Dialect conversion infrastructure supports 1->N type conversions by requiring
individual conversions to provide facilities to generate operations
retrofitting N values into 1 of the original type when N > 1. This
functionality can also be used to materialize explicit "cast"-like operations,
but it did not support 1->1 type conversions until now. Modify TypeConverter to
support materialization of cast operations for 1-1 conversions.
This also makes materialization specification more extensible following the
same pattern as type conversions. Instead of overloading a virtual function,
users or subclasses of TypeConversion can now register type-specific
materialization callbacks that will be called in order for the given type.
Differential Revision: https://reviews.llvm.org/D79729
One header guard was overlooked when renaming LoopOps to SCF, rename it.
Also drop two unused macros, one of which referred to LoopOp (not "Ops",
hence the overlook).
The lower-to-affine-loops pass in chapters 5-7 of the Toy tutorial has
been creating affine loops, erasing their terminator and creating it
anew using a PatternRewriter instance to work around the fact that
implicit terminators were created without notifying the rewriter. Now
that has been fixed in 3ccf4a5bd1, remove the code erasing and
re-creating the terminators and rely on the default ones.
Add BufferAssignmentCallOpConverter as a pattern rewriter for Buffer
Placement. It matches the signature of the caller operation with the callee
after rewriting the callee with FunctionAndBlockSignatureConverter.
Differential Revision: https://reviews.llvm.org/D80785
Keeping in the affine.for to gpu.launch conversions, which should
probably be the affine.parallel to gpu.launch conversion as well.
Differential Revision: https://reviews.llvm.org/D80747
Summary:
MSVC does not seem to like certain forward declarations.
https://reviews.llvm.org/D80728 introduces an error where
seemingly unrelated .cpp files that include the .h
(but do not otherwise use the class that depends on the forward declaration).
Instead of forward declaration, include the full vector ops definition.
Subscribers: mehdi_amini, rriddle, jpienaar, shauheen, antiagainst, arpith-jacob, mgester, lucyrfox, aartbik, liufengdb, stephenneuendorffer, Joonsoo, grosul1, frgossen, Kayjukh, jurahul, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D80841
This revision replaces the load + vector.type_cast by appropriate vector transfer
operations. These play more nicely with other vector abstractions and canonicalization
patterns and lower to load/store with or without masks when appropriate.
Differential Revision: https://reviews.llvm.org/D80809
Summary:
Implemented the basic changes for defining master operation in OpenMP.
It uses the generic parser and printer.
Reviewed By: kiranchandramohan, ftynse
Differential Revision: https://reviews.llvm.org/D80689
This revision adds custom rewrites for patterns that arise during linalg structured
ops vectorization. These patterns allow the composition of linalg promotion,
vectorization and removal of redundant copies.
The patterns are voluntarily limited and restrictive atm.
More robust behavior will be implemented once more powerful side effect modeling and analyses are available on view/subview.
On the transfer_read side, the following pattern is rewritten:
```
%alloc = ...
[optional] %view = std.view %alloc ...
%subView = subview %allocOrView ...
[optional] linalg.fill(%allocOrView, %cst) ...
...
linalg.copy(%in, %subView) ...
vector.transfer_read %allocOrView[...], %cst ...
```
into
```
[unchanged] %alloc = ...
[unchanged] [optional] %view = std.view %alloc ...
[unchanged] [unchanged] %subView = subview %allocOrView ...
...
vector.transfer_read %in[...], %cst ...
```
On the transfer_write side, the following pattern is rewriten:
```
%alloc = ...
[optional] %view = std.view %alloc ...
%subView = subview %allocOrView...
...
vector.transfer_write %..., %allocOrView[...]
linalg.copy(%subView, %out)
```
Differential Revision: https://reviews.llvm.org/D80728
This utility factors out the machinery required to add iterArgs and yield values to an scf.ForOp.
Differential Revision: https://reviews.llvm.org/D80656
Buffer placement can now operates on functions that return buffers. These
buffers escape from the deallocation phase of buffer placement.
Differential Revision: https://reviews.llvm.org/D80696
Summary: Add a test to check if the standalone dialect is registered within standalone-opt. Similar to the mlir-opt commandline.mlir test.
Reviewers: Kayjukh, stephenneuendorffer
Reviewed By: Kayjukh
Subscribers: mehdi_amini, rriddle, jpienaar, shauheen, antiagainst, nicolasvasilache, arpith-jacob, mgester, lucyrfox, liufengdb, Joonsoo, grosul1, frgossen, jurahul, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D80764
https://reviews.llvm.org/D79246 introduces alignment propagation for vector transfer operations. Unfortunately, the alignment calculation is incorrect and can result in crashes.
This revision fixes the calculation by using the natural alignment of the memref elemental type, instead of the resulting vector type.
If more alignment is desired, it can be done in 2 ways:
1. use a proper vector.type_cast to transform a memref<axbxcxdxf32> into a memref<axbxvector<cxdxf32>> giving a natural alignment of vector<cxdxf32>
2. add an alignment attribute to vector transfer operations and propagate it.
With this change the alignment in the relevant tests goes down from 128 to 4.
Lastly, a few minor cleanups are performed and the custom `isMinorIdentityMap` is deprecated.
Differential Revision: https://reviews.llvm.org/D80734
operands of Generic ops.
Unit-extent dimensions are typically used for achieving broadcasting
behavior. The pattern added (along with canonicalization patterns
added previously) removes the use of unit-extent dimensions, and
instead uses a more canonical representation of the computation. This
new pattern is not added as a canonicalization for now since it
entails adding additional reshape operations. A pass is added to
exercise these patterns, along with an API entry to populate a
patterns list with these patterns.
Differential Revision: https://reviews.llvm.org/D79766
D80142 restructured MLIR-to-GPU-binary conversion to support multiple
targets. It also modified cmake files to link relevant LLVM components
in test/lib, which broke shared-library builds, and likely made the
conversions unusable outside mlir-opt (or other tools that link in test
library targets). Link these components to GPUCommon instead.
