This class used to serve a few useful purposes:
* Allowed containing a null DictionaryAttr
* Provided some simple mutable API around a DictionaryAttr
The first of which is no longer an issue now that there is much better caching support for attributes in general, and a cache in the context for empty dictionaries. The second results in more trouble than it's worth because it mutates the internal dictionary on every action, leading to a potentially large number of dictionary copies. NamedAttrList is a much better alternative for the second use case, and should be modified as needed to better fit it's usage as a DictionaryAttrBuilder.
Differential Revision: https://reviews.llvm.org/D93442
Reductions in innermost loops become harder for the backend to disambiguate
after bufferization into memrefs, resulting in less efficient load-update-store
cycles. By scalarizing innermost reductions, the backend is more likely to assign
a register to perform the reduction (also prepares vectorization). Even though
we could scalarize reductions for more outer loops and while-loops as well,
currently scalarization is only done for chains of innermost for-loops, where
it matters most, to avoid complicating codegen unnecessary (viz. adding lots
of yield instructions).
This CL also refactors condition simplification into the merger class,
where it belongs, so that conditions are simplified only once per loop
nest and not repeatedly as was currently done. This CL also fixes a few
minor bugs, some layout issues, and comments.
Reviewed By: penpornk
Differential Revision: https://reviews.llvm.org/D93143
This operation is used to materialize a tensor of a particular
shape. The shape could be specified as a mix of static and dynamic
values.
The use of this operation is to be an `init` tensor for Linalg
structured operation on tensors where the bounds of the computation
depends on the shape of the output of the linalg operation. The result
of this operation will be used as the `init` tensor of such Linalg
operations. To note,
1) The values in the tensor materialized is not used. Any operation to
which this is an init tensor is expected to overwrite the entire
tensor.
2) The tensor is materialized only for the shape of the output and to
make the loop bounds depend only on operands of the structured
operation.
Based on (1) and (2) it is assumed that these operations eventually go
away since they are only used in `dim` operations that can be
canonicalized to make this operation dead. Such canonicalization are
added here too.
Differential Revision: https://reviews.llvm.org/D93374
This better matches the rest of the infrastructure, is much simpler, and makes it easier to move these types to being declaratively specified.
Differential Revision: https://reviews.llvm.org/D93432
This document was not updated after the LLVM dialect type system had been
reimplemented and was using an outdated syntax. Rewrite the part of the
document that concerns type conversion and prepare the ground for splitting it
into a document that explains how built-in types are converted and a separate
document that explains how standard types and functions are converted, which
will better correspond to the fact that built-in types do not belong to the
standard dialect.
Reviewed By: rriddle
Differential Revision: https://reviews.llvm.org/D93486
The LLVM IR 'switch' instruction allows control flow to be transferred
to one of any number of branches depending on an integer control value,
or a default value if the control does not match any branch values. This patch
adds `llvm.switch` to the MLIR LLVMIR dialect, as well as translation routines
for lowering it to LLVM IR.
To store a variable number of operands for a variable number of branch
destinations, the new op makes use of the `AttrSizedOperandSegments`
trait. It stores its default branch operands as one segment, and all
remaining case branches' operands as another. It also stores pairs of
begin and end offset values to delineate the sub-range of each case branch's
operands. There's probably a better way to implement this, since the
offset computation complicates several parts of the op definition. This is the
approach I settled on because in doing so I was able to delegate to the default
op builder member functions. However, it may be preferable to instead specify
`skipDefaultBuilders` in the op's ODS, or use a completely separate
approach; feedback is welcome!
Another contentious part of this patch may be the custom printer and
parser functions for the op. Ideally I would have liked the MLIR to be
printed in this way:
```
llvm.switch %0, ^bb1(%1 : !llvm.i32) [
1: ^bb2,
2: ^bb3(%2, %3 : !llvm.i32, !llvm.i32)
]
```
The above would resemble how LLVM IR is formatted for the 'switch'
instruction. But I found it difficult to print and parse something like
this, whether I used the declarative assembly format or custom functions.
I also was not sure a multi-line format would be welcome -- it seems
like most MLIR ops do not use newlines. Again, I'd be happy to hear any
feedback here as well, or on any other aspect of the patch.
Differential Revision: https://reviews.llvm.org/D93005
This commit shuffles SPIR-V code around to better follow MLIR
convention. Specifically,
* Created IR/, Transforms/, Linking/, and Utils/ subdirectories and
moved suitable code inside.
* Created SPIRVEnums.{h|cpp} for SPIR-V C/C++ enums generated from
SPIR-V spec. Previously they are cluttered inside SPIRVTypes.{h|cpp}.
* Fixed include guards in various header files (both .h and .td).
* Moved serialization tests under test/Target/SPIRV.
* Renamed TableGen backend -gen-spirv-op-utils into -gen-spirv-attr-utils
as it is only generating utility functions for attributes.
Reviewed By: mravishankar
Differential Revision: https://reviews.llvm.org/D93407
This was long overdue. The initial documentation for the LLVM dialect was
introduced before ODS had support for long descriptions. This is now possible,
so the documentation is moved to ODS, which can serve as a single source of
truth. The high-level description of the dialect structure is updated to
reflect that.
Depends On: D93315
Reviewed By: rriddle, mehdi_amini
Differential Revision: https://reviews.llvm.org/D93425
Rewrite the parts of the documentation that became stale: context/module
handling and type system. Expand the type system description.
