Don't assert fail on strided memrefs when dropping unit dims.
Instead just leave them unchanged.
Differential Revision: https://reviews.llvm.org/D108205
* This allows multiple MLIR-API embedding downstreams to co-exist in the same process.
* I believe this is the last thing needed to enable isolated embedding.
Differential Revision: https://reviews.llvm.org/D108605
An interface to allow for tiling of operations is introduced. The
tiling of the linalg.pad_tensor operation is modified to use this
interface.
Differential Revision: https://reviews.llvm.org/D108611
The StringAttr version doesn't need a context, so we can just use the
existing `SymbolRefAttr::get` form. The StringRef version isn't preferred
so we want to encourage people to use StringAttr.
There is an additional form of getSymbolRefAttr that takes a (SymbolTrait
implementing) operation. This should also be moved, but I'll do that as
a separate patch.
Differential Revision: https://reviews.llvm.org/D108922
* It is pretty clear that no one has tried this yet since it was both incomplete and broken.
* Fixes a symbol hiding issues keeping even the generic builder from constructing an operation with successors.
* Adds ODS support for successors.
* Adds CAPI `mlirBlockGetParentRegion`, `mlirRegionEqual` + tests (and missing test for `mlirBlockGetParentOperation`).
* Adds Python property: `Block.region`.
* Adds Python methods: `Block.create_before` and `Block.create_after`.
* Adds Python property: `InsertionPoint.block`.
* Adds new blocks.py test to verify a plausible CFG construction case.
Differential Revision: https://reviews.llvm.org/D108898
41d4aa7de6 introduced incorrect code in
extraTraitClassDeclaration: `this` refers to the trait class and not the
operation class so `->getContext()` is not valid. Use `$_op` instead.
SymbolRefAttr is fundamentally a base string plus a sequence
of nested references. Instead of storing the string data as
a copies StringRef, store it as an already-uniqued StringAttr.
This makes a lot of things simpler and more efficient because:
1) references to the symbol are already stored as StringAttr's:
there is no need to copy the string data into MLIRContext
multiple times.
2) This allows pointer comparisons instead of string
comparisons (or redundant uniquing) within SymbolTable.cpp.
3) This allows SymbolTable to hold a DenseMap instead of a
StringMap (which again copies the string data and slows
lookup).
This is a moderately invasive patch, so I kept a lot of
compatibility APIs around. It would be nice to explore changing
getName() to return a StringAttr for example (right now you have
to use getNameAttr()), and eliminate things like the StringRef
version of getSymbol.
Differential Revision: https://reviews.llvm.org/D108899
* Add `DimOfIterArgFolder`.
* Move existing cross-dialect canonicalization patterns to `LoopCanonicalization.cpp`.
* Rename `SCFAffineOpCanonicalization` pass to `SCFForLoopCanonicalization`.
* Expand documentaton of scf.for: The type of loop-carried variables may not change with iterations. (Not even the dynamic type.)
Differential Revision: https://reviews.llvm.org/D108806
* Add batched version of all `addId` variants, so that multiple IDs can be added at a time.
* Rename `addId` and variants to `insertId` and `appendId`. Most external users call `appendId`. Splitting `addId` into two functions also makes it possible to provide batched version for both. (Otherwise, the overloads are ambigious when calling `addId`.)
Differential Revision: https://reviews.llvm.org/D108532
Drop mgpuMemHostRegisterMemRef's dependence on LLVM Support. This
method is the only one in CUDA runtime wrappers library that creates
a dependence on libLLVMSupport due to its use of SmallVector and
ArrayRef. The code can be as easily/compactly written without those ADT.
The dependence on LLVMSupport adds a significant amount of additional
complexity for external things that want to link this library in (both
statically or as a shared object) since libLLVMSupport includes numerous
other objects that are sensitive to C++ compiler version and ABI.
Differential Revision: https://reviews.llvm.org/D108684
Needed to switch to extract to support tosa.reverse using dynamic shapes.
Reviewed By: NatashaKnk
Differential Revision: https://reviews.llvm.org/D108744
This prepares general sparse to sparse conversions. The code that
needs to be generated using this new feature is now simply:
(1) coo = sparse_tensor_1->asCOO(); // source format1
(2) sparse_tensor_2 = newSparseTensor(coo); // destination format2
By using COO as an intermediate, we can do *all* conversions without
having to implement the full O(N^2) conversion matrix. Note that we
can always improve particular conversions individually if a faster
solution is required.
Reviewed By: bixia
Differential Revision: https://reviews.llvm.org/D108681
This allows for using a different type when accessing a parameter than the
one used for storage. This allows for returning parameters by reference,
enables using more optimized/convient reference results, and more.