Differential Revision: https://reviews.llvm.org/D80739
This allows constructing operand adaptor from existing op (useful for commonalizing verification as I want to do in a follow up).
I also add ability to use member initializers for the generated adaptor constructors for convenience.
Differential Revision: https://reviews.llvm.org/D80667
The operand of `from_extent_tensor` is now of the same index type as the result
type of the inverse operation `to_extent_tensor`.
Differential Revision: https://reviews.llvm.org/D80283
Make ConvertKernelFuncToCubin pass to be generic:
- Rename to ConvertKernelFuncToBlob.
- Allow specifying triple, target chip, target features.
- Initializing LLVM backend is supplied by a callback function.
- Lowering process from MLIR module to LLVM module is via another callback.
- Change mlir-cuda-runner to adopt the revised pass.
- Add new tests for lowering to ROCm HSA code object (HSACO).
- Tests for CUDA and ROCm are kept in separate directories.
Differential Revision: https://reviews.llvm.org/D80142
The operation `num_elements` determines the number of elements for a given
shape.
That is the product of its dimensions.
Differential Revision: https://reviews.llvm.org/D80281
Add the two conversion operations `index_to_size` and `size_to_index` to the
shape dialect.
This facilitates the conversion of index types between the shape and the
standard dialect.
Differential Revision: https://reviews.llvm.org/D80280
Purely cosmetic change.
The operation implementations in `Shape.cpp` are now lexicographic order.
Differential Revision: https://reviews.llvm.org/D80277
This changes will catch error where C++ op are used without being
registered, either through creation with the OpBuilder or when trying to
cast to the C++ op.
Differential Revision: https://reviews.llvm.org/D80651
Summary:
Index is the proper type for storing shapes when constant folding, so
this fixes the previous code (which was using i64).
Differential Revision: https://reviews.llvm.org/D80600
I just spent a bunch of time debugging a mysterious bug that ended being due to my SmallVector getting passed to the Args&... overload instead of the MutableArrayRef overload, with disastrous results.
I appreciate the intent of this API, but for a function that does a bunch of unsafe casts, adding in potential overload confusion is just too much C++ footgun. If we end up needing this functionality, having something like a separate `packArgs(Args&...) -> SmallVector` overload would be preferable.
Turns out this API is unused and untested (even out of tree as far as I can tell, modulo the optional passing of no args to the other invoke as I fixed in this patch), so it's an easy fix -- just delete it and touch up the other overload.
Differential Revision: https://reviews.llvm.org/D80607
This allocation of a workgroup memory is lowered to a
spv.globalVariable. Only static size allocation with element type
being int or float is handled. The lowering does account for the
element type that are not supported in the lowered spv.module based on
the extensions/capabilities and adjusts the number of elements to get
the same byte length.
Differential Revision: https://reviews.llvm.org/D80411
Summary:
This includes a basic implementation for the OpenMP parallel
operation without a custom pretty-printer and parser.
The if, num_threads, private, shared, first_private, last_private,
proc_bind and default clauses are included in this implementation.
Currently the reduction clause is omitted as it is more complex and
requires analysis to see if we can share implementation with the loop
dialect. The allocate clause is also omitted.
A discussion about the design of this operation can be found here:
https://llvm.discourse.group/t/openmp-parallel-operation-design-issues/686
The current OpenMP Specification can be found here:
https://www.openmp.org/wp-content/uploads/OpenMP-API-Specification-5.0.pdf
Co-authored-by: Kiran Chandramohan <kiran.chandramohan@arm.com>
Reviewers: jdoerfert
Subscribers: mgorny, yaxunl, kristof.beyls, guansong, mehdi_amini, rriddle, jpienaar, shauheen, antiagainst, nicolasvasilache, arpith-jacob, mgester, lucyrfox, aartbik, liufengdb, Joonsoo, grosul1, frgossen, Kayjukh, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D79410
Take advantage of equality constrains to generate the type inference interface.
This is used for equality and trivially built types. The type inference method
is only generated when no type inference trait is specified already.
This reorders verification that changes some test error messages.
Differential Revision: https://reviews.llvm.org/D80484
Now that OpBuilder is available in `build` functions, it becomes possible to
populate the "then" and "else" regions directly when building the "if"
operation. This is desirable in more structured forms of builders, especially
in when conditionals are mixed with loops. Provide new `build` APIs taking
callbacks for body constructors, similarly to scf::ForOp, and replace more
clunky edsc::BlockBuilder uses with these. The original APIs remain available
and go through the new implementation.
Differential Revision: https://reviews.llvm.org/D80527
alloc/dealloc/copies.
Add options to LinalgPromotion to use callbacks for implementating the
allocation, deallocation of buffers used for the promoted subviews,
and to copy data into and from the original subviews to the allocated
buffers.
Also some misc. cleanup of the code.
Differential Revision: https://reviews.llvm.org/D80365
Modifying the loop nest builder for generating scf.parallel loops to
not generate scf.parallel loops for non-parallel iterator types in
Linalg operations. The existing implementation incorrectly generated
scf.parallel for all tiled loops. It is rectified by refactoring logic
used while lowering to loops that accounted for this.
Differential Revision: https://reviews.llvm.org/D80188
Summary:
This op extracts an extent from a shape.
This also is the first op which constant folds to shape.const_size,
which revealed that shape.const_size needs a folder (ConstantLike ops
seem to always need folders for the constant folding infra to work).
Differential Revision: https://reviews.llvm.org/D80394
This revision expands the types of vector contractions that can be lowered to vector.outerproduct.
All 8 permutation cases are support.
The idiomatic manipulation of AffineMap written declaratively makes this straightforward.
In the process a bug with the vector.contract verifier was uncovered.
The vector shape verification part of the contract op is rewritten to use AffineMap composition.