Reviewed By: nicolasvasilache
Differential Revision: https://reviews.llvm.org/D93315
This operation is designed to support partial conversion, more specifically the
IR state in which some operations expect or produce built-in types and some
operations produce and expect LLVM dialect types. It is reasonable for it to
support cast between built-in types and any equivalent that could be produced
by the type conversion. (At the same time, we don't want the dialect to depend
on the type conversion as it could lead to a dependency cycle). Introduce
support for casting from index to any integer type and back, and from memref to
bare pointer or memref descriptor type and back.
Contrary to what the TODO in the code stated, there are no particular
precautions necessary to handle the bare pointer conversion for memerfs. This
conversion applies exclusively to statically-shaped memrefs, so we can always
recover the full descriptor contents from the type.
This patch simultaneously tightens the verification for other types to only
accept matching pairs of types, e.g., i64 and !llvm.i64, as opposed to the
previous implementation that only checked if the types were generally allowed
byt not for matching, e.g. i64 could be "casted" to !llvm.bfloat, which is not
the intended semantics.
Move the relevant test under test/Dialect/LLVMIR because it is not specific to
the conversion pass, but rather exercises an op in the dialect. If we decide
this op does not belong to the LLVM dialect, both the dialect and the op should
move together.
Reviewed By: silvas, ezhulenev
Differential Revision: https://reviews.llvm.org/D93405
Test flakiness was fixed by: 9edcedf7f2
Runs these tests to verify that all parts of the lowering work correctly.
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D93384
Define Async runtime related typedefs in the `mlir::runtime` namespace.
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D93391
- the !gpu.async.token is the second result of 'gpu.alloc async', not the first.
- async.execute construction takes operand types not yet wrapped in !async.value.
- fix typo
Reviewed By: herhut
Differential Revision: https://reviews.llvm.org/D93156
Adds more support for `SpecConstantOperation` by defining a custom
syntax for the op and implementing its parsing and printing.
Reviewed By: mravishankar, antiagainst
Differential Revision: https://reviews.llvm.org/D92919
The current condition implies that the target materialization will be
called even if the type is the new operand type is legal, but slightly
different. For example, if there is a bufferization pattern that changes
memref layout, then target materialization for an illegal type
(TensorType) would be called.
Differential Revision: https://reviews.llvm.org/D93126
Previous behavior would fail if inserting an operation that already
existed. Now SymbolTable::insert can also be used as a way to make a
symbol's name unique even after insertion.
Further TODOs have been left over naming and consistent behavior
considerations.
Differential Revision: https://reviews.llvm.org/D93349
This parser does not include the general `pass_pipeline` option, the pass pipeline entries, or the options of pass entries. This is useful for cases such as `print-ir-after` that just want the user to select specific pass types. This revision greatly reduces the amount of text in --help for several MLIR based tools.
Fixes PR#45495
Differential Revision: https://reviews.llvm.org/D92987
These classes were initially introduced to factor out two common parts of LLVM
op definitions: the fact that they have no results or a single result of
LLVM_Type, and the default builders. Neither of the two parts is really
common anymore: many ops have more specific on the result type, and many ops
provide custom builders. The TableGen classes only add conceptual complexity
and make LLVM dialect definition dissimilar to other dialects. Remove them in
favor of explicitly specified builders (results are already specified).
Depends On D93329
Reviewed By: rriddle
Differential Revision: https://reviews.llvm.org/D93330
Now that we have predicates for LLVM dialect types in ODS, we can use them to
restrict the types allowed in results of LLVM dialect operations. This also
serves as additional documentation for these operations.
Reviewed By: rriddle
Differential Revision: https://reviews.llvm.org/D93329
This exposes several issues with the current generation that this revision also fixes.
* TypeDef now allows specifying the base class to use when generating.
* TypeDef now inherits from DialectType, which allows for using it as a TypeConstraint
* Parser/Printers are now no longer generated in the header(removing duplicate symbols), and are now only generated when necessary.
- Now that generatedTypeParser/Printer are only generated in the definition file,
existing users will need to manually expose this functionality when necessary.
* ::get() is no longer generated for singleton types, because it isn't necessary.
Differential Revision: https://reviews.llvm.org/D93270
Due to how the conversion infra works, the "clone" call that this
pattern was using required all the cloned ops to be immediately
legalized as part of this dialect conversion invocation.
That was previously working due to a couple factors:
- In the test case, there was scf.if, which we happen to mark as legal
as part of marking the entire SCF dialect as legal for the scf.parallel
we generate here.
- Originally, this test case had std.extract_element in the body, which
we happened to have a pattern for in this pass. After I migrated that to
`tensor.extract` (which removed the tensor.extract bufferization from
here), I hacked this up to use `std.dim` which we still have patterns
for in this pass.
This patch updates the test case to use a truly opaque op `test.source`
that properly stresses this aspect of the pattern.
(this also removes a stray dependency on the `tensor` dialect that I
must have left behind as part of my hacking this pass up when migrating
to `tensor.extract`)
Differential Revision: https://reviews.llvm.org/D93262
This is useful for scalar code that uses for/while loops.
This has also been confirmed to work for representing std.pow as an
scf.for loop on gpus.
Differential Revision: https://reviews.llvm.org/D93308
Now that passes have support for running nested pipelines, the inliner can now allow for users to provide proper nested pipelines to use for optimization during inlining. This revision also changes the behavior of optimization during inlining to optimize before attempting to inline, which should lead to a more accurate cost model and prevents the need for users to schedule additional duplicate cleanup passes before/after the inliner that would already be run during inlining.
Differential Revision: https://reviews.llvm.org/D91211
This revision adds a new `StaticVerifierFunctionEmitter` class that emits local static functions in the .cpp file for shared operation verification. This class deduplicates shared operation verification code by emitting static functions alongside the op definitions. These methods are local to the definition file, and are invoked within the operation verify methods. The first bit of shared verification is for the type constraints used when verifying operands and results. An example is shown below:
```
static LogicalResult localVerify(...) {
...
