, use 32-bit indices rather than 64-bit indices in the
SIMD computation of masks. This runs up to 2x and 4x faster on
a number of AVX2 and AVX512 microbenchmarks.
Reviewed By: bkramer
Differential Revision: https://reviews.llvm.org/D87116
Sizes of tiles (subviews) are bigger by 1 than they should. Let's consider
1D convolution without batches or channels. Furthermore let m iterate over
the output and n over the kernel then input is accessed with m + n. In tiling
subview sizes for convolutions are computed by applying requested tile size
together with kernel size to the above mentioned expression thus let's say
for tile size of 2 the subview size is 2 + size(n), which is bigger by one
than it should since we move kernel only once. The problem behind it is that
range is not turned into closed interval before the composition. This commit
fixes the problem by turning ranges first into closed intervals by substracting
1 and after the composition back to half open by adding 1.
Differential Revision: https://reviews.llvm.org/D86638
This patch adds basic support for vectorization of uniform values to SuperVectorizer.
For now, only invariant values to the target vector loops are considered uniform. This
enables the vectorization of loops that use function arguments and external definitions
to the vector loops. We could extend uniform support in the future if we implement some
kind of divergence analysis algorithm.
Reviewed By: nicolasvasilache, aartbik
Differential Revision: https://reviews.llvm.org/D86756
Make use of affine memory op interfaces in AffineLoopInvariantCodeMotion so
that it can also work on affine.vector_load and affine.vector_store ops.
Reviewed By: bondhugula
Differential Revision: https://reviews.llvm.org/D86986
Make sure that memory ops that are defined inside the loop are registered
as such in 'defineOp'. In the test provided, the 'mulf' op was hoisted
outside the loop nest even when its 'affine.load' operand was not.
Reviewed By: bondhugula
Differential Revision: https://reviews.llvm.org/D86982
In this PR, the users of BufferPlacement can configure
BufferAssginmentTypeConverter. These new configurations would give the user more
freedom in the process of converting function signature, and return and call
operation conversions.
These are the new features:
- Accepting callback functions for decomposing types (i.e. 1 to N type
conversion such as unpacking tuple types).
- Defining ResultConversionKind for specifying whether a function result
with a certain type should be appended to the function arguments list or
should be kept as function result. (Usage:
converter.setResultConversionKind<MemRefType>(AppendToArgumentList))
- Accepting callback functions for composing or decomposing values (i.e. N
to 1 and 1 to N value conversion).
Differential Revision: https://reviews.llvm.org/D85133
This reverts commit 94f5d24877 because
of failing the following tests:
MLIR :: Dialect/Linalg/tensors-to-buffers.mlir
MLIR :: Transforms/buffer-placement-preparation-allowed-memref-results.mlir
MLIR :: Transforms/buffer-placement-preparation.mlir
Added 128 byte alignment to alloc ops created in VectorToSCF pass.
128b alignment was already introduced to this pass but not to all alloc
ops. This commit changes that by adding 128b alignment to the remaining ops.
The point of specifying alignment is to prevent possible memory alignment errors
on weakly tested architectures.
Differential Revision: https://reviews.llvm.org/D86454
In this PR, the users of BufferPlacement can configure
BufferAssginmentTypeConverter. These new configurations would give the user more
freedom in the process of converting function signature, and return and call
operation conversions.
These are the new features:
- Accepting callback functions for decomposing types (i.e. 1 to N type
conversion such as unpacking tuple types).
- Defining ResultConversionKind for specifying whether a function result
with a certain type should be appended to the function arguments list or
should be kept as function result. (Usage:
converter.setResultConversionKind<MemRefType>(AppendToArgumentList))
- Accepting callback functions for composing or decomposing values (i.e. N
to 1 and 1 to N value conversion).
Differential Revision: https://reviews.llvm.org/D85133
Based on the PyType and PyConcreteType classes, this patch implements the bindings of Complex Type, Vector Type and Tuple Type subclasses.
For the convenience of type checking, this patch defines a `mlirTypeIsAIntegerOrFloat` function to check whether the given type is an integer or float type.
These three subclasses in this patch have similar binding strategy:
- The function pointer `isaFunction` points to `mlirTypeIsA***`.
- The `mlir***TypeGet` C API is bound with the `get_***` method in the python side.
- The Complex Type and Vector Type check whether the given type is an integer or float type.
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D86785
Unsigned and Signless attributes use uintN_t and signed attributes use intN_t, where N is the fixed width. The 1-bit variants use bool.
Differential Revision: https://reviews.llvm.org/D86739
This patch add the missing operands to the acc.loop operation. Only the device_type
information is not part of the operation for now.
Reviewed By: rriddle, kiranchandramohan
Differential Revision: https://reviews.llvm.org/D86753
This adds some initial support for regions and does not support formatting the specific arguments of a region. For now this can be achieved by using a custom directive that formats the arguments and then parses the region.
Differential Revision: https://reviews.llvm.org/D86760
Symbol names are a special form of StringAttr that get treated specially in certain areas, such as formatting. This revision adds a special derived attr for them in ODS and adds support in the assemblyFormat for formatting them properly.
Differential Revision: https://reviews.llvm.org/D86759
This revision adds support for custom directives to the declarative assembly format. This allows for users to use C++ for printing and parsing subsections of an otherwise declaratively specified format. The custom directive is structured as follows:
```
custom-directive ::= `custom` `<` UserDirective `>` `(` Params `)`
```
`user-directive` is used as a suffix when this directive is used during printing and parsing. When parsing, `parseUserDirective` will be invoked. When printing, `printUserDirective` will be invoked. The first parameter to these methods must be a reference to either the OpAsmParser, or OpAsmPrinter. The type of rest of the parameters is dependent on the `Params` specified in the assembly format.
Differential Revision: https://reviews.llvm.org/D84719
* This is just enough to create regions/blocks and iterate over them.
* Does not yet implement the preferred iteration strategy (python pseudo containers).
* Refinements need to come after doing basic mappings of operations and values so that the whole hierarchy can be used.
Differential Revision: https://reviews.llvm.org/D86683
The tensor_reshape op was only fusible only if it is a collapsing case. Now we
propagate the op to all the operands so there is a further chance to fuse it
with generic op. The pre-conditions are:
1) The producer is not an indexed_generic op.
2) All the shapes of the operands are the same.
3) All the indexing maps are identity.
4) All the loops are parallel loops.
5) The producer has a single user.
It is possible to fuse the ops if the producer is an indexed_generic op. We
still can compute the original indices. E.g., if the reshape op collapses the d0
and d1, we can use DimOp to get the width of d1, and calculate the index
`d0 * width + d1`. Then replace all the uses with it. However, this pattern is
not implemented in the patch.
Reviewed By: mravishankar
Differential Revision: https://reviews.llvm.org/D86314
The prior diff that introduced `addAffineIfOpDomain` missed appending
constraints from the ifOp domain. This revision fixes this problem.
Differential Revision: https://reviews.llvm.org/D86421
Adding a conversion pattern for the parallel Operation. This will
help the conversion of parallel operation with standard dialect to
parallel operation with llvm dialect. The type conversion of the block
arguments in a parallel region are controlled by the pattern for the
parallel Operation. Without this pattern, a parallel Operation with
block arguments cannot be converted from standard to LLVM dialect.
Other OpenMP operations without regions are marked as legal. When
translation of OpenMP operations with regions are added then patterns
for these operations can also be added.
Also uses all the standard to llvm patterns. Patterns of other dialects
can be added later if needed.
Reviewed By: rriddle
Differential Revision: https://reviews.llvm.org/D86273
When dealing with dialects that will results in function calls to
external libraries, it is important to be able to handle maps as some
dialects may require mapped data. Before this patch, the detection of
whether normalization can apply or not, operations are compared to an
explicit list of operations (`alloc`, `dealloc`, `return`) or to the
presence of specific operation interfaces (`AffineReadOpInterface`,
`AffineWriteOpInterface`, `AffineDMAStartOp`, or `AffineDMAWaitOp`).
This patch add a trait, `MemRefsNormalizable` to determine if an
operation can have its `memrefs` normalized.
This trait can be used in turn by dialects to assert that such
operations are compatible with normalization of `memrefs` with
nontrivial memory layout specification. An example is given in the
literal tests.
Differential Revision: https://reviews.llvm.org/D86236
This patch allows to pass the gpu module name to SPIR-V
module during conversion. This has many benefits as we can lookup
converted to SPIR-V kernel in the symbol table.
In order to avoid symbol conflicts, `"__spv__"` is added to the
gpu module name to form the new one.
Reviewed By: mravishankar
Differential Revision: https://reviews.llvm.org/D86384
This patch introduces a hook to encode descriptor set
and binding number into `spv.globalVariable`'s symbolic name. This
allows to preserve this information, and at the same time legalize
the global variable for the conversion to LLVM dialect.
This is required for `mlir-spirv-cpu-runner` to convert kernel
arguments into LLVM.
Also, a couple of some nits added:
- removed unused comment
- changed to a capital letter in the comment
Reviewed By: mravishankar
Differential Revision: https://reviews.llvm.org/D86515
This patch adds an optional name to SPIR-V module.
This will help with lowering from GPU dialect (so that we
can pass the kernel module name) and will be more naturally
aligned with `GPUModuleOp`/`ModuleOp`.
Reviewed By: mravishankar
Differential Revision: https://reviews.llvm.org/D86386
As discussed in D86576, noundef attribute is removed from masked store/load/gather/scatter's
pointer operands.
Reviewed By: efriedma
Differential Revision: https://reviews.llvm.org/D86656
This patch adds NoUndef to Intrinsics.td.
The attribute is attached to llvm.assume's operand, because llvm.assume(undef)
is UB.
It is attached to pointer operands of several memory accessing intrinsics
as well.
This change makes ValueTracking::getGuaranteedNonPoisonOps' intrinsic check
unnecessary, so it is removed.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D86576
The PDL Interpreter dialect provides a lower level abstraction compared to the PDL dialect, and is targeted towards low level optimization and interpreter code generation. The dialect operations encapsulates low-level pattern match and rewrite "primitives", such as navigating the IR (Operation::getOperand), creating new operations (OpBuilder::create), etc. Many of the operations within this dialect also fuse branching control flow with some form of a predicate comparison operation. This type of fusion reduces the amount of work that an interpreter must do when executing.
An example of this representation is shown below:
```mlir
// The following high level PDL pattern:
pdl.pattern : benefit(1) {
%resultType = pdl.type
%inputOperand = pdl.input
%root, %results = pdl.operation "foo.op"(%inputOperand) -> %resultType
pdl.rewrite %root {
pdl.replace %root with (%inputOperand)
}
}
// May be represented in the interpreter dialect as follows:
module {
func @matcher(%arg0: !pdl.operation) {
pdl_interp.check_operation_name of %arg0 is "foo.op" -> ^bb2, ^bb1
^bb1:
pdl_interp.return
^bb2:
pdl_interp.check_operand_count of %arg0 is 1 -> ^bb3, ^bb1
^bb3:
pdl_interp.check_result_count of %arg0 is 1 -> ^bb4, ^bb1
^bb4:
%0 = pdl_interp.get_operand 0 of %arg0
pdl_interp.is_not_null %0 : !pdl.value -> ^bb5, ^bb1
^bb5:
%1 = pdl_interp.get_result 0 of %arg0
pdl_interp.is_not_null %1 : !pdl.value -> ^bb6, ^bb1
^bb6:
pdl_interp.record_match @rewriters::@rewriter(%0, %arg0 : !pdl.value, !pdl.operation) : benefit(1), loc([%arg0]), root("foo.op") -> ^bb1
}
module @rewriters {
func @rewriter(%arg0: !pdl.value, %arg1: !pdl.operation) {
pdl_interp.replace %arg1 with(%arg0)
pdl_interp.return
}
}
}
```
Differential Revision: https://reviews.llvm.org/D84579
Provides fast, generic way of setting a mask up to a certain
point. Potential use cases that may benefit are create_mask
and transfer_read/write operations in the vector dialect.
Reviewed By: bkramer
Differential Revision: https://reviews.llvm.org/D86501
Based on the PyType and PyConcreteType classes, this patch implements the bindings of Index Type, Floating Point Type and None Type subclasses.
These three subclasses share the same binding strategy:
- The function pointer `isaFunction` points to `mlirTypeIsA***`.
- The `mlir***TypeGet` C API is bound with the `***Type` constructor in the python side.
Reviewed By: stellaraccident
Differential Revision: https://reviews.llvm.org/D86466
* Generic mlir.ir.Attribute class.
* First standard attribute (mlir.ir.StringAttr), following the same pattern as generic vs standard types.
* NamedAttribute class.
Differential Revision: https://reviews.llvm.org/D86250
Binding MemRefs of f16 needs special handling as the type is not supported on
CPU. There was a bug in the type used.
Differential Revision: https://reviews.llvm.org/D86328
Add a folder to the affine.parallel op so that loop bounds expressions are canonicalized.
Additionally, a new AffineParallelNormalizePass is added to adjust affine.parallel ops so that the lower bound is always 0 and the upper bound always represents a range with a step size of 1.
Differential Revision: https://reviews.llvm.org/D84998
Removed the Standard to LLVM conversion patterns that were previously
pulled in for testing purposes. This helps to separate the conversion
to LLVM dialect of the MLIR module with both SPIR-V and Standard
dialects in it (particularly helpful for SPIR-V cpu runner). Also,
tests were changed accordingly.
Reviewed By: mravishankar
Differential Revision: https://reviews.llvm.org/D86285
- This utility to merge a block anywhere into another one can help inline single
block regions into other blocks.
- Modified patterns test to use the new function.