One bug in the vector `ops.mlir` test is fixed and a new case not yet captured is added
to the vector`invalid.mlir` test.
Differential Revision: https://reviews.llvm.org/D80393
Summary:
Add DynamicMemRefType which can reference one of the statically ranked StridedMemRefType or a UnrankedMemRefType so that runner utils only need to be implemented once.
There is definitely room for more clean up and unification, but I will keep that for follow-ups.
Reviewers: nicolasvasilache
Reviewed By: nicolasvasilache
Subscribers: mehdi_amini, rriddle, jpienaar, shauheen, antiagainst, nicolasvasilache, arpith-jacob, mgester, lucyrfox, liufengdb, stephenneuendorffer, Joonsoo, grosul1, frgossen, Kayjukh, jurahul, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D80513
This revision adds the additional lowering and exposes the patterns at a finer granularity for better programmatic reuse. The unit test makes use of the finer grained pattern for simpler checks.
As the ContractionOpLowering is exposed programmatically, cleanup opportunities appear and static class methods are turned into free functions with static visibility.
Differential Revision: https://reviews.llvm.org/D80375
This still allows `if (value)` while requiring an explicit cast when not
in a boolean context. This means things like `std::set<Value>` will no
longer compile.
Differential Revision: https://reviews.llvm.org/D80497
* Enables using with more variadic sized operands;
* Generate convenience accessors for attributes;
- The accessor are named the same as their name in ODS and returns attribute
type (not convenience type) and no derived attributes.
This is first step to changing adapter to support verifying argument
constraints before the op is even created. This does not change the name of
adaptor nor does it require it except for ops with variadic operands to keep this change smaller.
Considered creating separate adapter but decided against that given operands also require attributes in general (and definitely for verification of operands and attributes).
Differential Revision: https://reviews.llvm.org/D80420
Enable inset/extract/construct composite ops as well as access chain for
cooperative matrix. ConstantComposite requires more change and will be done in
a separate patch. Also fix the getNumElements function for coopMatrix per
feedback from Jeff Bolz. The number of element is implementation dependent so
it cannot be known at compile time.
Differential Revision: https://reviews.llvm.org/D80321
Adds support for cooperative matrix support for arithmetic and cast
instructions. It also adds cooperative matrix store, muladd and matrixlength
instructions which are part of the extension.
Differential Revision: https://reviews.llvm.org/D80181
Due to similar APIs between CUDA and ROCm (HIP),
ConvertGpuLaunchFuncToCudaCalls pass could be used on both platforms with some
refactoring.
In this commit:
- Migrate ConvertLaunchFuncToCudaCalls from GPUToCUDA to GPUCommon, and rename.
- Rename runtime wrapper APIs be platform-neutral.
- Let GPU binary annotation attribute be specifiable as a PassOption.
- Naming changes within the implementation and tests.
Subsequent patches would introduce ROCm-specific tests and runtime wrapper
APIs.
Differential Revision: https://reviews.llvm.org/D80167
This reverts commit cdb6f05e2d.
The build is broken with:
You have called ADD_LIBRARY for library obj.MLIRGPUtoCUDATransforms without any source files. This typically indicates a problem with your CMakeLists.txt file
Due to similar APIs between CUDA and ROCm (HIP),
ConvertGpuLaunchFuncToCudaCalls pass could be used on both platforms with some
refactoring.
In this commit:
- Migrate ConvertLaunchFuncToCudaCalls from GPUToCUDA to GPUCommon, and rename.
- Rename runtime wrapper APIs be platform-neutral.
- Let GPU binary annotation attribute be specifiable as a PassOption.
- Naming changes within the implementation and tests.
Subsequent patches would introduce ROCm-specific tests and runtime wrapper
APIs.
Differential Revision: https://reviews.llvm.org/D80167
The subview semantics changes recently to allow for more natural
representation of constant offsets and strides. The legalization of
subview op for lowering to SPIR-V needs to account for this.
Also change the linearization to use the strides from the affine map
of a memref.
Differential Revision: https://reviews.llvm.org/D80270
Summary:
Previously, the only support partial lowering from vector transfers to SCF was
going through loops. This requires a dedicated allocation and extra memory
roundtrips because LLVM aggregates cannot be indexed dynamically (for more
details see the [deep-dive](https://mlir.llvm.org/docs/Dialects/Vector/#deeperdive)).
This revision allows specifying full unrolling which removes this additional roundtrip.
This should be used carefully though because full unrolling will spill, negating the
benefits of removing the interim alloc in the first place.
Proper heuristics are left for a later time.
Differential Revision: https://reviews.llvm.org/D80100
The SingleBlockImplicitTerminator op trait provides a function
`ensureRegionTerminator` that injects an appropriate terminator into the block
if necessary, which is used during operation constructing and parsing.
Currently, this function directly modifies the IR using low-level APIs on
Operation and Block. If this function is called from a conversion pattern,
these manipulations are not reflected in the ConversionPatternRewriter and thus
cannot be undone or, worse, lead to tricky memory errors and malformed IR.
Change `ensureRegionTerminator` to take an instance of `OpBuilder` instead of
`Builder`, and use it to construct the block and the terminator when required.
Maintain overloads taking an instance of `Builder` and creating a simple
`OpBuilder` to use in parsers, which don't have an `OpBuilder` and cannot
interact with the dialect conversion mechanism. This change was one of the
reasons to make `<OpTy>::build` accept an `OpBuilder`.
Differential Revision: https://reviews.llvm.org/D80138
Originally, the SCFToStandard conversion only declared Ops from the Standard
dialect as legal after conversion. This is undesirable as it would fail the
conversion if the SCF ops contained ops from any other dialect. Furthermore,
this would be problematic for progressive lowering of `scf.parallel` to
`scf.for` after `ensureRegionTerminator` is made aware of the pattern rewriting
infrastructure because it creates temporary `scf.yield` operations declared
illegal. Change the legalization target to declare any op other than `scf.for`,
`scf.if` and `scf.parallel` legal.