}
LogicalResult OpA::verify(...) {
if (failed(localVerify(...)))
return failure();
...
}
LogicalResult OpB::verify(...) {
if (failed(localVerify(...)))
return failure();
...
}
```
This allowed for saving >400kb of code size from a downstream TensorFlow project (~15% of MLIR code size).
Differential Revision: https://reviews.llvm.org/D91381
This will allow for caching pattern lists across multiple pass instances, such as when multithreading. This is an extremely important invariant for PDL patterns, which are compiled at runtime when the FrozenRewritePatternList is built.
Differential Revision: https://reviews.llvm.org/D93146
Some operations use integer literals as part of their custom format that don't necessarily map to an internal IntegerAttr. This revision exposes the same `parseInteger` functions as the DialectAsmParser to allow for these operations to parse integer literals without incurring the otherwise unnecessary roundtrip through IntegerAttr.
Differential Revision: https://reviews.llvm.org/D93152
This revision adds a new `printNewline` hook to OpAsmPrinter that allows for printing a newline within the custom format of an operation, that is then indented to the start of the operation. Support for the declarative assembly format is also added, in the form of a `\n` literal.
Differential Revision: https://reviews.llvm.org/D93151
Fix a bug causing to pick the wrong vector size to broadcast to when the source
vectors have different ranks.
Differential Revision: https://reviews.llvm.org/D93118
This commit splits SPIR-V's serialization and deserialization code
into separate libraries. The motiviation being that the serializer
is used more often the deserializer and therefore lumping them
together unnecessarily increases binary size for the most common
case.
This commit also moves these libraries into the Target/ directory
to follow MLIR convention.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D91548
This avoids dumping the module post emitting a reproducer, which results in
many MB logs where a reproducer has already been neatly generated.
Differential Revision: https://reviews.llvm.org/D93165
Adds support for 3 ternary ops from SPIR-V extended instructions for
GLSL. Namely, adds support for FClamp, UClamp, and SClamp.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D92859
Zero bit integer types are supported by IntegerType for consistency,
but the asmparser never got updated. Allow them to be parsed, as
required to fix CIRCT issue #316
Differential Revision: https://reviews.llvm.org/D93089
When printing verification errors for ops with the incorrect number of
operand segments, print the required number as well as the actual
number. Split off from D93005.
Differential Revision: https://reviews.llvm.org/D93145
This mirror the C++ API for NamedAttribute, and has the advantage or
internalizing earlier in the Context and not requiring the caller to
keep the StringRef alive beyong this call.
Differential Revision: https://reviews.llvm.org/D93133
This reverts commit 0d48d265db.
This reapplies the following commit, with a fix for CAPI/ir.c:
[mlir] Start splitting the `tensor` dialect out of `std`.
This starts by moving `std.extract_element` to `tensor.extract` (this
mirrors the naming of `vector.extract`).
Curiously, `std.extract_element` supposedly works on vectors as well,
and this patch removes that functionality. I would tend to do that in
separate patch, but I couldn't find any downstream users relying on
this, and the fact that we have `vector.extract` made it seem safe
enough to lump in here.
This also sets up the `tensor` dialect as a dependency of the `std`
dialect, as some ops that currently live in `std` depend on
`tensor.extract` via their canonicalization patterns.
Part of RFC: https://llvm.discourse.group/t/rfc-split-the-tensor-dialect-from-std/2347/2
Differential Revision: https://reviews.llvm.org/D92991
This starts by moving `std.extract_element` to `tensor.extract` (this
mirrors the naming of `vector.extract`).
Curiously, `std.extract_element` supposedly works on vectors as well,
and this patch removes that functionality. I would tend to do that in
separate patch, but I couldn't find any downstream users relying on
this, and the fact that we have `vector.extract` made it seem safe
enough to lump in here.
This also sets up the `tensor` dialect as a dependency of the `std`
dialect, as some ops that currently live in `std` depend on
`tensor.extract` via their canonicalization patterns.
Part of RFC: https://llvm.discourse.group/t/rfc-split-the-tensor-dialect-from-std/2347/2
Differential Revision: https://reviews.llvm.org/D92991
Introduce support for inlining into affine operations. This uses the generic
inline infrastructure and boils down to checking that, if applied, the inlining
doesn't violate the affine dimension/symbol value categorization. Given valid
IR, only the values that are valid dimensions/symbols thanks to being top-level
in their affine scope need special handling.
Reviewed By: rriddle
Differential Revision: https://reviews.llvm.org/D92770
OperationFolder currently uses ConstantOp as a backup when trying to materialize a constant after an operation is folded. This dependency isn't really useful or necessary given that dialects can/should provide a `materializeConstant` implementation.
Fixes PR#44866
Differential Revision: https://reviews.llvm.org/D92980
This fixes a subtle bug where SCCP could incorrectly optimize a private callable while waiting for its arguments to be resolved.
Fixes PR#48457
Differential Revision: https://reviews.llvm.org/D92976
This fixes a crash when no elements are parsed, but the type expects at least one.
Fixes PR#47763
Differential Revision: https://reviews.llvm.org/D92982
This was missed when supported for unsigned/signed integer types was first added, and results in crashes if a user tries to create/print a constant with the incorrect integer type.
Fixes PR#46222
Differential Revision: https://reviews.llvm.org/D92981
- use ConvertOpToLLVMPattern to avoid explicit casting and in most cases the
constructor can be reused to save a few lines of code.