Differential Revision: https://reviews.llvm.org/D86251
Add the unsigned complements to the existing FPToSI and SIToFP operations in the
standard dialect, with one-to-one lowerings to the corresponding LLVM operations.
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D85557
PDL presents a high level abstraction for the rewrite pattern infrastructure available in MLIR. This abstraction allows for representing patterns transforming MLIR, as MLIR. This allows for applying all of the benefits that the general MLIR infrastructure provides, to the infrastructure itself. This means that pattern matching can be more easily verified for correctness, targeted by frontends, and optimized.
PDL abstracts over various different aspects of patterns and core MLIR data structures. Patterns are specified via a `pdl.pattern` operation. These operations contain a region body for the "matcher" code, and terminate with a `pdl.rewrite` that either dispatches to an external rewriter or contains a region for the rewrite specified via `pdl`. The types of values in `pdl` are handle types to MLIR C++ types, with `!pdl.attribute`, `!pdl.operation`, and `!pdl.type` directly mapping to `mlir::Attribute`, `mlir::Operation*`, and `mlir::Value` respectively.
An example pattern is shown below:
```mlir
// pdl.pattern contains metadata similarly to a `RewritePattern`.
pdl.pattern : benefit(1) {
// External input operand values are specified via `pdl.input` operations.
// Result types are constrainted via `pdl.type` operations.
%resultType = pdl.type
%inputOperand = pdl.input
%root, %results = pdl.operation "foo.op"(%inputOperand) -> %resultType
pdl.rewrite(%root) {
pdl.replace %root with (%inputOperand)
}
}
```
This is a culmination of the work originally discussed here: https://groups.google.com/a/tensorflow.org/g/mlir/c/j_bn74ByxlQ
Differential Revision: https://reviews.llvm.org/D84578
Provide C API for MLIR standard attributes. Since standard attributes live
under lib/IR in core MLIR, place the C APIs in the IR library as well (standard
ops will go in a separate library).
Affine map and integer set attributes are only exposed as placeholder types
with IsA support due to the lack of C APIs for the corresponding types.
Integer and floating point attribute APIs expecting APInt and APFloat are not
exposed pending decision on how to support APInt and APFloat.
Reviewed By: stellaraccident
Differential Revision: https://reviews.llvm.org/D86143
* The binding for Type is trivial and should be non-controversial.
* The way that I define the IntegerType should serve as a pattern for what I want to do next.
* I propose defining the rest of the standard types in this fashion and then generalizing for dialect types as necessary.
* Essentially, creating/accessing a concrete Type (vs interacting with the string form) is done by "casting" to the concrete type (i.e. IntegerType can be constructed with a Type and will throw if the cast is illegal).
* This deviates from some of our previous discussions about global objects but I think produces a usable API and we should go this way.
Differential Revision: https://reviews.llvm.org/D86179
If Memref has rank > 1 this pass emits N-1 loops around
TransferRead op and transforms the op itself to 1D read. Since vectors
must have static shape while memrefs don't the pass emits if condition
to prevent out of bounds accesses in case some memref dimension is smaller
than the corresponding dimension of targeted vector. This logic is fine
but authors forgot to apply `permutation_map` on loops upper bounds and
thus if condition compares induction variable to incorrect loop upper bound
(dimension of the memref) in case `permutation_map` is not identity map.
This commit aims to fix that.
This changes the behavior of constructing MLIRContext to no longer load globally
registered dialects on construction. Instead Dialects are only loaded explicitly
on demand:
- the Parser is lazily loading Dialects in the context as it encounters them
during parsing. This is the only purpose for registering dialects and not load
them in the context.
- Passes are expected to declare the dialects they will create entity from
(Operations, Attributes, or Types), and the PassManager is loading Dialects into
the Context when starting a pipeline.
This changes simplifies the configuration of the registration: a compiler only
need to load the dialect for the IR it will emit, and the optimizer is
self-contained and load the required Dialects. For example in the Toy tutorial,
the compiler only needs to load the Toy dialect in the Context, all the others
(linalg, affine, std, LLVM, ...) are automatically loaded depending on the
optimization pipeline enabled.
To adjust to this change, stop using the existing dialect registration: the
global registry will be removed soon.
1) For passes, you need to override the method:
virtual void getDependentDialects(DialectRegistry ®istry) const {}
and registery on the provided registry any dialect that this pass can produce.
Passes defined in TableGen can provide this list in the dependentDialects list
field.
2) For dialects, on construction you can register dependent dialects using the
provided MLIRContext: `context.getOrLoadDialect<DialectName>()`
This is useful if a dialect may canonicalize or have interfaces involving
another dialect.
3) For loading IR, dialect that can be in the input file must be explicitly
registered with the context. `MlirOptMain()` is taking an explicit registry for
this purpose. See how the standalone-opt.cpp example is setup:
mlir::DialectRegistry registry;
registry.insert<mlir::standalone::StandaloneDialect>();
registry.insert<mlir::StandardOpsDialect>();
Only operations from these two dialects can be in the input file. To include all
of the dialects in MLIR Core, you can populate the registry this way:
mlir::registerAllDialects(registry);
4) For `mlir-translate` callback, as well as frontend, Dialects can be loaded in
the context before emitting the IR: context.getOrLoadDialect<ToyDialect>()
Differential Revision: https://reviews.llvm.org/D85622
This changes the behavior of constructing MLIRContext to no longer load globally
registered dialects on construction. Instead Dialects are only loaded explicitly
on demand:
- the Parser is lazily loading Dialects in the context as it encounters them
during parsing. This is the only purpose for registering dialects and not load
them in the context.
- Passes are expected to declare the dialects they will create entity from
(Operations, Attributes, or Types), and the PassManager is loading Dialects into
the Context when starting a pipeline.
This changes simplifies the configuration of the registration: a compiler only
need to load the dialect for the IR it will emit, and the optimizer is
self-contained and load the required Dialects. For example in the Toy tutorial,
the compiler only needs to load the Toy dialect in the Context, all the others
(linalg, affine, std, LLVM, ...) are automatically loaded depending on the
optimization pipeline enabled.
To adjust to this change, stop using the existing dialect registration: the
global registry will be removed soon.
1) For passes, you need to override the method:
virtual void getDependentDialects(DialectRegistry ®istry) const {}
and registery on the provided registry any dialect that this pass can produce.
Passes defined in TableGen can provide this list in the dependentDialects list
field.
2) For dialects, on construction you can register dependent dialects using the
provided MLIRContext: `context.getOrLoadDialect<DialectName>()`
This is useful if a dialect may canonicalize or have interfaces involving
another dialect.
3) For loading IR, dialect that can be in the input file must be explicitly
registered with the context. `MlirOptMain()` is taking an explicit registry for
this purpose. See how the standalone-opt.cpp example is setup:
mlir::DialectRegistry registry;
registry.insert<mlir::standalone::StandaloneDialect>();
registry.insert<mlir::StandardOpsDialect>();
Only operations from these two dialects can be in the input file. To include all
of the dialects in MLIR Core, you can populate the registry this way:
mlir::registerAllDialects(registry);
4) For `mlir-translate` callback, as well as frontend, Dialects can be loaded in
the context before emitting the IR: context.getOrLoadDialect<ToyDialect>()
Differential Revision: https://reviews.llvm.org/D85622
This greatly simplifies a large portion of the underlying infrastructure, allows for lookups of singleton classes to be much more efficient and always thread-safe(no locking). As a result of this, the dialect symbol registry has been removed as it is no longer necessary.
For users broken by this change, an alert was sent out(https://llvm.discourse.group/t/removing-kinds-from-attributes-and-types) that helps prevent a majority of the breakage surface area. All that should be necessary, if the advice in that alert was followed, is removing the kind passed to the ::get methods.
Differential Revision: https://reviews.llvm.org/D86121
This changes the behavior of constructing MLIRContext to no longer load globally
registered dialects on construction. Instead Dialects are only loaded explicitly
on demand:
- the Parser is lazily loading Dialects in the context as it encounters them
during parsing. This is the only purpose for registering dialects and not load
them in the context.
- Passes are expected to declare the dialects they will create entity from
(Operations, Attributes, or Types), and the PassManager is loading Dialects into
the Context when starting a pipeline.
This changes simplifies the configuration of the registration: a compiler only
need to load the dialect for the IR it will emit, and the optimizer is
self-contained and load the required Dialects. For example in the Toy tutorial,
the compiler only needs to load the Toy dialect in the Context, all the others
(linalg, affine, std, LLVM, ...) are automatically loaded depending on the
optimization pipeline enabled.
To adjust to this change, stop using the existing dialect registration: the
global registry will be removed soon.
1) For passes, you need to override the method:
virtual void getDependentDialects(DialectRegistry ®istry) const {}
and registery on the provided registry any dialect that this pass can produce.
Passes defined in TableGen can provide this list in the dependentDialects list
field.
2) For dialects, on construction you can register dependent dialects using the
provided MLIRContext: `context.getOrLoadDialect<DialectName>()`
This is useful if a dialect may canonicalize or have interfaces involving
another dialect.
3) For loading IR, dialect that can be in the input file must be explicitly
registered with the context. `MlirOptMain()` is taking an explicit registry for
this purpose. See how the standalone-opt.cpp example is setup:
mlir::DialectRegistry registry;
mlir::registerDialect<mlir::standalone::StandaloneDialect>();
mlir::registerDialect<mlir::StandardOpsDialect>();
Only operations from these two dialects can be in the input file. To include all
of the dialects in MLIR Core, you can populate the registry this way:
mlir::registerAllDialects(registry);
4) For `mlir-translate` callback, as well as frontend, Dialects can be loaded in
the context before emitting the IR: context.getOrLoadDialect<ToyDialect>()
LinalgDistribution options to allow more general distributions.
Changing the signature of the callback to send in the ranges for all
the parallel loops and expect a vector with the Value to use for the
processor-id and number-of-processors for each of the parallel loops.
Differential Revision: https://reviews.llvm.org/D86095
There should be an equivalent std.floor op to std.ceil. This includes
matching lowerings for SPIRV, NVVM, ROCDL, and LLVM.
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D85940
Create a reduction pass that accepts an optimization pass as argument
and only replaces the golden module in the pipeline if the output of the
optimization pass is smaller than the input and still exhibits the
interesting behavior.
Add a -test-pass option to test individual passes in the MLIR Reduce
tool.
Reviewed By: jpienaar
Differential Revision: https://reviews.llvm.org/D84783
This patch adds more op/type conversion support
necessary for `spirv-runner`:
- EntryPoint/ExecutionMode: currently removed since we assume
having only one kernel function in the kernel module.
- StorageBuffer storage class is now supported. We are not
concerned with multithreading so this is fine for now.
- Type conversion enhanced, now regular offsets and strides
for structs and arrays are supported (based on
`VulkanLayoutUtils`).
- Support of `spc.AccessChain` that is modelled with GEP op
in LLVM dialect.
Reviewed By: mravishankar
Differential Revision: https://reviews.llvm.org/D86109
When the operand to the linalg.tensor_reshape op is a splat constant,
the result can be replaced with a splat constant of the same value but
different type.
Differential Revision: https://reviews.llvm.org/D86117
Provide C API for MLIR standard types. Since standard types live under lib/IR
in core MLIR, place the C APIs in the IR library as well (standard ops will go
into a separate library). This also defines a placeholder for affine maps that
are necessary to construct a memref, but are not yet exposed to the C API.
Reviewed By: stellaraccident
Differential Revision: https://reviews.llvm.org/D86094
According to the LLVM Language Reference, 'cmpxchg' accepts integer or pointer
types. Several MLIR tests were using it with floats as it appears possible to
programmatically construct and print such an instruction, but it cannot be
parsed back. Use integers instead.
Depends On D85899
Reviewed By: flaub, rriddle
Differential Revision: https://reviews.llvm.org/D85900
Legacy implementation of the LLVM dialect in MLIR contained an instance of
llvm::Module as it was required to parse LLVM IR types. The access to the data
layout of this module was exposed to the users for convenience, but in practice
this layout has always been the default one obtained by parsing an empty layout
description string. Current implementation of the dialect no longer relies on
wrapping LLVM IR types, but it kept an instance of DataLayout for
compatibility. This effectively forces a single data layout to be used across
all modules in a given MLIR context, which is not desirable. Remove DataLayout
from the LLVM dialect and attach it as a module attribute instead. Since MLIR
does not yet have support for data layouts, use the LLVM DataLayout in string
form with verification inside MLIR. Introduce the layout when converting a
module to the LLVM dialect and keep the default "" description for
compatibility.
This approach should be replaced with a proper MLIR-based data layout when it
becomes available, but provides an immediate solution to compiling modules with
different layouts, e.g. for GPUs.
This removes the need for LLVMDialectImpl, which is also removed.
Depends On D85650
Reviewed By: aartbik
Differential Revision: https://reviews.llvm.org/D85652
This changes the behavior of constructing MLIRContext to no longer load globally registered dialects on construction. Instead Dialects are only loaded explicitly on demand:
- the Parser is lazily loading Dialects in the context as it encounters them during parsing. This is the only purpose for registering dialects and not load them in the context.
- Passes are expected to declare the dialects they will create entity from (Operations, Attributes, or Types), and the PassManager is loading Dialects into the Context when starting a pipeline.
This changes simplifies the configuration of the registration: a compiler only need to load the dialect for the IR it will emit, and the optimizer is self-contained and load the required Dialects. For example in the Toy tutorial, the compiler only needs to load the Toy dialect in the Context, all the others (linalg, affine, std, LLVM, ...) are automatically loaded depending on the optimization pipeline enabled.
Differential Revision: https://reviews.llvm.org/D85622
This changes the behavior of constructing MLIRContext to no longer load globally registered dialects on construction. Instead Dialects are only loaded explicitly on demand:
- the Parser is lazily loading Dialects in the context as it encounters them during parsing. This is the only purpose for registering dialects and not load them in the context.