Differential Revision: https://reviews.llvm.org/D80137
Multiple places in the code base were erasing Blocks or operations in them
using in-place modifications (`Block::erase` or `Block::clear`) unknown to
ConversionPatternRewriter. These operations could not be undone if the pattern
failed and could lead to inconsistent in-memory state of the IR with dangling
pointers. Use `ConversionPatternRewriter::eraseOp` and `::eraseBlock` instead.
Differential Revision: https://reviews.llvm.org/D80136
PatternRewriter has support for erasing a Block from its parent region, but
this feature has not been implemented for ConversionPatternRewriter that needs
to keep track of and be able to undo block actions. Introduce support for
undoing block erasure in the ConversionPatternRewriter by marking all the ops
it contains for erasure and by detaching the block from its parent region. The
detached block is stored in the action description and is not actually deleted
until the rewrites are applied.
Differential Revision: https://reviews.llvm.org/D80135
Dialect conversion infrastructure may roll back op creation by erasing the
operations in the reverse order of their creation. While this guarantees uses
of values will be deleted before their definitions, this does not guarantee
that a parent operation will not be deleted before its child. (This may happen
in case of block inlining or if child operations, such as terminators, are
created in the parent's `build` function before the parent itself.) Handle the
parent/child relationship between ops by removing all child ops from the blocks
before erasing the parent. The child ops remain live, detached from a block,
and will be safely destroyed in their turn, which may come later than that of
the parent.
Differential Revision: https://reviews.llvm.org/D80134
When creating temporary `scf.for` loops in `toy.print` lowering, the block
insertion point was erronously set up to the beginning of the block rather than
to its end, contradicting the comment just above the insertion point change.
The code was nevertheless operational because `scf.for` was setting up its
`scf.yield` terminator in an opaque to the pattern rewriting infrastructure
way. Now that it is about to change, the problem would have been exposed and
lead to conversion failures.
Differential Revision: https://reviews.llvm.org/D80133
Summary:
This revision refactors the Linalg tiling pass to be written as pattern applications and retires the use of the folder in Linalg tiling.
In the early days, tiling was written as a pass that would create (partially) folded and canonicalized operations on the fly for better composability.
As this evolves towards composition of patterns, the pass-specific folder is counter-productive and is retired.
The tiling options struct evolves to take a tile size creation function which allows materializing tile sizes on the fly (in particular constant tile sizes). This plays better with folding and DCE.
With the folder going away in Tiling, the check on whether subviews are the same in linalg fusion needs to be more robust. This revision also implements such a check.
In the current form, there are still some canonicalizations missing due to AffineMin/Max ops fed by scf::ForOp. These will be improved at a later time.
Differential Revision: https://reviews.llvm.org/D80267
Summary:
Additionally, this adds traits and builder methods to AssumingYieldOp
and names the input witness to the AssumingOp.
Differential Revision: https://reviews.llvm.org/D80187
MLIR tests in "mlir/test/mlir-cpu-runner" fails in SystemZ (z14) because
of incompatible datalayout error. This patch fixes it by setting host
CPU name in createTargetMachine()
Differential Revision: https://reviews.llvm.org/D80130
This is consistent to the other methods of the class, as well as
AffineExpr::replaceDimsAndSymbols.
Differential Revision: https://reviews.llvm.org/D80266
This should fix the error ```
VectorToSCF.cpp:238:62: error: specialization of 'template<class
ConcreteOp> mlir::LogicalResult
{anonymous}::NDTransferOpHelper<ConcreteOp>::doReplace()' in different
namespace
```
Add a new type to SPIRV dialect for cooperative matrix and add new op for
cooperative matrix load. This is missing most instructions to support
cooperative matrix extension but this is a stop-gap patch to avoid creating big
review.
Differential Revision: https://reviews.llvm.org/D80043
This patch introduces interfaces for read and write ops with affine
restrictions. I used `read`/`write` intead of `load`/`store` for the
interfaces so that they can also be implemented by dma ops.
For now, they are only implemented by affine.load, affine.store,
affine.vector_load and affine.vector_store.
For testing purposes, this patch also migrates affine loop fusion and
required analysis to use the new interfaces. No other changes are made
beyond that.
Co-authored-by: Alex Zinenko <zinenko@google.com>
Reviewed By: bondhugula, ftynse
Differential Revision: https://reviews.llvm.org/D79829
Enclose verifier code for AttrSizedOperandSegments and AttrSizedResultSegments
in a nested code block to avoid symbol collision.
Differential Revision: https://reviews.llvm.org/D80250
Using LLVM components in LINK_LIBS means that the mechanisms for
replacing component dependencies with libLLVM.so break. Try to catch
this incorrect usage up front, instead of waiting until later when we
get difficult to understand runtime errors from incorrectly linked
libraries.
Differential Revision: https://reviews.llvm.org/D80103
Summary:
This is a basic op needed for creating shapes from SSA values
representing the extents.
Differential Revision: https://reviews.llvm.org/D79833
Summary:
Previously, after applying the mask, a negative number would convert to a
positive number because the sign flag was forgotten. This patch adds two more
shift operations to do the sign extension. This assumes that we're using two's
complement.
This patch applies sign extension unconditionally when loading a unspported integer width, and it relies the pattern to do the casting because the signedness semantic is carried by operator itself.
Differential Revision: https://reviews.llvm.org/D79753
Making these two converters more generic. FunctionAndBlockSignatureConverter now
moves only memref results (after type conversion) to the function argument and
keeps other legal function results unchanged. NonVoidToVoidReturnOpConverter is
renamed to NoBufferOperandsReturnOpConverter. It removes only the buffer
operands from the operands of the converted ReturnOp and inserts CopyOps to copy
each buffer to the target function argument.