Differential Revision: https://reviews.llvm.org/D92989
The current implementation of the translation to LLVM IR relies on the
existence of a one-to-one mapping between MLIR blocks and LLVM IR basic blocks
in order to configure PHI nodes with appropriate source blocks. The one-to-one
mapping model is broken in presence of OpenMP operations that use LLVM's
OpenMPIRBuilder, which produces multiple blocks under the hood. This can lead
to invalid LLVM IR being emitted if OpenMPIRBuilder moved the branch operation
into a basic block different from the one it was originally created in;
specifically, a block that is not a direct predecessor could be used in the PHI
node. Instead, keep track of the mapping between MLIR LLVM dialect branch
operations and their LLVM IR counterparts and take the parent basic block of
the LLVM IR instruction at the moment of connecting the PHI nodes to
predecessors.
This behavior cannot be triggered as of now, but will be once we introduce the
conversion of OpenMP workshare loops.
Reviewed By: kiranchandramohan
Differential Revision: https://reviews.llvm.org/D92845
The documentation has become a bit stale with age, and doesn't include great documentation for some newer concepts. This revision tidies up a majority of it, with some more cleanup to come in the future. The documentation for the declarative specification is also moved from OpDefinitions.md to Interfaces.md, which is a much more logical place for it to live.
Differential Revision: https://reviews.llvm.org/D92895
- Remove some unused types from the Shape dialect
- Fix from_extent_tensor to only allow 1D index tensors
- Fix assuming_yield to only allow shape.assuming as the parent op.
- Fix some documentation typos and reword some things.
Differential Revision: https://reviews.llvm.org/D92901
This is to prevent assertion failures on scf.if and shape.assuming
operations where this is not enough information currently to handle any
aliasing information.
Differential Revision: https://reviews.llvm.org/D92963
This patch fixes a bug that allowed vectorizing of loops with loads and
stores having indexing functions varying along different memory
dimensions.
Reviewed By: aartbik, dcaballe
Differential Revision: https://reviews.llvm.org/D92702
Update all reference from the specification to the new OpenACC 3.1
document.
Reviewed By: SouraVX
Differential Revision: https://reviews.llvm.org/D92120
Add an option to pass the number of worker threads to select the number of async regions for parallel for transformation.
```
std::unique_ptr<OperationPass<FuncOp>> createAsyncParallelForPass(int numWorkerThreads);
```
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D92835
This reverts commit 02c9050155.
Painted myself into a corner with -Wvirtual_overload, private access, and final.
Differential Revision: https://reviews.llvm.org/D92855
This is needed so a listener hears all changes during the dialect
conversion to allow correct rollbacks upon failure.
Differential Revision: https://reviews.llvm.org/D92685
In RewritePattern, only expose `matchAndRewrite` as a public function. `match` can be protected (but needs to be protected because we want to call it from an override of `matchAndRewrite`). `rewrite` can be private.
For classes deriving from RewritePattern, all 3 functions can be private.
Side note: I didn't understand the need for the `using RewritePattern::matchAndRewrite` in derived classes, and started poking around. They are gone now, and I think the result is (only very slightly) cleaner.
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D92670
This commit adds initial support for SPIR-V OpSpecConstantOp
instruction. The following is introdcued:
- A new `spv.specConstantOperation` operation consisting of a single
region and of 2 operations within that regions (more details in the
docs of the op itself).
- A new `spv.yield` instruction that acts a terminator for
`spv.specConstantOperation`.
For now, the generic form of the new op is supported (i.e. no custom
parsing or printing). This will be done in a follow up patch.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D92232
Add a pass option to control the number of nested parallel loops produced by
the parallelization passes. This is useful to build end-to-end passes targeting
systems that don't need multiple parallel dimensions (e.g., CPUs typically need
only one).
Reviewed By: wsmoses, chelini
Differential Revision: https://reviews.llvm.org/D92765
The existing implementation of the affine parallelization silently copies over
the lower and upper bound maps from affine.for to affine.parallel. However, the
semantics of these maps differ between these two ops: in affine.for, a max(min)
of results is taken for the lower(upper) bound; in affine.parallel, multiple
induction variables can be defined an each result corresponds to one induction
variable. Thus the existing implementation could generate invalid IR or IR that
passes the verifier but has different semantics than the original code. Fix the
parallelization utility to emit dedicated min/max operations before the
affine.parallel in such cases. Disallow parallelization if min/max would have
been in an operation without the AffineScope trait, e.g., in another loop,
since the result of these operations is not considered a valid affine dimension
identifier and may not be properly handled by the affine analyses.
Reviewed By: wsmoses
Differential Revision: https://reviews.llvm.org/D92763
The default exception handling isn't very user friendly and does not
point accurately to the issue. Instead we can indicate which of the
operands isn't valid and provide contextual information in the error
message.
Differential Revision: https://reviews.llvm.org/D92710
After bufferization, the backend has much more trouble hoisting loop invariant
loads from the loops generated by the sparse compiler. Therefore, this is done
during sparse code generation. Note that we don't bother hoisting derived
invariant expressions on SSA values, since the backend does that very well.
Still TBD: scalarize reductions to avoid load-add-store cycles
Reviewed By: penpornk
Differential Revision: https://reviews.llvm.org/D92534
Add support to normalize affine.for ops i.e., convert the lower bound to zero
and loop step to one. The Upper bound is set to the trip count of the loop.
The exact value of loopIV is calculated just inside the body of affine.for.
Currently loops with lower bounds having single result are supported. No such
restriction exists on upper bounds.
Differential Revision: https://reviews.llvm.org/D92233
Some Ops in OMP dialect have regions associated with them i.e
`ParallelOp` `MasterOp`. Lowering of these regions involves interfacing
with `OMPIRBuilder` using callbacks, yet there still exist opportunities
for sharing common code in between.