Differential Revision: https://reviews.llvm.org/D108593
This allows for parsing strings that have escape sequences, which require constructing
a string (as they can't be represented by looking at the Token contents directly).
Differential Revision: https://reviews.llvm.org/D108589
This allows for iterating and interacting with the uses of a specific subset of
results as opposed to just the full range.
Differential Revision: https://reviews.llvm.org/D108586
Includes the quantized version of average pool lowering to linalg dialect.
This includes a lit test for the transform. It is not 100% correct as the
multiplier / shift should be done in i64 however this is negligable rounding
difference.
Reviewed By: NatashaKnk
Differential Revision: https://reviews.llvm.org/D108676
Lowering to table was incorrect as it did not apply a 128 offset before
extracting the value from the table. Fixed and correct tensor length on input
table.
Reviewed By: NatashaKnk
Differential Revision: https://reviews.llvm.org/D108436
* Add support for affine.max ops to SCF loop peeling pattern.
* Add support for affine.max ops to `AffineMinSCFCanonicalizationPattern`.
* Rename `AffineMinSCFCanonicalizationPattern` to `AffineOpSCFCanonicalizationPattern`.
* Rename `AffineMinSCFCanonicalization` pass to `SCFAffineOpCanonicalization`.
Differential Revision: https://reviews.llvm.org/D108009
The emplace commands are variadic and should take all the constructor arguments directly, since they implicitly call the constructor themselves in order to avoid the cost of constructing and then moving/copying temporaries.
Reviewed By: aartbik
Differential Revision: https://reviews.llvm.org/D108670
When padding quantized operations, the padding needs to equal the zero point
of the input value. Corrected the pass to change the padding value if quantized.
Reviewed By: NatashaKnk
Differential Revision: https://reviews.llvm.org/D108440
Add notes for discarding private-visible functions in the Toy tutorial chapter 4.
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D108026
Recent changes outside sparse compiler exposed the requirement of running a
new pass (lower-affine) but this only became apparent with private testing.
By adding some vectorized runs to integration test, we will detect the need
for such changes earlier and also widen codegen coverage of course.
Reviewed By: gussmith23
Differential Revision: https://reviews.llvm.org/D108667
This canonicalization simplifies affine.min operations inside "for loop"-like operations (e.g., scf.for and scf.parallel) based on two invariants:
* iv >= lb
* iv < lb + step * ((ub - lb - 1) floorDiv step) + 1
This commit adds a new pass `canonicalize-scf-affine-min` (instead of being a canonicalization pattern) to avoid dependencies between the Affine dialect and the SCF dialect.
Differential Revision: https://reviews.llvm.org/D107731
Introduces new Ops to represent 1. alias.scope metadata in LLVM, and 2. domains for these scopes. These correspond to the metadata described in https://llvm.org/docs/LangRef.html#noalias-and-alias-scope-metadata. Lists of scopes are modeled the same way as access groups - as an ArrayAttr on the Op (added in https://reviews.llvm.org/D97944).
Lowering 'noalias' attributes on function parameters is already supported. However, lowering `noalias` metadata on individual Ops is not, which is added in this change. LLVM uses the same keyword for these, but this change introduces a separate attribute name 'noalias_scopes' to represent this distinct concept.
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D107870
I found myself typing this code several times at different places
by now, so time to make this a general utility instead. Given
a permutation, it returns the permuted position of the input,
for example (i,j,k) -> (k,i,j) yields position 1 for input 0.
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D108347
If additional static type information can be deduced from a insert_slice's size operands, insert an explicit cast of the op's source operand.
This enables other canonicalization patterns that are matching for tensor_cast ops such as `ForOpTensorCastFolder` in SCF.
Differential Revision: https://reviews.llvm.org/D108617
Rationale:
Passing in a pointer to the memref data in order to implement the
dense to sparse conversion was a bit too low-level. This revision
improves upon that approach with a cleaner solution of generating
a loop nest in MLIR code itself that prepares the COO object before
passing it to our "swiss army knife" setup. This is much more
intuitive *and* now also allows for dynamic shapes.
Reviewed By: bixia
Differential Revision: https://reviews.llvm.org/D108491
This revision adds native ODS support for VariadicOfVariadic operand
groups. An example of this is the SwitchOp, which has a variadic number
of nested operand ranges for each of the case statements, where the
number of case statements is variadic. Builtin ODS support allows for
generating proper accessors for the nested operand ranges, builder
support, and declarative format support. VariadicOfVariadic operands
are supported by providing a segment attribute to use to store the
operand groups, mapping similarly to the AttrSizedOperand trait
(but with a user defined attribute name).