Differential Revision: https://reviews.llvm.org/D79329
Thanks to a recent change that made `::build` functions take an instance of
`OpBuilder`, it is now possible to build operations within a region attached to
the operation about to be created. Exercise this on `scf::ForOp` by taking a
callback that populates the loop body while the loop is being created.
Additionally, provide helper functions to build perfect nests of `ForOp`s,
with support for iteration arguments. These functions provide the same
functionality as EDSC LoopNestBuilder with simpler implementation, without
relying on edsc::ScopedContext, and using `OpBuilder` in an unambiguous way.
Compatibility functions for EDSC are provided, but may be removed in the
future.
Differential Revision: https://reviews.llvm.org/D79688
Implemented tangent op from SPIR-V's GLSL extended instruction set.
Added a round-trip and serialization/deserialization tests for the op.
Differential Revision: https://reviews.llvm.org/D80152
Summary:
This patch adds support for flush operation in OpenMP dialect and translation of this construct to LLVM IR.
The OpenMP IRBuilder is used for this translation.
The patch includes code changes and testcase modifications.
Reviewed By: ftynse, kiranchandramohan
Differential Revision: https://reviews.llvm.org/D79937
Summary: This revision adds support for assembly formats with optional attributes. It elides optional attributes that are part of the syntax from the attribute dictionary.
Reviewers: ftynse, Kayjukh
Reviewed By: ftynse, Kayjukh
Subscribers: mehdi_amini, rriddle, jpienaar, shauheen, antiagainst, nicolasvasilache, arpith-jacob, mgester, lucyrfox, liufengdb, stephenneuendorffer, Joonsoo, grosul1, frgossen, jurahul, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D80113
For now the promoted buffer is indexed using the `full view`. The full view might be
slightly bigger than the partial view (which is accounting for boundaries).
Unfortunately this does not compose easily with other transformations when multiple buffers
with shapes related to each other are involved.
Take `linalg.matmul A B C` (with A of size MxK, B of size KxN and C of size MxN) and suppose we are:
- Tiling over M by 100
- Promoting A only
This is producing a `linalg.matmul promoted_A B subview_C` where `promoted_A` is a promoted buffer
of `A` of size (100xK) and `subview_C` is a subview of size mxK where m could be smaller than 100 due
to boundaries thus leading to a possible incorrect behavior.
We propose to:
- Add a new parameter to the tiling promotion allowing to enable the use of the full tile buffer.
- By default all promoted buffers will be indexed by the partial view.
Note that this could be considered as a breaking change in comparison to the way the tiling promotion
was working.
Differential Revision: https://reviews.llvm.org/D79927
Summary:
Vector transfer ops semantic is extended to allow specifying a per-dimension `masked`
attribute. When the attribute is false on a particular dimension, lowering to LLVM emits
unmasked load and store operations.
Differential Revision: https://reviews.llvm.org/D80098
Summary:
This revision makes the use of vector transfer operatons more idiomatic by
allowing to omit and inferring the permutation_map.
Differential Revision: https://reviews.llvm.org/D80092
Summary:
When NamedAttrList::get is called against a StringRef and the
entry is not present, the Identifier::operator== crashes.
Interestingly there seems to be no use of the NamedAttrList::get(StringRef) in
the codebase so far. A subsequent commit will introduce such a use.
Differential Revision: https://reviews.llvm.org/D80089
unittest/Tablegen generates an executable that depends on MLIRIR and
LLVMMLIRTableGen. Avoid specifying linkage dependence on LLVM
libraries here because then everyone has to depend on those libraries.
Differential Revision: https://reviews.llvm.org/D80093
Generally:
1) don't use target_link_libraries() and add_mlir_library() on the same target, use LINK_LIBS PUBLIC instead.
2) don't use LINK_LIBS to specify LLVM libraries. Use LINK_COMPONENTS instead
3) no need to link against LLVMSupport. We pull it in by default.
Differential Revision: https://reviews.llvm.org/D80076
Along the lines of D77454 and D79968. Unlike loads and stores, the
default alignment is getPrefTypeAlign, to match the existing handling in
various places, including SelectionDAG and InstCombine.
Differential Revision: https://reviews.llvm.org/D80044
The JitRunner library is logically very close to the execution engine,
and shares similar dependencies.
find -name "*.cpp" -exec sed -i "s/Support\/JitRunner/ExecutionEngine\/JitRunner/" "{}" \;
Differential Revision: https://reviews.llvm.org/D79899
Summary:
First, compact implementation of lowering to LLVM IR. A bit more
challenging than the constant mask due to the dynamic indices, of course.
I like to hear if there are more efficient ways of doing this in LLVM,
but this for now at least gives us a functional reference implementation.
Reviewers: nicolasvasilache, ftynse, bkramer, reidtatge, andydavis1, mehdi_amini
Reviewed By: nicolasvasilache
Subscribers: mehdi_amini, rriddle, jpienaar, shauheen, antiagainst, nicolasvasilache, arpith-jacob, mgester, lucyrfox, liufengdb, stephenneuendorffer, Joonsoo, grosul1, frgossen, Kayjukh, jurahul, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D79954
DimOp folding is using bare accesses to underlying SubViewOp operands.
This is generally incorrect and is fixed in this revision.
Differential Revision: https://reviews.llvm.org/D80017
The "null-pointer-is-valid" attribute needs to be checked by many
pointer-related combines. To make the check more efficient, convert
it from a string into an enum attribute.
In the future, this attribute may be replaced with data layout
properties.
Differential Revision: https://reviews.llvm.org/D78862
It is possible for optimizations to create SSA code which violates
the dominance property in unreachable blocks. Equivalently, dominance
computed using normal mechanisms is undefined in unreachable blocks.
See discussion here: https://llvm.discourse.group/t/rfc-allowing-dialects-to-relax-the-ssa-dominance-condition/833/51
This patch only checks the dominance condition inside blocks which are
reachable from the the entry block of their region. Note that the
dominance conditions of regions contained in an unreachable block are
still checked.