This patch factors out common code into a separate function and adds
support for lowering `MasterOp` using that. Lowering of `ParallelOp` is
also modified appropriately.
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D87247
Trailing objects are really nice for storing additional data inline with the main class, and is something that we heavily take advantage of for Operation(and many other classes). To get the address of the inline data you need to compute the address by doing some pointer arithmetic taking into account any objects stored before the object you want to access. Most classes keep the count of the number of objects, so this is relatively cheap to compute. This is not the case for results though, which have two different types(inline and trailing) that are not necessarily as cheap to compute as the count for other objects. This revision moves the storage for results to before the operation and stores them in reverse order. This allows for getting results to still be very fast given that they are never iterated directly in order, and also greatly improves the speed when accessing the other trailing objects of an operation(operands/regions/blocks/etc).
This reduced compile time when compiling a decently sized mlir module by about ~400ms, or 2.17s -> 1.76s.
Differential Revision: https://reviews.llvm.org/D92687
The check for formatting enum attributes was missing a call to get the base attribute, which is necessary to strip off the top-level OptionalAttr<> wrapper.
Differential Revision: https://reviews.llvm.org/D92713
Make UnrollVectorPattern inherit from RewritePattern instead of
OpRewritePattern so that we don't need to create many patterns when applying to
many different type of ops. Since we may want to apply the pattern to all
arithmetic op, it is more convenient to filter dynamically.
Differential Revision: https://reviews.llvm.org/D92635
- Instead of hardcoding the parameters and return types of 'inferReturnTypes', use the
InferTypeOpInterface trait to generate the method declaration.
- Fix InferTypeOfInterface to use fully qualified type for inferReturnTypes results.
Differential Revision: https://reviews.llvm.org/D92585
In the past, the reshape op can be folded only if the indexing map is
permutation in consumer's usage. We can relax to condition to be projected
permutation.
This patch still limits the fusion for scalar cases. Scalar case is a corner
case, because we need to decide where to put extra dims.
Reviewed By: mravishankar
Differential Revision: https://reviews.llvm.org/D92466
This is part of a larger refactoring the better congregates the builtin structures under the BuiltinDialect. This also removes the problematic "standard" naming that clashes with the "standard" dialect, which is not defined within IR/. A temporary forward is placed in StandardTypes.h to allow time for downstream users to replaced references.
Differential Revision: https://reviews.llvm.org/D92435
The definitions of ModuleOp and FuncOp are now within BuiltinOps.h, making the individual files obsolete.
Differential Revision: https://reviews.llvm.org/D92622
A separate AVX512 lowering pass does not compose well with the regular
vector lowering pass. As such, it is at risk of code duplication and
lowering inconsistencies. This change removes the separate AVX512 lowering
pass and makes it an "option" in the regular vector lowering pass
(viz. vector dialect "augmented" with AVX512 dialect).
Reviewed By: rriddle
Differential Revision: https://reviews.llvm.org/D92614
This was important when ModuleOp was the only top level operation, but that isn't necessarily the case anymore. This is one of the last remaining aspects of the infrastructure that is hardcoded to ModuleOp.
Differential Revision: https://reviews.llvm.org/D92605
This was a somewhat important restriction in the past when ModuleOp was distinctly the top-level container operation, as well as before the pass manager had support for running nested pass managers natively. With these two issues fading away, there isn't really a good reason to enforce that a ModuleOp is the thing running within a pass manager. As such, this revision removes the restriction and allows for users to pass in the name of the operation that the pass manager will be scheduled on.
The only remaining dependency on BuiltinOps from Pass after this revision is due to FunctionPass, which will be resolved in a followup revision.
Differential Revision: https://reviews.llvm.org/D92450
There isn't a good reason for anything within IR to specifically reference any of the builtin operations. The only place that had a good reason in the past was AsmPrinter, but the behavior there doesn't need to hardcode ModuleOp anymore.
Differential Revision: https://reviews.llvm.org/D92448
Add support for vectorization for linalg.generic representing element-wise ops.
Those are converted to transfer_read + vector ops + transfer_write.
Also re-organize the vectorization tests to be together.
Implementation derived from the work of @burmako, @agrue and
@fedelebron.
Differential Revision: https://reviews.llvm.org/D92540
This fixes the build on gcc5 toolchain where sizeof is required for
types used in SmallVector now.
This is a consequence of using std::is_trivially_copy_constructible
instead of the LLVM variant: https://reviews.llvm.org/D92543
This reduces the chances of segfault. While it is a good practice to ensure
robust custom printers, it is unfortunately common to have them crash on
invalid input.
Reviewed By: stellaraccident
Differential Revision: https://reviews.llvm.org/D92536
Memrefs with affine_map in the results of normalizable operation were
not normalized by `--normalize-memrefs` option. This patch normalizes
them.
Differential Revision: https://reviews.llvm.org/D88719
Extended promote buffers to stack pass to support dynamically shaped allocas.
The conversion is limited by the rank of the underlying tensor.
An option is added to the pass to adjust the given rank.
Differential Revision: https://reviews.llvm.org/D91969
Given that OpState already implicit converts to Operator*, this seems reasonable.
The alternative would be to add more functions to OpState which forward to Operation.
Reviewed By: rriddle, ftynse
Differential Revision: https://reviews.llvm.org/D92266
OpenMPIRBuilder::createParallel outlines the body region of the parallel
construct into a new function that accepts any value previously defined outside
the region as a function argument. This function is called back by OpenMP
runtime function __kmpc_fork_call, which expects trailing arguments to be
pointers. If the region uses a value that is not of a pointer type, e.g. a
struct, the produced code would be invalid. In such cases, make createParallel
emit IR that stores the value on stack and pass the pointer to the outlined
function instead. The outlined function then loads the value back and uses as
normal.