- Passes are expected to declare the dialects they will create entity from (Operations, Attributes, or Types), and the PassManager is loading Dialects into the Context when starting a pipeline.
This changes simplifies the configuration of the registration: a compiler only need to load the dialect for the IR it will emit, and the optimizer is self-contained and load the required Dialects. For example in the Toy tutorial, the compiler only needs to load the Toy dialect in the Context, all the others (linalg, affine, std, LLVM, ...) are automatically loaded depending on the optimization pipeline enabled.
The convresion of memref cast operaitons from the Standard dialect to the LLVM
dialect has been emitting bitcasts from a struct type to itself. Beyond being
useless, such casts are invalid as bitcast does not operate on aggregate types.
This kept working by accident because LLVM IR bitcast construction API skips
the construction if types are equal before it verifies that the types are
acceptable in a bitcast. Do not emit such bitcasts, the memref cast that only
adds/erases size information is in fact a noop on the current descriptor as it
always contains dynamic values for all sizes.
Reviewed By: pifon2a
Differential Revision: https://reviews.llvm.org/D85899
Masked loading/storing in various forms can be optimized
into simpler memory operations when the mask is all true
or all false. Note that the backend does similar optimizations
but doing this early may expose more opportunities for further
optimizations. This further prepares progressively lowering
transfer read and write into 1-D memory operations.
Reviewed By: ThomasRaoux
Differential Revision: https://reviews.llvm.org/D85769
This exercises the corner case that was fixed in
https://reviews.llvm.org/rG8979a9cdf226066196f1710903d13492e6929563.
The bug can be reproduced when there is a @callee with a custom type argument and @caller has a producer of this argument passed to the @callee.
Example:
func @callee(!test.test_type) -> i32
func @caller() -> i32 {
%arg = "test.type_producer"() : () -> !test.test_type
%out = call @callee(%arg) : (!test.test_type) -> i32
return %out : i32
}
Even though there is a type conversion for !test.test_type, the output IR (before the fix) contained a DialectCastOp:
module {
llvm.func @callee(!llvm.ptr<i8>) -> !llvm.i32
llvm.func @caller() -> !llvm.i32 {
%0 = llvm.mlir.null : !llvm.ptr<i8>
%1 = llvm.mlir.cast %0 : !llvm.ptr<i8> to !test.test_type
%2 = llvm.call @callee(%1) : (!test.test_type) -> !llvm.i32
llvm.return %2 : !llvm.i32
}
}
instead of
module {
llvm.func @callee(!llvm.ptr<i8>) -> !llvm.i32
llvm.func @caller() -> !llvm.i32 {
%0 = llvm.mlir.null : !llvm.ptr<i8>
%1 = llvm.call @callee(%0) : (!llvm.ptr<i8>) -> !llvm.i32
llvm.return %1 : !llvm.i32
}
}
Differential Revision: https://reviews.llvm.org/D85914
-- This commit handles the returnOp in memref map layout normalization.
-- An initial filter is applied on FuncOps which helps us know which functions can be
a suitable candidate for memref normalization which doesn't lead to invalid IR.
-- Handles memref map normalization for external function assuming the external function
is normalizable.
Differential Revision: https://reviews.llvm.org/D85226
Provide printing functions for most IR objects in C API (except Region that
does not have a `print` function, and Module that is expected to be printed as
Operation instead). The printing is based on a callback that is called with
chunks of the string representation and forwarded user-defined data.
Reviewed By: stellaraccident, Jing, mehdi_amini
Differential Revision: https://reviews.llvm.org/D85748
This patch adds the translation of the proc_bind clause in a
parallel operation.
The values that can be specified for the proc_bind clause are
specified in the OMP.td tablegen file in the llvm/Frontend/OpenMP
directory. From this single source of truth enumeration for
proc_bind is generated in llvm and mlir (used in specification of
the parallel Operation in the OpenMP dialect). A function to return
the enum value from the string representation is also generated.
A new header file (DirectiveEmitter.h) containing definitions of
classes directive, clause, clauseval etc is created so that it can
be used in mlir as well.
Reviewers: clementval, jdoerfert, DavidTruby
Differential Revision: https://reviews.llvm.org/D84347
Inital conversion of `spv._address_of` and `spv.globalVariable`.
In SPIR-V, the global returns a pointer, whereas in LLVM dialect
the global holds an actual value. This difference is handled by
`spv._address_of` and `llvm.mlir.addressof`ops that both return
a pointer. Moreover, only current invocation is in conversion's
scope.
Reviewed By: antiagainst, mravishankar
Differential Revision: https://reviews.llvm.org/D84626
Now that LLVM dialect types are implemented directly in the dialect, we can use
MLIR hooks for verifying type construction invariants. Implement the verifiers
and use them in the parser.
Reviewed By: rriddle
Differential Revision: https://reviews.llvm.org/D85663
Linalg to processors.
This changes adds infrastructure to distribute the loops generated in
Linalg to processors at the time of generation. This addresses use
case where the instantiation of loop is done just to distribute
them. The option to distribute is added to TilingOptions for now and
will allow specifying the distribution as a transformation option,
just like tiling and promotion are specified as options.
Differential Revision: https://reviews.llvm.org/D85147
- Fix ODS framework to suppress build methods that infer result types and are
ambiguous with collective variants. This applies to operations with a single variadic
inputs whose result types can be inferred.
- Extended OpBuildGenTest to test these kinds of ops.
Differential Revision: https://reviews.llvm.org/D85060
This diff attempts to resolve the TODO in `getOpIndexSet` (formerly
known as `getInstIndexSet`), which states "Add support to handle IfInsts
surronding `op`".
Major changes in this diff:
1. Overload `getIndexSet`. The overloaded version considers both
`AffineForOp` and `AffineIfOp`.
2. The `getInstIndexSet` is updated accordingly: its name is changed to
`getOpIndexSet` and its implementation is based on a new API `getIVs`
instead of `getLoopIVs`.
3. Add `addAffineIfOpDomain` to `FlatAffineConstraints`, which extracts
new constraints from the integer set of `AffineIfOp` and merges it to
the current constraint system.
4. Update how a `Value` is determined as dim or symbol for
`ValuePositionMap` in `buildDimAndSymbolPositionMaps`.
Differential Revision: https://reviews.llvm.org/D84698
This patch also fixes a minor issue that shape.rank should allow
returning !shape.size. The dialect doc has such an example for
shape.rank.
Differential Revision: https://reviews.llvm.org/D85556
This reverts commit 9f24640b7e.
We hit some dead-locks on thread exit in some configurations: TLS exit handler is taking a lock.
Temporarily reverting this change as we're debugging what is going on.
This revision aims to provide a new API, `checkTilingLegality`, to
verify that the loop tiling result still satisifes the dependence
constraints of the original loop nest.
Previously, there was no check for the validity of tiling. For instance:
```
func @diagonal_dependence() {
%A = alloc() : memref<64x64xf32>
affine.for %i = 0 to 64 {
affine.for %j = 0 to 64 {
%0 = affine.load %A[%j, %i] : memref<64x64xf32>
%1 = affine.load %A[%i, %j - 1] : memref<64x64xf32>
%2 = addf %0, %1 : f32
affine.store %2, %A[%i, %j] : memref<64x64xf32>
}
}
return
}
```
You can find more information about this example from the Section 3.11
of [1].
In general, there are three types of dependences here: two flow
dependences, one in direction `(i, j) = (0, 1)` (notation that depicts a
vector in the 2D iteration space), one in `(i, j) = (1, -1)`; and one
anti dependence in the direction `(-1, 1)`.
Since two of them are along the diagonal in opposite directions, the
default tiling method in `affine`, which tiles the iteration space into
rectangles, will violate the legality condition proposed by Irigoin and
Triolet [2]. [2] implies two tiles cannot depend on each other, while in
the `affine` tiling case, two rectangles along the same diagonal are
indeed dependent, which simply violates the rule.
This diff attempts to put together a validator that checks whether the
rule from [2] is violated or not when applying the default tiling method
in `affine`.
The canonical way to perform such validation is by examining the effect
from adding the constraint from Irigoin and Triolet to the existing
dependence constraints.
Since we already have the prior knowlegde that `affine` tiles in a
hyper-rectangular way, and the resulting tiles will be scheduled in the
same order as their respective loop indices, we can simplify the
solution to just checking whether all dependence components are
non-negative along the tiling dimensions.
We put this algorithm into a new API called `checkTilingLegality` under
`LoopTiling.cpp`. This function iterates every `load`/`store` pair, and
if there is any dependence between them, we get the dependence component
and check whether it has any negative component. This function returns
`failure` if the legality condition is violated.
[1]. Bondhugula, Uday. Effective Automatic parallelization and locality optimization using the Polyhedral model. https://dl.acm.org/doi/book/10.5555/1559029
[2]. Irigoin, F. and Triolet, R. Supernode Partitioning. https://dl.acm.org/doi/10.1145/73560.73588
Differential Revision: https://reviews.llvm.org/D84882
Implement the Reduction Tree Pass framework as part of the MLIR Reduce tool. This is a parametarizable pass that allows for the implementation of custom reductions passes in the tool.
Implement the FunctionReducer class as an example of a Reducer class parameter for the instantiation of a Reduction Tree Pass.
Create a pass pipeline with a Reduction Tree Pass with the FunctionReducer class specified as parameter.
Reviewed By: jpienaar
Differential Revision: https://reviews.llvm.org/D83969
This also beefs up the test coverage:
- Make unranked memref testing consistent with ranked memrefs.
- Add testing for the invalid element type cases.
This is not quite NFC: index types are now allowed in unranked memrefs.
Differential Revision: https://reviews.llvm.org/D85541
This simple patch translates the num_threads and if clauses of the parallel
operation. Also includes test cases.
A minor change was made to parsing of the if clause to parse AnyType and
return the parsed type. Updates to test cases also.
Reviewed by: SouraVX
Differential Revision: https://reviews.llvm.org/D84798
This revision refactors the default definition of the attribute and type `classof` methods to use the TypeID of the concrete class instead of invoking the `kindof` method. The TypeID is already used as part of uniquing, and this allows for removing the need for users to define any of the type casting utilities themselves.
Differential Revision: https://reviews.llvm.org/D85356
This class allows for defining thread local objects that have a set non-static lifetime. This internals of the cache use a static thread_local map between the various different non-static objects and the desired value type. When a non-static object destructs, it simply nulls out the entry in the static map. This will leave an entry in the map, but erase any of the data for the associated value. The current use cases for this are in the MLIRContext, meaning that the number of items in the static map is ~1-2 which aren't particularly costly enough to warrant the complexity of pruning. If a use case arises that requires pruning of the map, the functionality can be added.
This is especially useful in the context of MLIR for implementing thread-local caching of context level objects that would otherwise have very high lock contention. This revision adds a thread local cache in the MLIRContext for attributes, identifiers, and types to reduce some of the locking burden. This led to a speedup of several hundred miliseconds when compiling a conversion pass on a very large mlir module(>300K operations).
Differential Revision: https://reviews.llvm.org/D82597
This change adds initial support needed to generate OpenCL compliant SPIRV.
If Kernel capability is declared then memory model becomes OpenCL.
If Addresses capability is declared then addressing model becomes Physical64.
Additionally for Kernel capability interface variable ABI attributes are not
generated as entry point function is expected to have normal arguments.
Differential Revision: https://reviews.llvm.org/D85196
This revision adds a folding pattern to replace affine.min ops by the actual min value, when it can be determined statically from the strides and bounds of enclosing scf loop .
This matches the type of expressions that Linalg produces during tiling and simplifies boundary checks. For now Linalg depends both on Affine and SCF but they do not depend on each other, so the pattern is added there.
In the future this will move to a more appropriate place when it is determined.
The canonicalization of AffineMinOp operations in the context of enclosing scf.for and scf.parallel proceeds by:
1. building an affine map where uses of the induction variable of a loop
are replaced by `%lb + %step * floordiv(%iv - %lb, %step)` expressions.
2. checking if any of the results of this affine map divides all the other
results (in which case it is also guaranteed to be the min).
3. replacing the AffineMinOp by the result of (2).
The algorithm is functional in simple parametric tiling cases by using semi-affine maps. However simplifications of such semi-affine maps are not yet available and the canonicalization does not succeed yet.
Differential Revision: https://reviews.llvm.org/D82009
Using a shuffle for the last recursive step in progressive lowering not only
results in much more compact IR, but also more efficient code (since the
backend is no longer confused on subvector aliasing for longer vectors).
E.g. the following
%f = vector.shape_cast %v0: vector<1024xf32> to vector<32x32xf32>
yields much better x86-64 code that runs 3x faster than the original.
Reviewed By: bkramer, nicolasvasilache
Differential Revision: https://reviews.llvm.org/D85482
This patch moves the registration to a method in the MLIRContext: getOrCreateDialect<ConcreteDialect>()
This method requires dialect to provide a static getDialectNamespace()
and store a TypeID on the Dialect itself, which allows to lazyily
create a dialect when not yet loaded in the context.
As a side effect, it means that duplicated registration of the same
dialect is not an issue anymore.
To limit the boilerplate, TableGen dialect generation is modified to
emit the constructor entirely and invoke separately a "init()" method
that the user implements.
Differential Revision: https://reviews.llvm.org/D85495
This new pattern mixes vector.transpose and direct lowering to vector.reduce.
This allows more progressive lowering than immediately going to insert/extract and
composes more nicely with other canonicalizations.