Differential Revision: https://reviews.llvm.org/D79922
The following Conversions are affected: LoopToStandard -> SCFToStandard,
LoopsToGPU -> SCFToGPU, VectorToLoops -> VectorToSCF. Full file paths are
affected. Additionally, drop the 'Convert' prefix from filenames living under
lib/Conversion where applicable.
API names and CLI options for pass testing are also renamed when applicable. In
particular, LoopsToGPU contains several passes that apply to different kinds of
loops (`for` or `parallel`), for which the original names are preserved.
Differential Revision: https://reviews.llvm.org/D79940
have abi attributes.
To ensure there is no conflict, use the default ABI only when none of
the arguments have the spv.interface_var_abi attribute. This also
implies that if one of the arguments has a spv.interface_var_abi
attribute, all of them should have it as well.
Differential Revision: https://reviews.llvm.org/D77232
This revision starts decoupling the include the kitchen sink behavior of Linalg to LLVM lowering by inserting a -convert-linalg-to-std pass.
The lowering of linalg ops to function calls was previously lowering to memref descriptors by having both linalg -> std and std -> LLVM patterns in the same rewrite.
When separating this step, a new issue occurred: the layout is automatically type-erased by this process. This revision therefore introduces memref casts to perform these type erasures explicitly. To connect everything end-to-end, the LLVM lowering of MemRefCastOp is relaxed because it is artificially more restricted than the op semantics. The op semantics already guarantee that source and target MemRefTypes are cast-compatible. An invalid lowering test now becomes valid and is removed.
Differential Revision: https://reviews.llvm.org/D79468
This patch adds `affine.vector_load` and `affine.vector_store` ops to
the Affine dialect and lowers them to `vector.transfer_read` and
`vector.transfer_write`, respectively, in the Vector dialect.
Reviewed By: bondhugula, nicolasvasilache
Differential Revision: https://reviews.llvm.org/D79658
For IR generated by a compiler, this is really simple: you just take the
datalayout from the beginning of the file, and apply it to all the IR
later in the file. For optimization testcases that don't care about the
datalayout, this is also really simple: we just use the default
datalayout.
The complexity here comes from the fact that some LLVM tools allow
overriding the datalayout: some tools have an explicit flag for this,
some tools will infer a datalayout based on the code generation target.
Supporting this properly required plumbing through a bunch of new
machinery: we want to allow overriding the datalayout after the
datalayout is parsed from the file, but before we use any information
from it. Therefore, IR/bitcode parsing now has a callback to allow tools
to compute the datalayout at the appropriate time.
Not sure if I covered all the LLVM tools that want to use the callback.
(clang? lli? Misc IR manipulation tools like llvm-link?). But this is at
least enough for all the LLVM regression tests, and IR without a
datalayout is not something frontends should generate.
This change had some sort of weird effects for certain CodeGen
regression tests: if the datalayout is overridden with a datalayout with
a different program or stack address space, we now parse IR based on the
overridden datalayout, instead of the one written in the file (or the
default one, if none is specified). This broke a few AVR tests, and one
AMDGPU test.
Outside the CodeGen tests I mentioned, the test changes are all just
fixing CHECK lines and moving around datalayout lines in weird places.
Differential Revision: https://reviews.llvm.org/D78403
Summary:
The assembly format for std.rank expects the operand type and not the
result type after the colon.
Differential Revision: https://reviews.llvm.org/D79857
Generally speaking, this is bad practice. It also causes the build to
break if there are editor temporary files.
Differential Revision: https://reviews.llvm.org/D79906
The Vulkan runtime wrapper will be compiled to a shared library
that are loaded by the JIT runner. Depending on LLVM libraries
means that LLVM symbols will be compiled into the shared library.
That can cause problems if we are using it with other shared
libraries depending on LLVM, notably Mesa, the open-source graphics
driver framework. The Vulkan API wrappers invoked by the JIT runner
links to the system libvulkan.so. If it's Mesa providing the
implementation, Mesa will normally try to load the system libLLVM.so
for its shader compilation. That causes issues because the JIT runner
already loaded the Vulkan runtime wrapper which has LLVM sybmols
compiled in. So system linker will instruct Mesa to use those symbols
instead.
Differential Revision: https://reviews.llvm.org/D79860
The CMake structure of the toy example is non-standard. encourage people to
copy the standalone example instead.
Differential Revision: https://reviews.llvm.org/D79889
All ops of the SCF dialect now use the `scf.` prefix instead of `loop.`. This
is a part of dialect renaming.
Differential Revision: https://reviews.llvm.org/D79844
The existing implementation of SubViewOp::getRanges relies on all
offsets/sizes/strides to be dynamic values and does not work in
combination with canonicalization. This revision adds a
SubViewOp::getOrCreateRanges to create the missing constants in the
canonicalized case.
This allows reactivating the fused pass with staged pattern
applications.
However another issue surfaces that the SubViewOp verifier is now too
strict to allow folding. The existing folding pattern is turned into a
canonicalization pattern which rewrites memref_cast + subview into
subview + memref_cast.
The transform-patterns-matmul-to-vector can then be reactivated.
Differential Revision: https://reviews.llvm.org/D79759
Due to the extension of Liveness, Buffer Assignment can now work on nested regions. This PR provides a test case to show that existing functionally of BA works properly.
Differential Revision: https://reviews.llvm.org/D79332
The current standard Alloca node is not annotated with the
MemEffect<Alloc> trait. This CL updates the Alloc and Alloca
memory-effect annotations using the latest Resource objects. Alloca
nodes will use a newly defined AutomaticAllocationScopeResource to
distinguish between Alloc and Alloca memory effects.
Differential Revision: https://reviews.llvm.org/D79620
This is only valid if the source tensors (result tensor) is static
shaped with all unit-extents when the reshape is collapsing
(expanding) dimensions.