Reviewed By: jdoerfert, llitchev
Differential Revision: https://reviews.llvm.org/D92189
The test process of the ir_array_attributes.py depends on numpy. This patch checks numpy in Python bindings configuration.
- Add NumPy in find_package as a required component to check numpy.
- If numpy is found, print the version and include directory.
Differential Revision: https://reviews.llvm.org/D92276
PDL patterns are now supported via a new `PDLPatternModule` class. This class contains a ModuleOp with the pdl::PatternOp operations representing the patterns, as well as a collection of registered C++ functions for native constraints/creations/rewrites/etc. that may be invoked via the pdl patterns. Instances of this class are added to an OwningRewritePatternList in the same fashion as C++ RewritePatterns, i.e. via the `insert` method.
The PDL bytecode is an in-memory representation of the PDL interpreter dialect that can be efficiently interpreted/executed. The representation of the bytecode boils down to a code array(for opcodes/memory locations/etc) and a memory buffer(for storing attributes/operations/values/any other data necessary). The bytecode operations are effectively a 1-1 mapping to the PDLInterp dialect operations, with a few exceptions in cases where the in-memory representation of the bytecode can be more efficient than the MLIR representation. For example, a generic `AreEqual` bytecode op can be used to represent AreEqualOp, CheckAttributeOp, and CheckTypeOp.
The execution of the bytecode is split into two phases: matching and rewriting. When matching, all of the matched patterns are collected to avoid the overhead of re-running parts of the matcher. These matched patterns are then considered alongside the native C++ patterns, which rewrite immediately in-place via `RewritePattern::matchAndRewrite`, for the given root operation. When a PDL pattern is matched and has the highest benefit, it is passed back to the bytecode to execute its rewriter.
Differential Revision: https://reviews.llvm.org/D89107
- Change InferTypeOpInterface::inferResultTypes to use fully qualified types matching
the ones generated by genTypeInterfaceMethods, so the redundancy can be detected.
- Move genTypeInterfaceMethods() before genOpInterfaceMethods() so that the
inferResultTypes method generated by genTypeInterfaceMethods() takes precedence
over the declaration that might be generated by genOpInterfaceMethods()
- Modified an op in the test dialect to exercise this (the modified op would fail to
generate valid C++ code due to duplicate inferResultTypes methods).
Differential Revision: https://reviews.llvm.org/D92414
ExecutionEngine/LLJIT do not run globals destructors in loaded dynamic libraries when destroyed, and threads managed by ThreadPool can race with program termination, and it leads to segfaults.
TODO: Re-enable threading after fixing a problem with destructors, or removing static globals from dynamic library.
Differential Revision: https://reviews.llvm.org/D92368
- Address TODO in scf-bufferize: the argument materialization issue is
now fixed and the code is now in Transforms/Bufferize.cpp
- Tighten up finalizing-bufferize to avoid creating invalid IR when
operand types potentially change
- Tidy up the testing of func-bufferize, and move appropriate tests
to a new finalizing-bufferize.mlir
- The new stricter checking in finalizing-bufferize revealed that we
needed a DimOp conversion pattern (found when integrating into npcomp).
Previously, the converion infrastructure was blindly changing the
operand type during finalization, which happened to work due to
DimOp's tensor/memref polymorphism, but is generally not encouraged
(the new pattern is the way to tell the conversion infrastructure that
it is legal to change that type).
The InlineAsmOp mirrors the underlying LLVM semantics with a notable
exception: the embedded `asm_string` is not allowed to define or reference
any symbol or any global variable: only the operands of the op may be read,
written, or referenced.
Attempting to define or reference any symbol or any global behavior is
considered undefined behavior at this time.
The asm dialect syntax is currently specified with an integer (0 [default] for the "att dialect", 1 for the intel dialect) to circumvent the ODS limitation on string enums.
Translation to LLVM is provided and raises the fact that the asm constraints string must be well-formed with respect to in/out operands. No check is performed on the asm_string.
An InlineAsm instruction in LLVM is a special call operation to a function that is constructed on the fly.
It does not fit the current model of MLIR calls with symbols.
As a consequence, the current implementation constructs the function type in ModuleTranslation.cpp.
This should be refactored in the future.
The mlir-cpu-runner is augmented with the global initialization of the X86 asm parser to allow proper execution in JIT mode. Previously, only the X86 asm printer was initialized.
Differential revision: https://reviews.llvm.org/D92166
* If ODS redefines this, it is fine, but I have found this accessor to be universally useful in the old npcomp bindings and I'm closing gaps that will let me switch.
Differential Revision: https://reviews.llvm.org/D92287
* Add capsule get/create for Attribute and Type, which already had capsule interop defined.
* Add capsule interop and get/create for Location.
* Add Location __eq__.
* Use get() and implicit cast to go from PyAttribute, PyType, PyLocation to MlirAttribute, MlirType, MlirLocation (bundled with this change because I didn't want to continue the pattern one more time).
Differential Revision: https://reviews.llvm.org/D92283
Op with mapping from ops to corresponding shape functions for those op
in the library and mechanism to associate shape functions to functions.
The mapping of operand to shape function is kept separate from the shape
functions themselves as the operation is associated to the shape
function and not vice versa, and one could have a common library of
shape functions that can be used in different contexts.
Use fully qualified names and require a name for shape fn lib ops for
now and an explicit print/parse (based around the generated one & GPU
module op ones).
This commit reverts d9da4c3e73. Fixes
missing headers (don't know how that was working locally).