This has 2 use cases:
1. for very wide vectors the generated IR may be much smaller
2. when we have a custom lowering for transpose ops we can target it directly
rather than rely LLVM
Differential Revision: https://reviews.llvm.org/D85428
When any of the memrefs in a structured linalg op has a zero dimension, it becomes dead.
This is consistent with the fact that linalg ops deduce their loop bounds from their operands.
Note however that this is not the case for the `tensor<0xelt_type>` which is a special convention
that must be lowered away into either `memref<elt_type>` or just `elt_type` before this
canonicalization can kick in.
Differential Revision: https://reviews.llvm.org/D85413
The intrinsics were already supported and vector.transfer_read/write lowered
direclty into these operations. By providing them as individual ops, however,
clients can used them directly, and it opens up progressively lowering transfer
operations at higher levels (rather than direct lowering to LLVM IR as done now).
Reviewed By: bkramer
Differential Revision: https://reviews.llvm.org/D85357
Previous type model in the LLVM dialect did not support identified structure
types properly and therefore could use stateless translations implemented as
free functions. The new model supports identified structs and must keep track
of the identified structure types present in the target context (LLVMContext or
MLIRContext) to avoid creating duplicate structs due to LLVM's type
auto-renaming. Expose the stateful type translation classes and use them during
translation, storing the state as part of ModuleTranslation.
Drop the test type translation mechanism that is no longer necessary and update
the tests to exercise type translation as part of the main translation flow.
Update the code in vector-to-LLVM dialect conversion that relied on stateless
translation to use the new class in a stateless manner.
Reviewed By: rriddle
Differential Revision: https://reviews.llvm.org/D85297
Per Vulkan's SPIR-V environment spec: "While the OpSRem and OpSMod
instructions are supported by the Vulkan environment, they require
non-negative values and thus do not enable additional functionality
beyond what OpUMod provides."
The `getOffsetForBitwidth` function is used for lowering std.load
and std.store, whose indices are of `index` type and cannot be
negative. So we should be okay to use spv.UMod directly here to
be exact. Also made the comment explicit about the assumption.
Differential Revision: https://reviews.llvm.org/D83714
If Int16 is not available, 16-bit integers inside Workgroup storage
class should be emulated via 32-bit integers. This was previously
broken because the capability querying logic was incorrectly
intercepting all storage classes where it meant to only handle
interface storage classes. Adjusted where we return to fix this.
Differential Revision: https://reviews.llvm.org/D85308
Introduce an initial version of C API for MLIR core IR components: Value, Type,
Attribute, Operation, Region, Block, Location. These APIs allow for both
inspection and creation of the IR in the generic form and intended for wrapping
in high-level library- and language-specific constructs. At this point, there
is no stability guarantee provided for the API.
Reviewed By: stellaraccident, lattner
Differential Revision: https://reviews.llvm.org/D83310
This dialect was introduced during the bring-up of the new LLVM dialect type
system for testing purposes. The main LLVM dialect now uses the new type system
and the test dialect is no longer necessary, so remove it.
Reviewed By: rriddle
Differential Revision: https://reviews.llvm.org/D85224
Handle the case where the ViewOp takes in a memref that has
an memory space.
Reviewed By: ftynse, bondhugula, nicolasvasilache
Differential Revision: https://reviews.llvm.org/D85048
This patch introduces a conversion of `spv.loop` to LLVM dialect.
Similarly to `spv.selection`, op's control attributes are not mapped
to LLVM yet and therefore the conversion fails if the loop control is
not `None`. Also, all blocks within the loop should be reachable in
order for conversion to succeed.
Reviewed By: mravishankar
Differential Revision: https://reviews.llvm.org/D84245
Always define a remapping for the memref replacement (`indexRemap`)
with the proper number of inputs, including all the `outerIVs`, so that
the number of inputs and the operands provided for the map don't mismatch.
Reviewed By: bondhugula, andydavis1
Differential Revision: https://reviews.llvm.org/D85177
Introduces the expand and compress operations to the Vector dialect
(important memory operations for sparse computations), together
with a first reference implementation that lowers to the LLVM IR
dialect to enable running on CPU (and other targets that support
the corresponding LLVM IR intrinsics).
Reviewed By: reidtatge
Differential Revision: https://reviews.llvm.org/D84888
Simplify semi-affine expression for the operations like ceildiv,
floordiv and modulo by any given symbol by checking divisibilty by that
symbol.
Some properties used in simplification are:
1) Commutative property of the floordiv and ceildiv:
((expr1 floordiv expr2) floordiv expr3 ) = ((expr1 floordiv expr3) floordiv expr2)
((expr1 ceildiv expr2) ceildiv expr3 ) = ((expr1 ceildiv expr3) ceildiv expr2)
While simplification if operations are different no simplification is
possible as there is no property that simplify expressions like these:
((expr1 ceildiv expr2) floordiv expr3) or ((expr1 floordiv expr2)
ceildiv expr3).
2) If both expr1 and expr2 are divisible by the expr3 then:
(expr1 % expr2) / expr3 = ((expr1 / expr3) % (expr2 / expr3))
where / is divide symbol.
3) If expr1 is divisible by expr2 then expr1 % expr2 = 0.
Signed-off-by: Yash Jain <yash.jain@polymagelabs.com>
Differential Revision: https://reviews.llvm.org/D84920
This revision adds a transformation and a pattern that rewrites a "maybe masked" `vector.transfer_read %view[...], %pad `into a pattern resembling:
```
%1:3 = scf.if (%inBounds) {
scf.yield %view : memref<A...>, index, index
} else {
%2 = linalg.fill(%extra_alloc, %pad)
%3 = subview %view [...][...][...]
linalg.copy(%3, %alloc)
memref_cast %extra_alloc: memref<B...> to memref<A...>
scf.yield %4 : memref<A...>, index, index
}
%res= vector.transfer_read %1#0[%1#1, %1#2] {masked = [false ... false]}
```
where `extra_alloc` is a top of the function alloca'ed buffer of one vector.
This rewrite makes it possible to realize the "always full tile" abstraction where vector.transfer_read operations are guaranteed to read from a padded full buffer.
The extra work only occurs on the boundary tiles.
A new first-party modeling for LLVM IR types in the LLVM dialect has been
developed in parallel to the existing modeling based on wrapping LLVM `Type *`
instances. It resolves the long-standing problem of modeling identified
structure types, including recursive structures, and enables future removal of
LLVMContext and related locking mechanisms from LLVMDialect.
This commit only switches the modeling by (a) renaming LLVMTypeNew to LLVMType,
(b) removing the old implementaiton of LLVMType, and (c) updating the tests. It
is intentionally minimal. Separate commits will remove the infrastructure built
for the transition and update API uses where appropriate.
Depends On D85020
Reviewed By: rriddle
Differential Revision: https://reviews.llvm.org/D85021
With new LLVM dialect type modeling, the dialect types no longer wrap LLVM IR
types. Therefore, they need to be translated to and from LLVM IR during export
and import. Introduce the relevant functionality for translating types. It is
currently exercised by an ad-hoc type translation roundtripping test that will
be subsumed by the actual translation test when the type system transition is
complete.
Depends On D84339
Reviewed By: herhut
Differential Revision: https://reviews.llvm.org/D85019
This is a first patch that sweeps over tests to fix
indentation (tabs to spaces). It also adds label checks and
removes redundant matching of `%{{.*}} = `.
The following tests have been fixed:
- arithmetic-ops-to-llvm
- bitwise-ops-to-llvm
- cast-ops-to-llvm
- comparison-ops-to-llvm
- logical-ops-to-llvm (renamed to match the rest)
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D85181
Unit attributes are given meaning by their existence, and thus have no meaningful value beyond "is it present". As such, in the format of an operation unit attributes are generally used to guard the printing of other elements and aren't generally printed themselves; as the presence of the group when parsing means that the unit attribute should be added. This revision adds support to the declarative format for eliding unit attributes in situations where they anchor an optional group, but aren't the first element.
For example,
```
let assemblyFormat = "(`is_optional` $unit_attr^)? attr-dict";
```
would print `foo.op is_optional` when $unit_attr is present, instead of the current `foo.op is_optional unit`.
Differential Revision: https://reviews.llvm.org/D84577
This revision adds a transformation and a pattern that rewrites a "maybe masked" `vector.transfer_read %view[...], %pad `into a pattern resembling:
```
%1:3 = scf.if (%inBounds) {
scf.yield %view : memref<A...>, index, index
} else {
%2 = vector.transfer_read %view[...], %pad : memref<A...>, vector<...>
%3 = vector.type_cast %extra_alloc : memref<...> to
memref<vector<...>> store %2, %3[] : memref<vector<...>> %4 =
memref_cast %extra_alloc: memref<B...> to memref<A...> scf.yield %4 :
memref<A...>, index, index
}
%res= vector.transfer_read %1#0[%1#1, %1#2] {masked = [false ... false]}
```
where `extra_alloc` is a top of the function alloca'ed buffer of one vector.
This rewrite makes it possible to realize the "always full tile" abstraction where vector.transfer_read operations are guaranteed to read from a padded full buffer.
The extra work only occurs on the boundary tiles.
Differential Revision: https://reviews.llvm.org/D84631
This reverts commit 35b65be041.
Build is broken with -DBUILD_SHARED_LIBS=ON with some undefined
references like:
VectorTransforms.cpp:(.text._ZN4llvm12function_refIFvllEE11callback_fnIZL24createScopedInBoundsCondN4mlir25VectorTransferOpInterfaceEE3$_8EEvlll+0xa5): undefined reference to `mlir::edsc::op::operator+(mlir::Value, mlir::Value)'
The current modeling of LLVM IR types in MLIR is based on the LLVMType class
that wraps a raw `llvm::Type *` and delegates uniquing, printing and parsing to
LLVM itself. This model makes thread-safe type manipulation hard and is being
progressively replaced with a cleaner MLIR model that replicates the type
system. Introduce a set of classes reflecting the LLVM IR type system in MLIR
instead of wrapping the existing types. These are currently introduced as
separate classes without affecting the dialect flow, and are exercised through
a test dialect. Once feature parity is reached, the old implementation will be
gradually substituted with the new one.
Depends On D84171
Reviewed By: rriddle
Differential Revision: https://reviews.llvm.org/D84339
This revision adds a transformation and a pattern that rewrites a "maybe masked" `vector.transfer_read %view[...], %pad `into a pattern resembling:
```
%1:3 = scf.if (%inBounds) {
scf.yield %view : memref<A...>, index, index
} else {
%2 = vector.transfer_read %view[...], %pad : memref<A...>, vector<...>
%3 = vector.type_cast %extra_alloc : memref<...> to
memref<vector<...>> store %2, %3[] : memref<vector<...>> %4 =
memref_cast %extra_alloc: memref<B...> to memref<A...> scf.yield %4 :
memref<A...>, index, index
}
%res= vector.transfer_read %1#0[%1#1, %1#2] {masked = [false ... false]}
```
where `extra_alloc` is a top of the function alloca'ed buffer of one vector.
This rewrite makes it possible to realize the "always full tile" abstraction where vector.transfer_read operations are guaranteed to read from a padded full buffer.
The extra work only occurs on the boundary tiles.
Differential Revision: https://reviews.llvm.org/D84631
This patch handles loopControl and selectionControl in parsing and
printing. In order to reuse the functionality, and avoid handling cases when
`{` of the region is parsed as a dictionary attribute, `control` keyword was
introduced.`None` is a default control attribute. This functionality can be
later extended to `spv.func`.
Also, loopControl and selectionControl can now be (de)serialized.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D84175
This is an operation that can returns a new ValueShape with a different shape. Useful for composing shape function calls and reusing existing shape transfer functions.
Just adding the op in this change.
Differential Revision: https://reviews.llvm.org/D84217
This change allow CooperativeMatrix Load/Store operations to use pointer type
that may not match the matrix element type. This allow us to declare buffer
with a larger type size than the matrix element type. This follows SPIR-V spec
and this is needed to be able to use cooperative matrix in combination with
shared local memory efficiently.
Differential Revision: https://reviews.llvm.org/D84993
In a context in which `shape.broadcast` is known not to produce an error value,
we want it to operate solely on extent tensors. The operation's behavior is
then undefined in the error case as the result type cannot hold this value.
Differential Revision: https://reviews.llvm.org/D84933
Replaced definition of named ND ConvOps with tensor comprehension
syntax which reduces boilerplate code significantly. Furthermore,
new ops to support TF convolutions added (without strides and dilations).
Reviewed By: nicolasvasilache
Differential Revision: https://reviews.llvm.org/D84628
Now that we can have a memref of index type, we no longer need to materialize shapes in i64 and then index_cast.
Differential Revision: https://reviews.llvm.org/D84938
-- Introduces a pass that normalizes the affine layout maps to the identity layout map both within and across functions by rewriting function arguments and call operands where necessary.
-- Memref normalization is now implemented entirely in the module pass '-normalize-memrefs' and the limited intra-procedural version has been removed from '-simplify-affine-structures'.
-- Run using -normalize-memrefs.
-- Return ops are not handled and would be handled in the subsequent revisions.
Signed-off-by: Abhishek Varma <abhishek.varma@polymagelabs.com>
Differential Revision: https://reviews.llvm.org/D84490
When lowering to the standard dialect, we currently support only the extent
tensor variant of the shape.rank operation. This change lets the conversion
pattern fail in a well-defined manner.
Differential Revision: https://reviews.llvm.org/D84852
This patch introduces new intrinsics in LLVM dialect:
- `llvm.intr.floor`
- `llvm.intr.maxnum`
- `llvm.intr.minnum`
- `llvm.intr.smax`
- `llvm.intr.smin`
These intrinsics correspond to SPIR-V ops from GLSL
extended instruction set (`spv.GLSL.Floor`, `spv.GLSL.FMax`,
`spv.GLSL.FMin`, `spv.GLSL.SMax` and `spv.GLSL.SMin`
respectively). Also conversion patterns for them were added.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D84661
This is a second patch on conversion of GLSL ops to LLVM dialect.