Differential Revision: https://reviews.llvm.org/D79764
Summary:
Makes this operation runnable on CPU by generating MLIR instructions
that are eventually folded into an LLVM IR constant for the mask.
Reviewers: nicolasvasilache, ftynse, reidtatge, bkramer, andydavis1
Reviewed By: nicolasvasilache, ftynse, andydavis1
Subscribers: mehdi_amini, rriddle, jpienaar, shauheen, antiagainst, nicolasvasilache, arpith-jacob, mgester, lucyrfox, liufengdb, stephenneuendorffer, Joonsoo, grosul1, frgossen, Kayjukh, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D79815
The main objective of this revision is to change the way static information is represented, propagated and canonicalized in the SubViewOp.
In the current implementation the issue is that canonicalization may strictly lose information because static offsets are combined in irrecoverable ways into the result type, in order to fit the strided memref representation.
The core semantics of the op do not change but the parser and printer do: the op always requires `rank` offsets, sizes and strides. These quantities can now be either SSA values or static integer attributes.
The result type is automatically deduced from the static information and more powerful canonicalizations (as powerful as the representation with sentinel `?` values allows). Previously static information was inferred on a best-effort basis from looking at the source and destination type.
Relevant tests are rewritten to use the idiomatic `offset: x, strides : [...]`-form. Bugs are corrected along the way that were not trivially visible in flattened strided memref form.
Lowering to LLVM is updated, simplified and now supports all cases.
A mixed static-dynamic mode test that wouldn't previously lower is added.
It is an open question, and a longer discussion, whether a better result type representation would be a nicer alternative. For now, the subview op carries the required semantic.
Differential Revision: https://reviews.llvm.org/D79662
Conversion/ folders were originally intended to store patterns for
DialectA->DialectB conversions that depend on both dialects and do not
conceptually belong to either of the dialects. As such, DialectA->DialectA
conversion does not make sense under Conversion/ and should rather live with
the dialect it operates on.
Differential Revision: https://reviews.llvm.org/D79569
cmake does not truly support dependencies on automatically generated files
which are not in the same directory as the targets which depend on them.
It works with ninja, but doesn't work with make
This patch adds an explicit dependence so that all dialects are built
before the analysis libraries.
Differential Revision: https://reviews.llvm.org/D79805
This normalize the name of the tablegen file with the name of the generated
files (SideEffectInterfaces.h.inc) and the other Interface tablegen files,
which all end in Interface(s).td
Differential Revision: https://reviews.llvm.org/D79517
This reverts commit 80d133b24f.
Per Stephan Herhut: The canonicalizer pattern that was added creates
forms of the subview op that cannot be lowered.
This is shown by failing Tensorflow XLA tests such as:
tensorflow/compiler/xla/service/mlir_gpu/tests:abs.hlo.test
Will provide more details offline, they rely on logs from private CI.
Summary:
The scalar zero + splat yields more intermediate code than the direct
dense zero constant, and ultimately is lowered to exactly the same
LLVM IR operations, so no point wasting the intermediate code.
Reviewers: nicolasvasilache, andydavis1, reidtatge
Reviewed By: nicolasvasilache
Subscribers: mehdi_amini, rriddle, jpienaar, shauheen, antiagainst, nicolasvasilache, arpith-jacob, mgester, lucyrfox, liufengdb, stephenneuendorffer, Joonsoo, grosul1, frgossen, Kayjukh, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D79758
Summary:
- Fix comments in several places
- Eliminate extra ' in AST dump and adjust tests accordingly
Differential Revision: https://reviews.llvm.org/D78399
Summary:
The main objective of this revision is to change the way static information is represented, propagated and canonicalized in the SubViewOp.
In the current implementation the issue is that canonicalization may strictly lose information because static offsets are combined in irrecoverable ways into the result type, in order to fit the strided memref representation.
The core semantics of the op do not change but the parser and printer do: the op always requires `rank` offsets, sizes and strides. These quantities can now be either SSA values or static integer attributes.
The result type is automatically deduced from the static information and more powerful canonicalizations (as powerful as the representation with sentinel `?` values allows). Previously static information was inferred on a best-effort basis from looking at the source and destination type.
Relevant tests are rewritten to use the idiomatic `offset: x, strides : [...]`-form. Bugs are corrected along the way that were not trivially visible in flattened strided memref form.
It is an open question, and a longer discussion, whether a better result type representation would be a nicer alternative. For now, the subview op carries the required semantic.
Reviewers: ftynse, mravishankar, antiagainst, rriddle!, andydavis1, timshen, asaadaldien, stellaraccident
Reviewed By: mravishankar
Subscribers: aartbik, bondhugula, mehdi_amini, rriddle, jpienaar, shauheen, antiagainst, arpith-jacob, mgester, lucyrfox, liufengdb, stephenneuendorffer, Joonsoo, bader, grosul1, frgossen, Kayjukh, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D79662
Summary:
This revision introduces a helper function to allow applying rewrite patterns, interleaved with more global transformations, in a staged fashion:
1. the first stage consists of an OwningRewritePatternList. The RewritePattern in this list are applied once and in order.
2. the second stage consists of a single OwningRewritePattern that is applied greedily until convergence.
3. the third stage consists of applying a lambda, generally used for non-local transformation effects.
This allows creating custom fused transformations where patterns can be ordered and applied at a finer granularity than a sequence of traditional compiler passes.
A test that exercises these behaviors is added.
Differential Revision: https://reviews.llvm.org/D79518
Summary:
This makes a common pattern of
`dyn_cast_or_null<OpTy>(v.getDefiningOp())` more concise.
Differential Revision: https://reviews.llvm.org/D79681
This [discussion](https://llvm.discourse.group/t/viewop-isnt-expressive-enough/991/2) raised some concerns with ViewOp.
In particular, the handling of offsets is incorrect and does not match the op description.
Note that with an elemental type change, offsets cannot be part of the type in general because sizeof(srcType) != sizeof(dstType).