Differential Revision: https://reviews.llvm.org/D91672
Op with mapping from ops to corresponding shape functions for those op
in the library and mechanism to associate shape functions to functions.
The mapping of operand to shape function is kept separate from the shape
functions themselves as the operation is associated to the shape
function and not vice versa, and one could have a common library of
shape functions that can be used in different contexts.
Use fully qualified names and require a name for shape fn lib ops for
now and an explicit print/parse (based around the generated one & GPU
module op ones).
Differential Revision: https://reviews.llvm.org/D91672
A splat attribute have a single element during printing so we should
treat it as such when we decide if we elide it or not based on the flag
intended to elide large attributes.
Reviewed By: rriddle, mehdi_amini
Differential Revision: https://reviews.llvm.org/D92165
Many pages have had their titles renamed over time,
causing broken links to spread throughout the documentation.
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D92093
The ops are very similar to the std variants, but support async GPU execution.
gpu.alloc does not currently support an alignment attribute, and the new ops do not have
canonicalizers/folders like their std siblings do.
Reviewed By: herhut
Differential Revision: https://reviews.llvm.org/D91698
The rewrite logic has an optimization to drop a cast operation after
rewriting block arguments if the cast operation has no users. This is
unsafe as there might be a pending rewrite that replaced the cast operation
itself and hence would trigger a second free.
Instead, do not remove the casts and leave it up to a later canonicalization
to do so.
Differential Revision: https://reviews.llvm.org/D92184
This change is required so that bufferization can properly identify
the linalg.yield as a terminator with an associated parent op.
Differential Revision: https://reviews.llvm.org/D92173
This enables partial bufferization that includes function signatures. To test this, this
change also makes the func-bufferize partial and adds a dedicated finalizing-bufferize pass.
Differential Revision: https://reviews.llvm.org/D92032
This change gives sparse compiler clients more control over selecting
individual types for the pointers and indices in the sparse storage schemes.
Narrower width obviously results in smaller memory footprints, but the
range should always suffice for the maximum number of entries or index value.
Reviewed By: penpornk
Differential Revision: https://reviews.llvm.org/D92126
Adding missing custom builders for AffineVectorLoadOp & AffineVectorStoreOp. In practice, it is difficult to correctly construct these ops without these builders (because the AffineMap is not included at construction time).
Differential Revision: https://reviews.llvm.org/D86380
This CL adds the ability to request different parallelization strategies
for the generate code. Every "parallel" loop is a candidate, and converted
to a parallel op if it is an actual for-loop (not a while) and the strategy
allows dense/sparse outer/inner parallelization.
This will connect directly with the work of @ezhulenev on parallel loops.
Still TBD: vectorization strategy
Reviewed By: penpornk
Differential Revision: https://reviews.llvm.org/D91978
CHECK_* directives for message() where added in Cmake 3.17, LLVM
requires 3.14 as minimum so they may not be intepreted correctly and
just print "CHECK_*" into the message stream. Replace them with STATUS.
Reviewed By: stellaraccident
Differential Revision: https://reviews.llvm.org/D91959
SameOperandsAndResultShape and ElementwiseMappable have similar
verification, but in general neither is strictly redundant with the
other.
Examples:
- SameOperandsAndResultShape allows
`"foo"(%0) : tensor<2xf32> -> tensor<?xf32> but ElementwiseMappable
does not.
- ElementwiseMappable allows
`select %scalar_pred, %true_tensor, %false_tensor` but
SameOperandsAndResultShape does not.
SameOperandsAndResultShape is redundant with ElementwiseMappable when
we can prove that the mixed scalar/non-scalar case cannot happen. In
those situations, `ElementwiseMappable & SameOperandsAndResultShape ==
ElementwiseMappable`:
- Ops with 1 operand: the case of mixed scalar and non-scalar operands
cannot happen since there is only one operand.
- When SameTypeOperands is also present, the mixed scalar/non-scalar
operand case cannot happen.
Differential Revision: https://reviews.llvm.org/D91396
Generalizes invariant handling to anything defined outside the Linalg op
(parameters and SSA computations). Fixes bug that was using parameter number
as tensor number.
Reviewed By: penpornk
Differential Revision: https://reviews.llvm.org/D91985
Introduce a conversion pass from SCF parallel loops to OpenMP dialect
constructs - parallel region and workshare loop. Loops with reductions are not
supported because the OpenMP dialect cannot model them yet.
The conversion currently targets only one level of parallelism, i.e. only
one top-level `omp.parallel` operation is produced even if there are nested
`scf.parallel` operations that could be mapped to `omp.wsloop`. Nested
parallelism support is left for future work.
Reviewed By: kiranchandramohan
Differential Revision: https://reviews.llvm.org/D91982
Print part of an op of the form:
```
<optional-offset-prefix>`[` offset-list `]`
<optional-size-prefix>`[` size-list `]`
<optional-stride-prefix>[` stride-list `]`
```
Also address some leftover nits.
Differential revision: https://reviews.llvm.org/D92031
Parse trailing part of an op of the form:
```
<optional-offset-prefix>`[` offset-list `]`
<optional-size-prefix>`[` size-list `]`
<optional-stride-prefix>[` stride-list `]`
```
Each entry in the offset, size and stride list either resolves to an integer
constant or an operand of index type.
Constants are added to the `result` as named integer array attributes with
name `OffsetSizeAndStrideOpInterface::getStaticOffsetsAttrName()` (resp.
`getStaticSizesAttrName()`, `getStaticStridesAttrName()`).
Append the number of offset, size and stride operands to `segmentSizes`
before adding it to `result` as the named attribute:
`OpTrait::AttrSizedOperandSegments<void>::getOperandSegmentSizeAttr()`.