It introduces patterns to convert `spv.InverseSqrt` and `spv.Tanh`.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D84633
This is the first patch that adds support for GLSL extended
instruction set ops. These are direct conversions, apart from `spv.Tan`
that is lowered to `sin() / cos()`.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D84627
This commit is part of a greater project which aims to add
full end-to-end support for convolutions inside mlir. The
reason behind having conv ops for each rank rather than
having one generic ConvOp is to enable better optimizations
for every N-D case which reflects memory layout of input/kernel
buffers better and simplifies code as well. We expect plain linalg.conv
to be progressively retired.
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D83879
The lowering does not support all types for its source operations. This change
makes the patterns fail in a well-defined manner.
Differential Revision: https://reviews.llvm.org/D84443
Operating on indices and extent tensors directly, the type conversion is no
longer needed for the supported cases.
Differential Revision: https://reviews.llvm.org/D84442
This adds conversions for const_size and to_extent_tensor. Also, cast-like operations are now folded away if the source and target types are the same.
Differential Revision: https://reviews.llvm.org/D84745
Conversion of `spv.BranchConditional` now supports branch weights
that are mapped to weights vector in `llvm.cond_br`.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D84657
Added a check for 'Function' storage class in `spv.globalVariable`
verifier since it only can be used with `spv.Variable`.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D84731
This patch adds support of Volatile and Nontemporal
memory accesses to `spv.Load` and `spv.Store`. These attributes are
modelled with a `volatile` and `nontemporal` flags.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D84739
- Initiate the unit test with a test that tests variants of build() methods
generated for ops with variadic operands and results.
- The intent is to migrate unit .td tests in mlir/test/mlir-tblgen that check for
generated C++ code to these unit tests which test both that the generated code
compiles and also is functionally correct.
Differential Revision: https://reviews.llvm.org/D84074
functions.
This allows using command line flags to lowere from GPU to SPIR-V. The
pass added is only for testing/example purposes. Most uses cases will
need more fine-grained control on setting workgroup sizes for kernel
functions.
Differential Revision: https://reviews.llvm.org/D84619
The MemRefDataFlow pass does store to load forwarding
only for affine store/loads. This patch updates the pass
to use affine read/write interface which enables vector
forwarding.
Reviewed By: dcaballe, bondhugula, ftynse
Differential Revision: https://reviews.llvm.org/D84302
For the purpose of vector transforms, the Tablegen-based infra is subsumed by simple C++ pattern application. Deprecate declarative transforms whose complexity does not pay for itself.
Differential Revision: https://reviews.llvm.org/D84753
Do not return error code, instead return created resource handles or void. Error reporting is done by the library function.
Reviewed By: herhut
Differential Revision: https://reviews.llvm.org/D84660
The current transformation to shape.reduce does not support tensor values.
This adds the required changes to make that work, including fixing the builder
for shape.reduce.
Differential Revision: https://reviews.llvm.org/D84744
- replace DotOp, now that DRR rules have been dropped.
- Capture arguments mismatch in the parser. The number of parsed arguments must
equal the number of expected arguments.
Reviewed By: ftynse, nicolasvasilache
Differential Revision: https://reviews.llvm.org/D82952
The LowerAffine psas was a FunctionPass only for legacy
reasons. Making this Op-agnostic allows it to be used from command
line when affine expressions are within operations other than
`std.func`.
Differential Revision: https://reviews.llvm.org/D84590
linalg.indexed_generic (consumer) with tensor arguments.
The implementation of fusing std.constant producer with a
linalg.indexed_generic consumer was already in place. It is exposed
with this change. Also cleaning up some of the patterns that implement
the fusion to not be templated, thereby avoiding lot of conditional
checks for calling the right instantiation.
Differential Revision: https://reviews.llvm.org/D84566
Introduce support for mutable storage in the StorageUniquer infrastructure.
This makes MLIR have key-value storage instead of just uniqued key storage. A
storage instance now contains a unique immutable key and a mutable value, both
stored in the arena allocator that belongs to the context. This is a
preconditio for supporting recursive types that require delayed initialization,
in particular LLVM structure types. The functionality is exercised in the test
pass with trivial self-recursive type. So far, recursive types can only be
printed in parsed in a closed type system. Removing this restriction is left
for future work.
Differential Revision: https://reviews.llvm.org/D84171
This patch introduces 2 new optional attributes to `llvm.load`
and `llvm.store` ops: `volatile` and `nontemporal`. These attributes
are translated into proper LLVM as a `volatile` marker and a metadata node
respectively. They are also helpful with SPIR-V to LLVM dialect conversion
since they are the mappings for `Volatile` and `NonTemporal` Memory Operands.
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D84396
This diff provides a concrete test case for the error that will be raised when the iteration space is non hyper-rectangular.
The corresponding emission method for this error message has been changed as well.
Differential Revision: https://reviews.llvm.org/D84531
Previous changes generalized some of the operands and results. Complete
a larger group of those to simplify progressive lowering. Also update
some of the declarative asm form due to generalization. Tried to keep it
mostly mechanical.
Based on https://reviews.llvm.org/D84439 but less restrictive, else we
don't allow shape_of to be able to produce a ranked output and doesn't
allow for iterative refinement here. We can consider making it more
restrictive later.
This patch introduces conversion pattern for `spv.Store` and `spv.Load`.
Only op with `Function` Storage Class is supported at the moment
because `spv.GlobalVariable` has not been introduced yet. If the op
has memory access attribute, then there are the following cases.
If the access is `Aligned`, add alignment to the op builder. Otherwise
the conversion fails as other cases are not supported yet because they
require additional attributes for `llvm.store`/`llvm.load` ops: e.g.
`volatile` and `nontemporal`.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D84236
The patch introduces the conversion pattern for function-level
`spv.Variable`. It is modelled as `llvm.alloca` op. If initialized, then
additional store instruction is used. Note that there is no initialization
for arrays and structs since constants of these types are not supported in
LLVM dialect yet. Also, at the moment initialisation is only possible via
`spv.constant` (since `spv.GlobalVariable` conversion is not implemented
yet).
The input code has some scoping is not taken into account and will be
addressed in a different patch.
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D84224
This concerns `from/to_extent_tensor`, `size_to_index`, `index_to_size`, and
`const_size` conversion patterns. The new lowering will work directly on indices
and extent tensors. The shape and size values will allow for error values but
are not yet supported by the dialect conversion.
Differential Revision: https://reviews.llvm.org/D84436
The operation `shape.shape_of` now returns an extent tensor `tensor<?xindex>` in
cases when no error are possible. All consuming operation will eventually accept
both, shapes and extent tensors.
Differential Revision: https://reviews.llvm.org/D84160
The default lowering of `assert` calls `abort` in case the assertion is
violated. The failure message is ignored but should be used by custom lowerings
that can assume more about their environment.
Differential Revision: https://reviews.llvm.org/D83886
The operation `shape.const_shape` was used for constants of type shape only.
We can now also use it to create constant extent tensors.
Differential Revision: https://reviews.llvm.org/D84157
This patch introduces branch weights metadata to `llvm.cond_br` op in
LLVM Dialect. It is modelled as optional `ElementsAttr`, for example:
```
llvm.cond_br %cond weights(dense<[1, 3]> : vector<2xi32>), ^bb1, ^bb2
```
When exporting to proper LLVM, this attribute is transformed into metadata
node. The test for metadata creation is added to `../Target/llvmir.mlir`.
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D83658
This revision adds support for much deeper type conversion integration into the conversion process, and enables auto-generating cast operations when necessary. Type conversions are now largely automatically managed by the conversion infra when using a ConversionPattern with a provided TypeConverter. This removes the need for patterns to do type cast wrapping themselves and moves the burden to the infra. This makes it much easier to perform partial lowerings when type conversions are involved, as any lingering type conversions will be automatically resolved/legalized by the conversion infra.
To support this new integration, a few changes have been made to the type materialization API on TypeConverter. Materialization has been split into three separate categories:
* Argument Materialization: This type of materialization is used when converting the type of block arguments when calling `convertRegionTypes`. This is useful for contextually inserting additional conversion operations when converting a block argument type, such as when converting the types of a function signature.
* Source Materialization: This type of materialization is used to convert a legal type of the converter into a non-legal type, generally a source type. This may be called when uses of a non-legal type persist after the conversion process has finished.
* Target Materialization: This type of materialization is used to convert a non-legal, or source, type into a legal, or target, type. This type of materialization is used when applying a pattern on an operation, but the types of the operands have not yet been converted.
Differential Revision: https://reviews.llvm.org/D82831
The `makeTiledViews` did not use the sizes of the tiled views based on
the result of the loop bound inference computation. This manifested as
an error in computing tile sizes with convolution where not all the
result expression of concatenated affine maps are simple
AffineDimExpr.
Differential Revision: https://reviews.llvm.org/D84366
linalg.conv does not support memrefs with rank smaller than 3 as stated here:
https://www.tensorflow.org/versions/r2.0/api_docs/python/tf/nn/convolution
However it does not verify it and thus crashes with "LLVM ERROR: out of memory"
error for 1D case and "nWin > 0 && "expected at least one window dimension"" assertion
for 2D case. This commit adds check for that in the verification method.
Differential Revision: https://reviews.llvm.org/D84317
Loop bound inference is right now very limited as it supports only permutation maps and thus
it is impossible to implement convolution with linalg.generic as it requires more advanced
loop bound inference. This commits solves it for the convolution case.
Depends On D83158
Differential Revision: https://reviews.llvm.org/D83191
The underlying infrastructure supports this already, just add the
pattern matching for linalg.generic.
Differential Revision: https://reviews.llvm.org/D84335
AllocOp is updated in normalizeMemref(AllocOp allocOp), but, when the
AllocOp has `alignment` attribute, it was ignored and updated AllocOp
does not have `alignment` attribute. This patch fixes it.
Differential Revision: https://reviews.llvm.org/D83656
Introduces the scatter/gather operations to the Vector dialect
(important memory operations for sparse computations), together
with a first reference implementation that lowers to the LLVM IR
dialect to enable running on CPU (and other targets that support
the corresponding LLVM IR intrinsics).
The operations can be used directly where applicable, or can be used
during progressively lowering to bring other memory operations closer to
hardware ISA support for a gather/scatter. The semantics of the operation
closely correspond to those of the corresponding llvm intrinsics.
Note that the operation allows for a dynamic index vector (which is
important for sparse computations). However, this first reference
lowering implementation "serializes" the address computation when
base + index_vector is converted to a vector of pointers. Exploring
how to use SIMD properly during these step is TBD. More general
memrefs and idiomatic versions of striding are also TBD.
Reviewed By: arpith-jacob
Differential Revision: https://reviews.llvm.org/D84039
This patch introduces conversion pattern for `spv.selection` op.
The conversion can only be applied to selection with all blocks being
reachable. Moreover, selection with control attributes "Flatten" and
"DontFlatten" is not supported.
Since the `PatternRewriter` hook for block merging has not been implemented
for `ConversionPatternRewriter`, merge and continue blocks are kept
separately.
Reviewed By: antiagainst, ftynse
Differential Revision: https://reviews.llvm.org/D83860
This patch introduces conversion for `spv.Branch` and `spv.BranchConditional`
ops. Branch weigths for `spv.BranchConditional` are not supported at the
moment, and conversion in this case fails.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D83784
This commit adds functionality needed for implementation of convolutions with
linalg.generic op. Since linalg.generic right now expects indexing maps to be
just permutations, offset indexing needed in convolutions is not possible.
Therefore in this commit we address the issue by adding support for symbols inside
indexing maps which enables more advanced indexing. The upcoming commit will
solve the problem of computing loop bounds from such maps.
Differential Revision: https://reviews.llvm.org/D83158
Linkage support is already present in the LLVM dialect, and is being translated
for globals other than functions. Translation support has been missing for
functions because their conversion goes through a different code path than
other globals.
Differential Revision: https://reviews.llvm.org/D84149
Summary: The logic was conservative but inverted: cases that should remain unmasked became 1-D masked.
Differential Revision: https://reviews.llvm.org/D84051
Summary: Vector contract patterns were only parameterized by a `vectorTransformsOptions`. As a result, even if an mlir file was containing several occurrences of `vector.contract`, all of them would be lowered in the same way. More granularity might be required . This Diff adds a `constraint` argument to each of these patterns which allows the user to specify with more precision on which `vector.contract` should each of the lowering apply.
Differential Revision: https://reviews.llvm.org/D83960
- Added more default values for `attributes` parameter for 2 more build methods
- Extend the op-decls.td unit test to test these build methods.
Differential Revision: https://reviews.llvm.org/D83839
Lower `shape.shape_eq` to the `scf` (and `std`) dialect. For now, this lowering
is limited to extent tensor operands.
Differential Revision: https://reviews.llvm.org/D82530
To make it clear when shape error values cannot occur the shape operations can
operate on extent tensors. This change updates the lowering for `shape.reduce`
accordingly.
Differential Revision: https://reviews.llvm.org/D83944
Some dialects have semantics which is not well represented by common
SSA structures with dominance constraints. This patch allows
operations to declare the 'kind' of their contained regions.
Currently, two kinds are allowed: "SSACFG" and "Graph". The only
difference between them at the moment is that SSACFG regions are
required to have dominance, while Graph regions are not required to
have dominance. The intention is that this Interface would be
generated by ODS for existing operations, although this has not yet
been implemented. Presumably, if someone were interested in code
generation, we might also have a "CFG" dialect, which defines control
flow, but does not require SSA.