`build` methods for VariadicOfVariadic operand expect inputs of the
form `ArrayRef<ValueRange>`. Accessors for the variadic ranges
return a new `OperandRangeRange` type, which represents a
contiguous range of `OperandRange`. In the declarative assembly
format, VariadicOfVariadic operands and types are by default
formatted as a comma delimited list of value lists:
`(<value>, <value>), (), (<value>)`.
Differential Revision: https://reviews.llvm.org/D107774
This allows for inlining into an empty block or to the beginning of a block. NFC as the existing implementations now foward to this overload.
Differential Revision: https://reviews.llvm.org/D108572
This revision fixes a bug where an operation would get replaced with
a pre-existing constant that didn't dominate it. This can occur when
a pattern inserts operations to be folded at the beginning of the
constants insertion block. This revision fixes the bug by moving the
existing constant before the replaced operation in such cases. This is
fine because if a constant didn't already exist, a new one would have
been inserted before this operation anyways.
Differential Revision: https://reviews.llvm.org/D108498
Do not apply loop peeling to loops that are contained in the partial iteration of an already peeled loop. This is to avoid code explosion when dealing with large loop nests. Can be controlled with a new pass option `skip-partial`.
Differential Revision: https://reviews.llvm.org/D108542
* This is the native data layout for PyTorch and npcomp was using the prior version before cleanup.
Differential Revision: https://reviews.llvm.org/D108527
* Resolves a TODO by making this configurable by downstreams.
* This seems to be the last thing allowing full use of the Python bindings as a library within another project (i.e. be embedding them).
Differential Revision: https://reviews.llvm.org/D108523
Presently, the lowering of nested scf.parallel loops to OpenMP creates one omp.parallel region, with two (nested) OpenMP worksharing loops on the inside. When lowered to LLVM and executed, this results in incorrect results. The reason for this is as follows:
An OpenMP parallel region results in the code being run with whatever number of threads available to OpenMP. Within a parallel region a worksharing loop divides up the total number of requested iterations by the available number of threads, and distributes accordingly. For a single ws loop in a parallel region, this works as intended.
Now consider nested ws loops as follows:
omp.parallel {
A: omp.ws %i = 0...10 {
B: omp.ws %j = 0...10 {
code(%i, %j)
}
}
}
Suppose we ran this on two threads. The first workshare loop would decide to execute iterations 0, 1, 2, 3, 4 on thread 0, and iterations 5, 6, 7, 8, 9 on thread 1. The second workshare loop would decide the same for its iteration. This means thread 0 would execute i \in [0, 5) and j \in [0, 5). Thread 1 would execute i \in [5, 10) and j \in [5, 10). This means that iterations i in [5, 10), j in [0, 5) and i in [0, 5), j in [5, 10) never get executed, which is clearly wrong.
This permits two options for a remedy:
1) Change the semantics of the omp.wsloop to be distinct from that of the OpenMP runtime call or equivalently #pragma omp for. This could then allow some lowering transformation to remedy the aforementioned issue. I don't think this is desirable for an abstraction standpoint.
2) When lowering an scf.parallel always surround the wsloop with a new parallel region (thereby causing the innermost wsloop to use the number of threads available only to it).
This PR implements the latter change.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D108426
Multiple operations were still defined as TC ops that had equivalent versions
as YAML operations. Reducing to a single compilation path guarantees that
frontends can lower to their equivalent operations without missing the
optimized fastpath.
Some operations are maintained purely for testing purposes (mainly conv{1,2,3}D
as they are included as sole tests in the vectorizaiton transforms.
Differential Revision: https://reviews.llvm.org/D108169
Folding in the MLIR uses the order of the type directly
but folding in the underlying implementation must take
the dim ordering into account. These tests clarify that
behavior and verify it is done right.
Reviewed By: bixia
Differential Revision: https://reviews.llvm.org/D108474
The boilerplate was setting up some arrays for testing. To fully illustrate
python - MLIR potential, however, this data should also come from numpy land.
Reviewed By: bixia
Differential Revision: https://reviews.llvm.org/D108336
Tosa rescale can contain uint8 types. Added support for these types
using an unrealized conversion cast. Optimistically it would be better to
use bitcast however it does not support unsigned integers.
Differential Revision: https://reviews.llvm.org/D108427
Previously, ExecuteRegionOps with multiple return values would fail a round-trip test due to missing parenthesis around the types.