Offset, size and stride operands resolution occurs after `preResolutionFn`
to give a chance to leading operands to resolve first, after parsing the
types.
```
ParseResult parseOffsetsSizesAndStrides(
OpAsmParser &parser, OperationState &result, ArrayRef<int> segmentSizes,
llvm::function_ref<ParseResult(OpAsmParser &, OperationState &)>
preResolutionFn = nullptr,
llvm::function_ref<ParseResult(OpAsmParser &)> parseOptionalOffsetPrefix =
nullptr,
llvm::function_ref<ParseResult(OpAsmParser &)> parseOptionalSizePrefix =
nullptr,
llvm::function_ref<ParseResult(OpAsmParser &)> parseOptionalStridePrefix =
nullptr);
```
Differential revision: https://reviews.llvm.org/D92030
MLIR C API use the `MlirStringRef` instead of `const char *` for the string type now. This patch sync the Python bindings with the C API modification.
Differential Revision: https://reviews.llvm.org/D92007
* Was missed in the initial submission and is required for a ConstantLike op.
* Also adds a materializeConstant hook to preserve it.
* Tightens up the argument constraint on tosa.const to match what is actually legal.
Differential Revision: https://reviews.llvm.org/D92040
This revision will make it easier to create new ops base on the strided memref abstraction outside of the std dialect.
OffsetSizeAndStrideOpInterface is an interface for ops that allow specifying mixed dynamic and static offsets, sizes and strides variadic operands.
Ops that implement this interface need to expose the following methods:
1. `getArrayAttrRanks` to specify the length of static integer
attributes.
2. `offsets`, `sizes` and `strides` variadic operands.
3. `static_offsets`, resp. `static_sizes` and `static_strides` integer
array attributes.
The invariants of this interface are:
1. `static_offsets`, `static_sizes` and `static_strides` have length
exactly `getArrayAttrRanks()`[0] (resp. [1], [2]).
2. `offsets`, `sizes` and `strides` have each length at most
`getArrayAttrRanks()`[0] (resp. [1], [2]).
3. if an entry of `static_offsets` (resp. `static_sizes`,
`static_strides`) is equal to a special sentinel value, namely
`ShapedType::kDynamicStrideOrOffset` (resp. `ShapedType::kDynamicSize`,
`ShapedType::kDynamicStrideOrOffset`), then the corresponding entry is
a dynamic offset (resp. size, stride).
4. a variadic `offset` (resp. `sizes`, `strides`) operand must be present
for each dynamic offset (resp. size, stride).
This interface is useful to factor out common behavior and provide support
for carrying or injecting static behavior through the use of the static
attributes.
Differential Revision: https://reviews.llvm.org/D92011
Use the correct interface base type name when generating attribute interfaces
with TabeGen.
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D92023
1. Move ThreadPool ownership to the runtime, and wait for the async tasks completion in the destructor.
2. Remove MLIR_ASYNCRUNTIME_EXPORT from method definitions because they are unnecessary in .cpp files, as only function declarations need to be exported, not their definitions.
3. Fix concurrency bugs in group emplace and potential use-after-free in token emplace.
Tested internally 10k runs in `async.mlir` and `async-group.mlir`.
Fixed: https://bugs.llvm.org/show_bug.cgi?id=48267
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D91988
This file is intended to be included by other files, including
out-of-tree dialects, and makes more sense in `include` than in `lib`.
Depends On D91652
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D91961
Attributes represent additional data about an operation and are intended to be
modifiable during the lifetime of the operation. In the dialect-specific Python
bindings, attributes are exposed as properties on the operation class. Allow
for assigning values to these properties. Also support creating new and
deleting existing attributes through the generic "attributes" property of an
operation. Any validity checking must be performed by the op verifier after the
mutation, similarly to C++. Operations are not invalidated in the process: no
dangling pointers can be created as all attributes are owned by the context and
will remain live even if they are not used in any operation.
Introduce a Python Test dialect by analogy with the Test dialect and to avoid
polluting the latter with Python-specific constructs. Use this dialect to
implement a test for the attribute access and mutation API.
Reviewed By: stellaraccident, mehdi_amini
Differential Revision: https://reviews.llvm.org/D91652
It is a simple conversion that only requires to change the region argument
types, generalize it from ParallelOp.
Reviewed By: kiranchandramohan
Differential Revision: https://reviews.llvm.org/D91989
While this makes the unit tests a bit more verbose, this simplifies the creation of bindings because only the bidirectional mapping between the host language's string type and MlirStringRef need to be implemented.
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D91905
Exposing some utility functions from Linalg to allow for promotion of
fused views outside of the core tile+fuse logic.
This is an alternative to patch D91322 which adds the promotion logic
to the tileAndFuse method. Downside with that approach is that it is
not easily customizable based on needs.
Differential Revision: https://reviews.llvm.org/D91503
Enhance the tile+fuse logic to allow fusing a sequence of operations.
Make sure the value used to obtain tile shape is a
SubViewOp/SubTensorOp. Current logic used to get the bounds of loop
depends on the use of `getOrCreateRange` method on `SubViewOp` and
`SubTensorOp`. Make sure that the value/dim used to compute the range
is from such ops. This fix is a reasonable WAR, but a btter fix would
be to make `getOrCreateRange` method be a method of `ViewInterface`.
Differential Revision: https://reviews.llvm.org/D90991
Previously, there was no way to add context to the diagnostic engine via the C API. Adding this ability makes it much easier to reason about memory ownership, particularly in reference-counted languages such as Swift. There are more details in the review comments.
Reviewed By: ftynse, mehdi_amini
Differential Revision: https://reviews.llvm.org/D91738