The new behavior is mostly identical to the previous behavior, since
registered operations without a RegionKindInterface are assumed to
contain SSACFG regions. However, the behavior has changed for
unregistered operations. Previously, these were checked for
dominance, however the new behavior allows dominance violations, in
order to allow the processing of unregistered dialects with Graph
regions. One implication of this is that regions in unregistered
operations with more than one op are no longer CSE'd (since it
requires dominance info).
I've also reorganized the LangRef documentation to remove assertions
about "sequential execution", "SSA Values", and "Dominance". Instead,
the core IR is simply "ordered" (i.e. totally ordered) and consists of
"Values". I've also clarified some things about how control flow
passes between blocks in an SSACFG region. Control Flow must enter a
region at the entry block and follow terminator operation successors
or be returned to the containing op. Graph regions do not define a
notion of control flow.
see discussion here:
https://llvm.discourse.group/t/rfc-allowing-dialects-to-relax-the-ssa-dominance-condition/833/53
Differential Revision: https://reviews.llvm.org/D80358
- Add function `verifyTypes` that Op's can call to do type checking verification
along the control flow edges described the Op's RegionBranchOpInterface.
- We cannot rely on the verify methods on the OpInterface because the interface
functions assume valid Ops, so they may crash if invoked on unverified Ops.
(For example, scf.for getSuccessorRegions() calls getRegionIterArgs(), which
dereferences getBody() block. If the scf.for is invalid with no body, this
can lead to a segfault). `verifyTypes` can be called post op-verification to
avoid this.
Differential Revision: https://reviews.llvm.org/D82829
This folds shape.broadcast where at least one operand is a scalar to the
other operand.
Also add an assemblyFormat for shape.broadcast and shape.concat.
Differential Revision: https://reviews.llvm.org/D83854
Add `shape.shape_eq` operation to the shape dialect.
The operation allows to test shapes and extent tensors for equality.
Differential Revision: https://reviews.llvm.org/D82528
In 2b3c505, the pointer arguments for the matrix load and store
intrinsics was changed to always be the element type of the vector
argument.
This patch updates the MatrixBuilder to not add the pointer type to the
overloaded types and adjusts the clang/mlir tests.
This should fix a few build failures on GreenDragon, including
http://green.lab.llvm.org/green/job/test-suite-verify-machineinstrs-x86_64-O0-g/7891/
Summary:
linalg.copy + linalg.fill can be used to create a padded local buffer.
The `masked` attribute is only valid on this padded buffer.
When forwarding to vector.transfer ops, the attribute must be reset
conservatively.
Differential Revision: https://reviews.llvm.org/D83782
This adds a `parseOptionalAttribute` method to the OpAsmParser that allows for parsing optional attributes, in a similar fashion to how optional types are parsed. This also enables the use of attribute values as the first element of an assembly format optional group.
Differential Revision: https://reviews.llvm.org/D83712
Up until now, there has been an implicit agreement that when an operation is marked as
"erased" all uses of that operation's results are guaranteed to be removed during conversion. How this works in practice is that there is either an assert/crash/asan failure/etc. This revision adds support for properly detecting when an erased operation has dangling users, emits and error and fails the conversion.
Differential Revision: https://reviews.llvm.org/D82830
- Arguments of the first block of a region are considered region arguments.
- Add API on Region class to deal with these arguments directly instead of
using the front() block.
- Changed several instances of existing code that can use this API
- Fixes https://bugs.llvm.org/show_bug.cgi?id=46535
Differential Revision: https://reviews.llvm.org/D83599
This patch introduces lowering of the OpenMP parallel operation to LLVM
IR using the OpenMPIRBuilder.
Functions topologicalSort and connectPhiNodes are generalised so that
they work with operations also. connectPhiNodes is also made static.
Lowering works for a parallel region with multiple blocks. Clauses and
arguments of the OpenMP operation are not handled.
Reviewed By: rriddle, anchu-rajendran
Differential Revision: https://reviews.llvm.org/D81660
Summary: The native alignment may generally not be used when lowering a vector.transfer to the underlying load/store operation. This revision fixes the unmasked load/store alignment to match that of the masked path.
Differential Revision: https://reviews.llvm.org/D83684
- Provide default value for `ArrayRef<NamedAttribute> attributes` parameter of
the collective params build method.
- Change the `genSeparateArgParamBuilder` function to not generate build methods
that may be ambiguous with the new collective params build method.
- This change should help eliminate passing empty NamedAttribue ArrayRef when the
collective params build method is used
- Extend op-decl.td unit test to make sure the ambiguous build methods are not
generated.
Differential Revision: https://reviews.llvm.org/D83517
Per the Vulkan's SPIR-V environment spec, "for the OpSRem and OpSMod
instructions, if either operand is negative the result is undefined."
So we cannot directly use spv.SRem/spv.SMod if either operand can be
negative. Emulate it via spv.UMod.
Because the emulation uses spv.SNegate, this commit also defines
spv.SNegate.
Differential Revision: https://reviews.llvm.org/D83679
Summary:
These are semantically equivalent, but fmuladd allows decaying the op
into fmul+fadd if there is no fma instruction available. llvm.fma lowers
to scalar calls to libm fmaf, which is a lot slower.
Reviewers: nicolasvasilache, aartbik, ftynse
Subscribers: mehdi_amini, rriddle, jpienaar, shauheen, antiagainst, arpith-jacob, mgester, lucyrfox, liufengdb, stephenneuendorffer, Joonsoo, grosul1, Kayjukh, jurahul, msifontes
Tags: #mlir
Differential Revision: https://reviews.llvm.org/D83666
Introduce pass to convert parallel affine.for op into 1-D affine.parallel op.
Run using --affine-parallelize. Removes test-detect-parallel: pass for checking
parallel affine.for ops.
Signed-off-by: Yash Jain <yash.jain@polymagelabs.com>
Differential Revision: https://reviews.llvm.org/D83193
- Create a pass that generates bugs based on trivially defined behavior for the purpose of testing the MLIR Reduce Tool.
- Implement the functionality inside the pass to crash mlir-opt in the presence of an operation with the name "crashOp".
- Register the pass as a test pass in the mlir-opt tool.
Reviewed by: jpienaar
Differential Revision: https://reviews.llvm.org/D83422
Improve the logic deciding if it is safe to hoist vector transfer read/write
out of the loop. Change the logic to prevent hoisting operations if there are
any unknown access to the memref in the loop no matter where the operation is.
For other transfer read/write in the loop check if we can prove that they
access disjoint memory and ignore them in this case.
Differential Revision: https://reviews.llvm.org/D83538
This revision folds vector.transfer operations by updating the `masked` bool array attribute when more unmasked dimensions can be discovered.
Differential revision: https://reviews.llvm.org/D83586
We temporarily had separate OUTER lowering (for matmat flavors) and
AXPY lowering (for matvec flavors). With the new generalized
"vector.outerproduct" semantics, these cases can be merged into
a single lowering method. This refactoring will simplify future
decisions on cost models and lowering heuristics.
Reviewed By: nicolasvasilache
Differential Revision: https://reviews.llvm.org/D83585
This specialization allows sharing more code where an AXPY follows naturally
in cases where an OUTERPRODUCT on a scalar would be generated.
Reviewed By: nicolasvasilache
Differential Revision: https://reviews.llvm.org/D83453
TransposeOp are often followed by ExtractOp.
In certain cases however, it is unnecessary (and even detrimental) to lower a TransposeOp to either a flat transpose (llvm.matrix intrinsics) or to unrolled scalar insert / extract chains.
Providing foldings of ExtractOp mitigates some of the unnecessary complexity.
Differential revision: https://reviews.llvm.org/D83487
This revision adds support for vectorizing named and generic contraction ops to vector.contract. Cases in which the memref is 0-D are special cased to emit std.load/std.store instead of vector.transfer. Relevant tests are added.
Differential revision: https://reviews.llvm.org/D83307
This patch introduces type conversion for SPIR-V structs. Since
handling offset case requires thorough testing, it was left out
for now. Hence, only structs with no offset are currently
supported. Also, structs containing member decorations cannot
be translated.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D83403
This commit augments spv.CopyMemory's implementation to support 2 memory
access operands. Hence, more closely following the spec. The following
changes are introduces:
- Customize logic for spv.CopyMemory serialization and deserialization.
- Add 2 additional attributes for source memory access operand.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D83241
Summary:
* Native '_mlir' extension module.
* Python mlir/__init__.py trampoline module.
* Lit test that checks a message.
* Uses some cmake configurations that have worked for me in the past but likely needs further elaboration.
Subscribers: mgorny, mehdi_amini, rriddle, jpienaar, shauheen, antiagainst, nicolasvasilache, arpith-jacob, mgester, lucyrfox, aartbik, liufengdb, stephenneuendorffer, Joonsoo, grosul1, Kayjukh, jurahul, msifontes
Tags: #mlir
Differential Revision: https://reviews.llvm.org/D83279
This patch adds type conversion for 4 SPIR-V types: array, runtime array, pointer
and struct. This conversion is integrated using a separate function
`populateSPIRVToLLVMTypeConversion()` that adds new type conversions. At the moment,
this is a basic skeleton that allows to perfom conversion from SPIR-V array,
runtime array and pointer types to LLVM typesystem. There is no support of array
strides or storage classes. These will be supported on the case by case basis.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D83399
Summary:
Added canonicalization and folding was:
- Folding when either input is an attribute indicating a scalar input
which can always be broadcasted.
- Canonicalization where it can be determined that either input shape is
a scalar.
- Canonicalization where the partially specified input shapes can be
proven to be broadcastable always.
Differential Revision: https://reviews.llvm.org/D83194
Summary:
* This allows these flags to be passed on the command line with normal CMake bool-interpreted values like ON/OFF instead of requiring 0/1.
* As-is, if passing ON/OFF, these will cause a parse error in lit.site.cfg.py because Python tries to interpret the string literally.
Reviewers: stephenneuendorffer
Subscribers: mgorny, mehdi_amini, rriddle, jpienaar, shauheen, antiagainst, nicolasvasilache, arpith-jacob, mgester, lucyrfox, aartbik, liufengdb, Joonsoo, grosul1, Kayjukh, jurahul, msifontes
Tags: #mlir
Differential Revision: https://reviews.llvm.org/D83451
Depending on where the 0 dimension is within the shape, the parser will currently reject .mlir generated by the printer.
Differential Revision: https://reviews.llvm.org/D83445
This patch adds conversion patterns for `spv.BitFieldSExtract` and `spv.BitFieldUExtract`.
As in the patch for `spv.BitFieldInsert`, `offset` and `count` have to be broadcasted in
vector case and casted to match the type of the base.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D82640
This patch introduces 3 new direct conversions for SPIR-V ops:
- `spv.Select`
- `spv.Undef`
- `spv.FMul` that was skipped in the patch with arithmetic ops
Differential Revision: https://reviews.llvm.org/D83291
Create the framework and testing environment for MLIR Reduce - a tool
with the objective to reduce large test cases into smaller ones while
preserving their interesting behavior.
Implement the functionality to parse command line arguments, parse the
MLIR test cases into modules and run the interestingness tests on
the modules.
Reviewed By: jpienaar
Differential Revision: https://reviews.llvm.org/D82803
This revision adds foldings for ExtractOp operations that come from previous InsertOp.
InsertOp have cumulative semantic where multiple chained inserts are necessary to produce the final value from which the extracts are obtained.
Additionally, TransposeOp may be interleaved and need to be tracked in order to follow the producer consumer relationships and properly compute positions.
Differential revision: https://reviews.llvm.org/D83150
An operation can specify that an operation or result type matches the
type of another operation, result, or attribute via the `AllTypesMatch`
or `TypesMatchWith` constraints.
Use these constraints to also automatically resolve types in the
automatically generated assembly parser.
This way, only the attribute needs to be listed in `assemblyFormat`,
e.g. for constant operations.
Reviewed By: rriddle
Differential Revision: https://reviews.llvm.org/D78434
with the objective to reduce large test cases into smaller ones while
preserving their interesting behavior.
Implement the framework to parse the command line arguments, parse the
input MLIR test case into a module and call reduction passes on the MLIR module.
Implement the Tester class which allows the different reduction passes to test the
interesting behavior of the generated reduced variants of the test case and keep track
of the most reduced generated variant.
scf.if currently lacks folding on true / false conditionals.
Such foldings are a bit more involved than can be addressed immediately.
This revision introduces an eager folding for lowering vector.transfer operations in the presence of unrolling.
Differential revision: https://reviews.llvm.org/D83146
The UnrollVectorPattern is can be used in a programmable fashion by:
```
OwningRewritePatternList patterns;
patterns.insert<UnrollVectorPattern<AddFOp>>(ArrayRef<int64_t>{2, 2}, ctx);
patterns.insert<UnrollVectorPattern<vector::ContractionOp>>(
ArrayRef<int64_t>{2, 2, 2}, ctx);
...
applyPatternsAndFoldGreedily(getFunction(), patterns);
```
Differential revision: https://reviews.llvm.org/D83064
Introduce pass to convert parallel affine.for op into 1-D
affine.parallel op. Run using --affine-parallelize. Removes
test-detect-parallel: pass for checking parallel affine.for ops.
Differential Revision: https://reviews.llvm.org/D82672
While lowering min/max pooling ops to loops, generate the right kind of
load/stores (std or affine) instead of always generating std
load/stores.
Differential Revision: https://reviews.llvm.org/D83080
ViewLikeOpInterfaces introduce new aliases that need to be added to the alias
list. This is necessary to place deallocs in the right positions.
Differential Revision: https://reviews.llvm.org/D83044
This pass removes redundant dialect-independent Copy operations in different
situations like the following:
%from = ...