Howerver, offset is a poorly chosen term for this purpose and is renamed to byte_shift.
Additionally, for all intended purposes, trying to support non-identity layouts for this op does not bring expressive power but rather increases code complexity.
This revision simplifies the existing semantics and implementation.
This simplification effort is voluntarily restrictive and acts as a stepping stone towards supporting richer semantics: treat the non-common cases as YAGNI for now and reevaluate based on concrete use cases once a round of simplification occurred.
Differential revision: https://reviews.llvm.org/D79541
Summary: This revision introduces LinalgPromotionOptions to more easily control the application of promotion patterns. It also simplifies the different entry points into Promotion in preparation for some behavior change in subsequent revisions.
Differential Revision: https://reviews.llvm.org/D79489
This dialect contains various structured control flow operaitons, not only
loops, reflect this in the name. Drop the Ops suffix for consistency with other
dialects.
Note that this only moves the files and changes the C++ namespace from 'loop'
to 'scf'. The visible IR prefix remains the same and will be updated
separately. The conversions will also be updated separately.
Differential Revision: https://reviews.llvm.org/D79578
Summary:
Cast from a value interpreted as floating-point to the corresponding signed
integer value. Similar to an element-wise `static_cast` in C++, performs an
element-wise conversion operation.
Differential Revision: https://reviews.llvm.org/D79373
Functions checking whether an SSA value is a valid dimension or symbol for
affine operations can be called on values defined in a detached region (a
region that is not yet attached to an operation), for example, during parsing
or operation construction. These functions will attempt to uncondtionally
dereference a pointer to the parent operation of a region, which may be null
(as fixed by the previous commit, uninitialized before that). Since one cannot
know to which operation a region will be attached, conservatively this
operation would not be a valid affine scope and act accordingly, instead of
crashing.
Region has a default constructor that is called when a region is constructed
while an operation is being created, and therefore before the region can be
attached to this operation. The `container` field is uninitialized, which makes
it impossible to check programmatically if a Region is attached to an operation
or not, leading to sly memory errors when this field is read. Initialize it to
nullptr by default and thus make sure one can check if a region is attached to
an operation or not.
Complex addition and substraction are the first two binary operations on complex
numbers.
Remaining operations will follow the same pattern.
Differential Revision: https://reviews.llvm.org/D79479
The SideEffect interface definitions currently use string expressions to
reference custom resource objects. This CL introduces Resource objects in
tablegen definitions to simplify linking of resource reference to resource
objects.
Differential Revision: https://reviews.llvm.org/D78917
This is a wrapper around vector of NamedAttributes that keeps track of whether sorted and does some minimal effort to remain sorted (doing more, e.g., appending attributes in sorted order, could be done in follow up). It contains whether sorted and if a DictionaryAttr is queried, it caches the returned DictionaryAttr along with whether sorted.
Change MutableDictionaryAttr to always return a non-null Attribute even when empty (reserve null cases for errors). To this end change the getter to take a context as input so that the empty DictionaryAttr could be queried. Also create one instance of the empty dictionary attribute that could be reused without needing to lock context etc.
Update infer type op interface to use DictionaryAttr and use NamedAttrList to avoid incurring multiple conversion costs.
Fix bug in sorting helper function.
Differential Revision: https://reviews.llvm.org/D79463
vulkan-runtime-wrappers does not need MLIRSPIRVSerialization,
which is used by the ConvertGpuLaunchFuncToVulkanLaunchFunc pass
under the hood.
Differential Revision: https://reviews.llvm.org/D79577
SPIR-V ops can mix operands and attributes in the definition. These
operands and attributes are serialized in the exact order of the definition
to match SPIR-V binary format requirements. It can cause excessive
generated code bloat because we are emitting code to handle each
operand/attribute separately. So here we probe first to check whether all
the operands are ahead of attributes. Then we can serialize all operands
together.
This removes ~1000 lines of code from the generated inc file.
Differential Revision: https://reviews.llvm.org/D79446
These template functions are used in the serializer, where we can
actually directly query the opcode from the op's definition and
use that in the auto-generated serialization logic.
This removes a set of templates accounting for 319 lines from
the auto-generated inc file.
Differential Revision: https://reviews.llvm.org/D79444
Originally, these operations were folded only if all expressions in their
affine maps could be folded to a constant expression that can be then subject
to numeric min/max computation. This introduces a more advanced version that
partially folds the affine map by lifting individual constant expression in it
even if some of the expressions remain variable. The folding can update the
operation in place to use a simpler map. Note that this is not as powerful as
canonicalization, in particular this does not remove dimensions or symbols that
became useless. This allows for better composition of Linalg tiling and
promotion transformation, where the latter can handle some canonical forms of
affine.min that the folding can now produce.
Differential Revision: https://reviews.llvm.org/D79502
In the Vector to LLVM conversion, the `replaceTransferOp` function calls
into a type converter that may fail and suppresses the status. Change
the function to return the failure status instead, Since it is called
from a pattern, the failure can be readily propagated to the rest of
infrastructure.
The list of destination load ops while evaluating producer-consumer
fusion wasn't being maintained as a set, and as such, duplicate load ops
were being added to it. Although this is harmless correctness-wise, it's
a killer efficiency-wise and it prevents interesting/useful fusions
(including for eg. reshapes into a matmul). The reason the latter
fusions would be missed is that a slice union would be unnecessarily
needed due to the duplicate load ops on a memref added to the 'dst
loads' list. Since slice union is unimplemented for the local var case,
a single destination load op that leads to local vars (like a floordiv /
mod producing fusion), a common case, would not get fused due to an
unnecessary union being tried with itself. (The union would actually be
the same thing but we would bail out.)
Besides the above, this would also significantly speed up fusion as all
the unnecessary slice computations / unions, checks, etc. due to the
duplicates go away.
Differential Revision: https://reviews.llvm.org/D79547
Jacques Pienaar