%to = ...
... (no user/alias for %to)
copy(%from, %to)
... (no user/alias for %from)
dealloc %from
use(%to)
Differential Revision: https://reviews.llvm.org/D82757
Default vector.contract lowering essentially yields a series of sdot/ddot
operations. However, for some layouts a series of saxpy/daxpy operations,
chained through fma are more efficient. This CL introduces a choice between
the two lowering paths. A default heuristic is to follow.
Some preliminary avx2 performance numbers for matrix-times-vector.
Here, dot performs best for 64x64 A x b and saxpy for 64x64 A^T x b.
```
------------------------------------------------------------
A x b A^T x b
------------------------------------------------------------
GFLOPS sdot (reassoc) saxpy sdot (reassoc) saxpy
------------------------------------------------------------
1x1 0.6 0.9 0.6 0.9
2x2 2.5 3.2 2.4 3.5
4x4 6.4 8.4 4.9 11.8
8x8 11.7 6.1 5.0 29.6
16x16 20.7 10.8 7.3 43.3
32x32 29.3 7.9 6.4 51.8
64x64 38.9 79.3
128x128 32.4 40.7
------------------------------------------------------------
```
Reviewed By: nicolasvasilache, ftynse
Differential Revision: https://reviews.llvm.org/D83012
This patch introduces conversion pattern for `spv.constant` with scalar
and vector types. There is a special case when the constant value is a
signed/unsigned integer (vector of integers). Since LLVM dialect does not
have signedness semantics, the types had to be converted to signless ints.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D82936
This commit augments spv.CopyMemory's implementation to support 2 memory
access operands. Hence, more closely following the spec. The following
changes are introduces:
- Customize logic for spv.CopyMemory serialization and deserialization.
- Add 2 additional attributes for source memory access operand.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D82710
Added conversion pattern for SPIR-V `FunctionCallOp`. Based on
specification, it returns no results or a single result, so
can be mapped directly to LLVM dialect's `llvm.call`.
Reviewed By: antiagainst, ftynse
Differential Revision: https://reviews.llvm.org/D83030
This patch introduces conversion pattern for `spv.BitFiledInsert` op,
as well as some utility functions to facilitate code reading.
Since `spv.BitFiledInsert` may take both vector and integer operands,
this case was specifically handled by broadcasting values (`count`
and `offset` here) to vectors. Moreover, the types had to be converted
to same bitwidth in order to conform with LLVM dialect rules.
This was done with `zext` when extending (Note that `count` and
`offset` are treated as unsigned) and `trunc` in the opposite case.
For the latter one, truncation is safe since the op is defined only when
`count`/`offset`/their sum is less than the bitwidth of the result.
This introduces a natural bound of the value of 64, which can be
expressed as `i8`.
Reviewed By: antiagainst, ftynse
Differential Revision: https://reviews.llvm.org/D82639
This enables better support for traits such as SameOperandsAndResultType, and other situations in which a variadic operand may be resolved from a non-variadic.
Differential Revision: https://reviews.llvm.org/D83011
This allow lowering to support scf.for and scf.if with results. As right now
spv region operations don't have return value the results are demoted to
Function memory. We create one allocation per result right before the region
and store the yield values in it. Then we can load back the value from
allocation to be able to use the results.
Differential Revision: https://reviews.llvm.org/D82246
This revision adds support to ODS for generating interfaces for attributes and types, in addition to operations. These interfaces can be specified using `AttrInterface` and `TypeInterface` in place of `OpInterface`. All of the features of `OpInterface` are supported except for the `verify` method, which does not have a matching representation in the Attribute/Type world. Generating these interface can be done using `gen-(attr|type)-interface-(defs|decls|docs)`.
Differential Revision: https://reviews.llvm.org/D81884
Summary:
This changes the casing of MLIRGPUtoGPURuntimeTransforms to be consistent
with other transform libraries.
Differential Revision: https://reviews.llvm.org/D82841
More efficient implementation of the multiply-reduce pair,
no need to add in a zero vector. Microbenchmarking on AVX2
yields the following difference in vector.contract speedup
(over strict-order scalar reduction).
SPEEDUP SIMD-fma SIMD-mul
4x4 1.45 2.00
8x8 1.40 1.90
32x32 5.32 5.80
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D82833
Summary: The current BufferPlacement implementation does not support
nested region control flow. This CL adds support for nested regions via
the RegionBranchOpInterface and the detection of branch-like
(ReturnLike) terminators inside nested regions.
Differential Revision: https://reviews.llvm.org/D81926
These map to the similar accessors on ArrayRef and other random access containers.
This fixes a compilation error on MLIR ODS for variadic operands/results, which relied on the availability of front in certain situations.
Added conversion pattern and tests for `spv.Bitcast` op. This one has
a direct mapping in LLVM dialect so `DirectConversionPattern` was used.
Differential Revision: https://reviews.llvm.org/D82748
Also fixed bug in type inferface generator to address bug where operands and
attributes are interleaved.
Differential Revision: https://reviews.llvm.org/D82819
This patch introduces new conversion patterns for bit and logical
negation op: `spv.Not` and `spv.LogicalNot`. They are implemented
by applying xor on the operand and mask with all bits set.
Differential Revision: https://reviews.llvm.org/D82637
Current Affine comparison builders, which use operator overload, default to signed comparison. This creates the possibility of misuse of these builders and potential correctness issues when dealing with unsigned integers. This change makes the distinction between signed and unsigned comparison builders and forces the caller to make a choice between the two.
Differential Revision: https://reviews.llvm.org/D82323
Summary:
The patch makes the index type lowering of the GPU to NVVM/ROCDL conversion configurable. It introduces a pass option that controls the bitwidth used when lowering index computations and uses the LowerToLLVMOptions structure to control the Standard to LLVM lowering.
This commit fixes a use-after-free bug introduced by the reverted commit d10b1a3. It implements the following changes:
- Added a getDefaultOptions method to the LowerToLLVMOptions struct that returns a reference to statically allocated default options.
- Use the getDefaultOptions method to provide default LowerToLLVMOptions (instead of an initializer list).
- Added comments to clarify the required lifetime of the LowerToLLVMOptions
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D82475
`llvm.mlir.constant` was originally introduced as an LLVM dialect counterpart
to `std.constant`. As such, it was supporting "function pointer" constants
derived from the symbol name. This is different from `std.constant` that allows
for creation of a "function" constant since MLIR, unlike LLVM IR, supports
this. Later, `llvm.mlir.addressof` was introduced as an Op that obtains a
constant pointer to a global in the LLVM dialect. It naturally extends to
functions (in LLVM IR, functions are globals) and should be used for defining
"function pointer" values instead.
Fixes PR46344.
Differential Revision: https://reviews.llvm.org/D82667
When the origin of a shape is an extent tensor the operation `get_extent` can be
lowered directly to `extract_element`.
This choice circumvents the necessity to materialize the shape in memory.
Differential Revision: https://reviews.llvm.org/D82645
When the shape is derived from a tensor argument the shape extent can be derived
directly from that tensor with `std.dim`.
This lowering pattern circumvents the necessity to materialize the shape in
memory.
Differential Revision: https://reviews.llvm.org/D82644
The error message in the `std.constant` verifier for function-typed constants
had the name of the undefined function hardcoded to `bar`. Report the actual
name instead.
Differential Revision: https://reviews.llvm.org/D82666
This test largely predates MLIR testing guidelines. Update it to match the
guidelines. In particular, avoid pattern-matching SSA value names, avoid
unnecessary CHECK-NEXT, relax assumptions about the form of SSA names.
Value-returning operations are still matched agaist _any_ name in order to
check that the operation indeed produces values.
Differential Revision: https://reviews.llvm.org/D82656
Rationale:
In general, passing "fastmath" from MLIR to LLVM backend is not supported, and even just providing such a feature for experimentation is under debate. However, passing fine-grained fastmath related attributes on individual operations is generally accepted. This CL introduces an option to instruct the vector-to-llvm lowering phase to annotate floating-point reductions with the "reassociate" fastmath attribute, which allows the LLVM backend to use SIMD implementations for such constructs. Oher lowering passes can start using this mechanism right away in cases where reassociation is allowed.
Benefit:
For some microbenchmarks on x86-avx2, speedups over 20 were observed for longer vector (due to cleaner, spill-free and SIMD exploiting code).
Usage:
mlir-opt --convert-vector-to-llvm="reassociate-fp-reductions"
Reviewed By: ftynse, mehdi_amini
Differential Revision: https://reviews.llvm.org/D82624
Implemented conversion for `spv.BitReverse` and `spv.BitCount`. Since ODS
generates builders in a different way for LLVM dialect intrinsics, I
added attributes to build method in `DirectConversionPattern` class. The
tests for these ops are in `bitwise-ops-to-llvm.mlir`.
Differential Revision: https://reviews.llvm.org/D82286
Add a pass to rewrite sequential chains of `spirv::CompositeInsert`
operations into `spirv::CompositeConstruct` operations.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D82198
This patch add support for 'spv.CopyMemory'. The following changes are
introduced:
- 'CopyMemory' op is added to SPIRVOps.td.
- Custom parse and print methods are introduced.
- A few Roundtripping tests are added.
Differential Revision: https://reviews.llvm.org/D82384
Summary: The patch fixes an off by one error in the method collapseParallelLoops. It ensures the same normalized bound is used for the computation of the division and the remainder.
Reviewers: herhut
Reviewed By: herhut
Subscribers: mehdi_amini, rriddle, jpienaar, shauheen, antiagainst, nicolasvasilache, arpith-jacob, mgester, lucyrfox, aartbik, liufengdb, stephenneuendorffer, Joonsoo, grosul1, Kayjukh, jurahul, msifontes
Tags: #mlir
Differential Revision: https://reviews.llvm.org/D82634
Initially, unranked memref descriptors in the LLVM dialect were designed only
to be passed into functions. An assertion was guarding against returning
unranked memrefs from functions in the standard-to-LLVM conversion. This is
insufficient for functions that wish to return an unranked memref such that the
caller does not know the rank in advance, and hence cannot allocate the
descriptor and pass it in as an argument.
Introduce a calling convention for returning unranked memref descriptors as
follows. An unranked memref descriptor always points to a ranked memref
descriptor stored on stack of the current function. When an unranked memref
descriptor is returned from a function, the ranked memref descriptor it points
to is copied to dynamically allocated memory, the ownership of which is
transferred to the caller. The caller is responsible for deallocating the
dynamically allocated memory and for copying the pointed-to ranked memref
descriptor onto its stack.
Provide default lowerings for std.return, std.call and std.indirect_call that
maintain the conversion defined above.
This convention is additionally exercised by a runtime test to guard against
memory errors.
Differential Revision: https://reviews.llvm.org/D82647
Using fully qualified names wherever possible avoids ambiguous class and function names. This is a follow-up to D82371.
Reviewed By: rriddle
Differential Revision: https://reviews.llvm.org/D82471
When there is a mix of affine load/store and non-affine operations (e.g. std.load, std.store),
affine-loop-fusion ignores the present of non-affine ops, thus changing the program semantics.
E.g. we have a program of three affine loops operating on the same memref in which one of them uses std.load and std.store, as follows.
```
affine.for
affine.store %1
affine.for
std.load %1
std.store %1
affine.for
affine.load %1
affine.store %1
```
affine-loop-fusion will produce the following result which changed the program semantics:
```
affine.for
std.load %1
std.store %1
affine.for
affine.store %1
affine.load %1
affine.store %1
```
This patch is to fix the above problem by checking non-affine users of the memref that are between the source and destination nodes of interest.
Differential Revision: https://reviews.llvm.org/D82158
Lower `shape.rank` to standard dialect.
A shape's size is the same as the extent of the first and only dimension of the
`tensor<?xindex>` it is represented by.
Differential Revision: https://reviews.llvm.org/D82080
Replace any `rank(shape_of(tensor))` that relies on a ranked tensor with the
corresponding constant `const_size`.
Differential Revision: https://reviews.llvm.org/D82077
Summary: The patch optimizes the tiling of parallel loops with static bounds if the number of loop iterations is an integer multiple of the tile size.
Reviewers: herhut, ftynse, bondhugula
Reviewed By: herhut, ftynse
Subscribers: bondhugula, mehdi_amini, rriddle, jpienaar, shauheen, antiagainst, nicolasvasilache, arpith-jacob, mgester, lucyrfox, aartbik, liufengdb, stephenneuendorffer, Joonsoo, grosul1, frgossen, Kayjukh, jurahul, msifontes
Tags: #mlir
Differential Revision: https://reviews.llvm.org/D82003
This patch introduces conversion patterns for `spv.module` and `spv._module_end`.
SPIR-V module is converted into `ModuleOp`. This will play a role of enclosing
scope to LLVM ops. At the moment, SPIR-V module attributes (such as memory model,
etc) are ignored.
Differential Revision: https://reviews.llvm.org/D82468
Use vector compares for the 1-D case. This approach scales much better
than generating insertion operations, and exposes SIMD directly to backend.
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D82402
Introduced `llvm.intr.bitreverse` and `llvm.intr.ctpop` LLVM bit
intrinsics to LLVM dialect. These intrinsics help with SPIR-V to
LLVM conversion, allowing a direct mapping from `spv.BitReverse`
and `spv.BitCount` respectively. Tests are added to `roundtrip.mlir`
and `llvm-intrinsics.mlir`.
Differential Revision: https://reviews.llvm.org/D82285
This patch provides an implementation for `spv.func` conversion. The pattern
is populated in a separate method added to the pass. At the moment, the type
signature conversion only includes the supported types. The conversion pattern
also matches SPIR-V function control attributes to LLVM function attributes.
Those are modelled as `passthrough` attributes in LLVM dialect. The following
mapping are used:
- None: no attributes passed
- Inline: `alwaysinline` seems to be the right equivalent (`inlinehint` is
semantically weaker in my opinion)
- DontInline: `noinline`
- Pure and Const: I think those can be modelled as `readonly` and `readnone`
attributes respectively.
Also, 2 patterns added for return ops conversion (`spv.Return` for void return
and `spv.ReturnValue` for a single value return).
Differential Revision: https://reviews.llvm.org/D81931
Add option to filter which op the OpDefinitionsGen run on. This enables having multiple ops together in the same TD file but generating different CC files for them (useful if one wants to use multiclasses or split out 1 dialect into multiple different libraries). There is probably more general query here (e.g., split out all ops that don't have a verify method, or that are commutative) but filtering based on op name (e.g., test.a_op) seemed a reasonable start and didn't require inventing a query specification mechanism here.
Differential Revision: https://reviews.llvm.org/D82319
Summary:
Currently, the TableGen rewrite generates redundant native calls in MLIR DRR files. This is a problem as some native calls may involve significant computations (e.g. when performing constant propagation where every values in a large tensor is touched).
The pattern was as follow:
```c++
if (native-call(args)) tblgen_attrs.emplace_back(rewriter, attribute, native-call(args))
```
The replacement pattern compute `native-call(args)` once and then use it both in the `if` condition and the `emplace_back` call.
Differential Revision: https://reviews.llvm.org/D82101
This patch extends the AccessChainOp index type handling to be able to deal with
all Integer type indices (i.e., all bit-widths and signedness symantics).
There were two ways of achieving this:
1- Backward compatible: The new way of handling the indices will assume that
an index type is i32 by default if not specified in the assembly format,
this way all the old tests would pass correctly.
2- Enforce the format: This unifies the spv.AccessChain Op format and all the old
tests had to be updated to reflect this change or else they fail.
I picked option-2 to unify the Op format and avoid having optional index-type fields
that can lead to somewhat confusing tests format and multiple representations for
the same Op with undocumented assumption that an index is i32 unless stated.
Nonetheless, reverting to option-1 should be straightforward if preferred or needed.
Differential Revision: https://reviews.llvm.org/D81763
The patch makes the index type lowering of the GPU to NVVM/ROCDL
conversion configurable. It introduces a pass option that controls the
bitwidth used when lowering index computations.
Differential Revision: https://reviews.llvm.org/D80285
Summary:
We already had a parallel loop specialization pass that is used to
enable unrolling and consecutive vectorization by rewriting loops
whose bound is defined as a min of a constant and a dynamic value
into a loop with static bound (the constant) and the minimum as
bound, wrapped into a conditional to dispatch between the two.
This adds the same rewriting for for loops.
Differential Revision: https://reviews.llvm.org/D82189
Subview operations are not natively supported downstream in the spirv path.
This change allows removing subview when used by vector transfer the same way
we already do it when they are used by LoadOp/StoreOp
Differential Revision: https://reviews.llvm.org/D82106
Allow lhs and rhs to have different type than accumulator/destination. Some
hardware like GPUs support natively operations like uint8xuint8xuint32.
Differential Revision: https://reviews.llvm.org/D82069
Use direct vector constants for the 1-D case. This approach
scales much better than generating elaborate insertion operations
that are eventually folded into a constant. We could of course
generalize the 1-D case to higher ranks, but this simplification
already helps in scaling some microbenchmarks that would formerly
crash on the intermediate IR length.
Reviewed By: reidtatge
Differential Revision: https://reviews.llvm.org/D82144
Lower `shape.shape_of` to standard dialect.
This lowering supports statically and dynamically shaped tensors.
Support for unranked tensors will be added as part of the lowering to `scf`.
Differential Revision: https://reviews.llvm.org/D82098
Summary:
With this change, a function argument attribute of the form
"llvm.align" = <int> will be translated to the corresponding align
attribute in LLVM by the ModuleConversion.
Differential Revision: https://reviews.llvm.org/D82161
This patch adds the `default_triple` feature to MLIR test suite.
This feature was added to LLVM in d178f4fc8 in order to be able to
run the LLVM tests without having the host targets configured in.
With this change, `ninja check-mlir` passes without the host
target, i.e. this config:
cmake ../llvm -DLLVM_TARGETS_TO_BUILD="" -DLLVM_DEFAULT_TARGET_TRIPLE="" -DLLVM_ENABLE_PROJECTS=mlir -GNinja
Differential Revision: https://reviews.llvm.org/D82142
This revision removes the TypeConverter parameter passed to the apply* methods, and instead moves the responsibility of region type conversion to patterns. The types of a region can be converted using the 'convertRegionTypes' method, which acts similarly to the existing 'applySignatureConversion'. This method ensures that all blocks within, and including those moved into, a region will have the block argument types converted using the provided converter.
This has the benefit of making more of the legalization logic controlled by patterns, instead of being handled explicitly by the driver. It also opens up the possibility to support multiple type conversions at some point in the future.
This revision also adds a new utility class `FailureOr<T>` that provides a LogicalResult friendly facility for returning a failure or a valid result value.
Differential Revision: https://reviews.llvm.org/D81681
Traditionally patterns have always had the root operation kind hardcoded to a specific operation name. This has worked well for quite some time, but it has certain limitations that make it undesirable. For example, some lowering have the same implementation for many different operations types with a few lowering entire dialects using the same pattern implementation. This problem has led to several "solutions":
a) Provide a template implementation to the user so that they can instantiate it for each operation combination, generally requiring the inclusion of the auto-generated operation definition file.
b) Use a non-templated pattern that allows for providing the name of the operation to match
- No one ever does this, because enumerating operation names can be cumbersome and so this quickly devolves into solution a.
This revision removes the restriction that patterns have a hardcoded root type, and allows for a class patterns that could match "any" operation type. The major downside of root-agnostic patterns is that they make certain pattern analyses more difficult, so it is still very highly encouraged that an operation specific pattern be used whenever possible.
Differential Revision: https://reviews.llvm.org/D82066
muladd can have differenti types for lhs/rhs and acc/destination. Change
verifier and update the test to use supported example.
Differential Revision: https://reviews.llvm.org/D82042
Summary:
The "i1" (viz. bool) type does not have a proper equivalent on the "C"
size. So, to avoid any ABIs issues, we simply use print_i32 on an i32
value of one or zero for true and false. This has the added advantage
that one less function needs to be implemented when porting the runtime
support library.
Reviewers: ftynse, bkramer, nicolasvasilache
Reviewed By: ftynse
Subscribers: mehdi_amini, rriddle, jpienaar, shauheen, antiagainst, nicolasvasilache, arpith-jacob, mgester, lucyrfox, liufengdb, stephenneuendorffer, Joonsoo, grosul1, frgossen, Kayjukh, jurahul, msifontes
Tags: #mlir
Differential Revision: https://reviews.llvm.org/D82048
Summary:
Fixed build of D81618
Add a pattern for expanding tanh op into exp form.
A `tanh` is expanded into:
1) 1-exp^{-2x} / 1+exp^{-2x}, if x => 0
2) exp^{2x}-1 / exp^{2x}+1 , if x < 0.
Differential Revision: https://reviews.llvm.org/D82040
Summary:
This is to provide a utility to remove unsupported constraints or for
pipelines that happen to receive these but cannot lower them due to not
supporting assertions.
Differential Revision: https://reviews.llvm.org/D81560
The ScopedBuilder class in EDSC is being gradually phased out in favor of core
OpBuilder-based helpers with callbacks. Provide helper functions that are
compatible with `edsc::ScopedContext` and can be used to create and populate
blocks using callbacks that take block arguments as callback arguments. This
removes the need for `edsc::BlockHandle`, forward-declaration of `Value`s used
for block arguments and the tag `edsc::Append` class, leading to noticable
reduction in the verbosity of the code using helper functions.
Remove "eager mode" construction tests that are only relevant to the
`BlockBuilder`-based approach.
`edsc::BlockHandle` and `edsc::BlockBuilder` are now deprecated and will be
removed soon.
Differential Revision: https://reviews.llvm.org/D82008
This patch adjust the load/store matrix intrinsics, formerly known as
llvm.matrix.columnwise.load/store, to improve the naming and allow
passing of extra information (volatile).
The patch performs the following changes:
* Rename columnwise.load/store to column.major.load/store. This is more
expressive and also more in line with the naming in Clang.
* Changes the stride arguments from i32 to i64. The stride can be
larger than i32 and this makes things more uniform with the way
things are handled in Clang.
* A new boolean argument is added to indicate whether the load/store
is volatile. The lowering respects that when emitting vector
load/store instructions
* MatrixBuilder is updated to require both Alignment and IsVolatile
arguments, which are passed through to the generated intrinsic. The
alignment is set using the `align` attribute.
The changes are grouped together in a single patch, to have a single
commit that breaks the compatibility. We probably should be fine with
updating the intrinsics, as we did not yet officially support them in
the last stable release. If there are any concerns, we can add
auto-upgrade rules for the columnwise intrinsics though.
Reviewers: anemet, Gerolf, hfinkel, andrew.w.kaylor, LuoYuanke, nicolasvasilache, rjmccall, ftynse
Reviewed By: anemet, nicolasvasilache
Differential Revision: https://reviews.llvm.org/D81472
We previously weren't properly updating the SCC iterator when nodes were removed, leading to asan failures in certain situations. This commit adds a CallGraphSCC class and defers operation deletion until inlining has finished.
Differential Revision: https://reviews.llvm.org/D81984
Summary:
- Define the MatrixTimesScalar operation and add roundtrip tests.
- Added a new base class for matrix-specific operations to avoid invalid operands type mismatch check.
- Created a separate Matrix arithmetic operations td file to add more operations in the future.
- Augmented the automatically generated verify method to print more fine-grained error messages.
- Made minor Updates to the matrix type tests.
Reviewers: antiagainst, rriddle, mravishankar
Reviewed By: antiagainst
Subscribers: mehdi_amini, jpienaar, shauheen, nicolasvasilache, arpith-jacob, mgester, lucyrfox, aartbik, liufengdb, stephenneuendorffer, Joonsoo, bader, grosul1, frgossen, Kayjukh, jurahul, msifontes
Tags: #mlir
Differential Revision: https://reviews.llvm.org/D81677
Added support of simple logical ops: `LogicalAnd`, `LogicalOr`,
`LogicalEqual` and `LogicalNotEqual`. Added a missing conversion
for `UMod` op.
Also, implemented SPIR-V cast ops conversion. There are 4 simple
case where there is a clear equivalent in LLVM (e.g. `ConvertFToS`
is `fptosi`). For `FConvert`, `SConvert` and `UConvert` we
distinguish between truncation and extension based on the bit
width of the operand.
Differential Revision: https://reviews.llvm.org/D81812
Summary:
Parallel loop tiling did not properly compute the updated loop
indices when tiling, which lead to wrong results.
Differential Revision: https://reviews.llvm.org/D82013
Implement the missing lowering from `std.dim` to the LLVM dialect in case of a
dynamic dimension.
Differential Revision: https://reviews.llvm.org/D81834
Summary:
This revision replaces MatmulOp, now that DRR rules have been dropped.
This revision also fixes minor parsing bugs and a plugs a few holes to get e2e paths working (e.g. library call emission).
During the replacement the i32 version had to be dropped because only the EDSC operators +, *, etc support type inference.
Deciding on a type-polymorphic behavior, and implementing it, is left for future work.
Reviewers: aartbik
Subscribers: mehdi_amini, rriddle, jpienaar, shauheen, antiagainst, arpith-jacob, mgester, lucyrfox, aartbik, liufengdb, stephenneuendorffer, Joonsoo, grosul1, frgossen, Kayjukh, jurahul, msifontes
Tags: #mlir
Differential Revision: https://reviews.llvm.org/D81935
Fix memref region compute for 0-d memref accesses in certain cases (when
there are loops surrounding such 0-d accesses).
Differential Revision: https://reviews.llvm.org/D81792
- Modify HasParent trait to allow one of several op's as a parent -
- Expose this trait in the ODS framework using the ParentOneOf<> trait.
Differential Revision: https://reviews.llvm.org/D81880
This allows for passing a lambda to addDynamicallyLegalDialect without needing to explicit wrap with Optional<DynamicLegalityCallbackFn>.
Differential Revision: https://reviews.llvm.org/D81680
It is quite common for the same type to be converted many types throughout the conversion process, and there isn't any good reason why we aren't caching that result. Especially given that we currently use identity conversion to signify legality. This revision also adds a few additional helpers to TypeConverter.
Differential Revision: https://reviews.llvm.org/D81679
This revision replaces MatmulOp, now that DRR rules have been dropped.
This revision also fixes minor parsing bugs and a plugs a few holes to get e2e paths working (e.g. library call emission).
During the replacement the i32 version had to be dropped because only the EDSC operators +, *, etc support type inference.
Deciding on a type-polymorphic behavior, and implementing it, is left for future work.
Differential Revision: https://reviews.llvm.org/D79762
This reverts commit 32c757e4f8.
Broke the build bot:
******************** TEST 'MLIR :: Examples/standalone/test.toy' FAILED ********************
[...]
/tmp/ci-KIMiRFcVZt/lib/libMLIRLinalgToLLVM.a(LinalgToLLVM.cpp.o): In function `(anonymous namespace)::ConvertLinalgToLLVMPass::runOnOperation()':
LinalgToLLVM.cpp:(.text._ZN12_GLOBAL__N_123ConvertLinalgToLLVMPass14runOnOperationEv+0x100): undefined reference to `mlir::populateExpandTanhPattern(mlir::OwningRewritePatternList&, mlir::MLIRContext*)'
Lubomir Litchev