Differential Revision: https://reviews.llvm.org/D108402
Apply the "for loop peeling" pattern from SCF dialect transforms. This pattern splits scf.for loops into full and partial iterations. In the full iteration, all masked loads/stores are canonicalized to unmasked loads/stores.
Differential Revision: https://reviews.llvm.org/D107733
Simplify affine.min ops, enabling various other canonicalizations inside the peeled loop body.
affine.min ops such as:
```
map = affine_map<(d0)[s0, s1] -> (s0, -d0 + s1)>
%r = affine.min #affine.min #map(%iv)[%step, %ub]
```
are rewritten them into (in the case the peeled loop):
```
%r = %step
```
To determine how an affine.min op should be rewritten and to prove its correctness, FlatAffineConstraints is utilized.
Differential Revision: https://reviews.llvm.org/D107222
This shares more code with existing utilities. Also, to be consistent,
we moved dimension permutation on the DimOp to the tensor lowering phase.
This way, both pre-existing DimOps on sparse tensors (not likely but
possible) as well as compiler generated DimOps are handled consistently.
Reviewed By: bixia
Differential Revision: https://reviews.llvm.org/D108309
Its possible for the clamp to have invalid min/max values on its range. To fix
this we validate the range of the min/max and clamp to a valid range.
Reviewed By: NatashaKnk
Differential Revision: https://reviews.llvm.org/D108256
LLVM considers global variables marked as externals to be defined within the module if it is initialized (including to an undef). Other external globals are considered as being defined externally and imported into the current translation unit. Lowering of MLIR Global Ops does not properly propagate undefined initializers, resulting in a global which is expected to be defined within the current TU, not being defined.
Differential Revision: https://reviews.llvm.org/D108252
MSVC needs to know where to put the archive (.lib) as well as the runtime
(.dll). If left to the default location, multiple rules to generate the same
file will be produced, creating a Ninja error.
Differential Revision: https://reviews.llvm.org/D108181
* Rename ids to values in FlatAffineValueConstraints.
* Overall cleanup of comments in FlatAffineConstraints and FlatAffineValueConstraints.
Differential Revision: https://reviews.llvm.org/D107947
* Extract "value" functionality of `FlatAffineConstraints` into a new derived `FlatAffineValueConstraints` class. Current users of `FlatAffineConstraints` can use `FlatAffineValueConstraints` without additional code changes, thus NFC.
* `FlatAffineConstraints` no longer associates dimensions with SSA Values. All functionality that requires this, is moved to `FlatAffineValueConstraints`.
* `FlatAffineConstraints` no longer makes assumptions about where Values associated with dimensions are coming from.
Differential Revision: https://reviews.llvm.org/D107725
This method bitcasts a DenseElementsAttr elementwise to one of the same
shape with a different element type.
Reviewed By: rriddle
Differential Revision: https://reviews.llvm.org/D107612
Reduction axis should come after all parallel axis to work with vectorization.
Reviewed By: NatashaKnk
Differential Revision: https://reviews.llvm.org/D108005
These operations are not lowered to from any source dialect and are only
used for redundant tests. Removing these named ops, along with their
associated tests, will make migration to YAML operations much more
convenient.
Reviewed By: stellaraccident
Differential Revision: https://reviews.llvm.org/D107993
Expand ParallelLoopTilingPass with an inbound_check mode.
In default mode, the upper bound of the inner loop is from the min op; in
inbound_check mode, the upper bound of the inner loop is the step of the outer
loop and an additional inbound check will be emitted inside of the inner loop.
This was 'FIXME' in the original codes and a typical usage is for GPU backends,
thus the outer loop and inner loop can be mapped to blocks/threads in seperate.
Differential Revision: https://reviews.llvm.org/D105455
The primary pattern for this pass clones many operations from producers
to consumers. Doing this top down prevents duplicated work when a
producer has multiple consumers, if it also is consuming another
linalg.generic.
As an example, a chain of ~2600 generics that are fused into ~70
generics was resulting in 16255 pattern invocations. This took 14
seconds on one machine but takes only 0.3 seconds with top-down
traversal.
Differential Revision: https://reviews.llvm.org/D107818
While the changes are extensive, they basically fall into a few
categories:
1) Moving the TestDialect itself.
2) Updating C++ code in tablegen to explicitly use ::mlir, since it
will be put in a headers that shouldn't expect a 'using'.
3) Updating some generic MLIR Interface definitions to do the same thing.
4) Updating the Tablegen generator in a few places to be explicit about
namespaces
5) Doing the same thing for llvm references, since we no longer pick
up the definitions from mlir/Support/LLVM.h
Differential Revision: https://reviews.llvm.org/D88251
The approach for handling reductions in the outer most
dimension follows that for inner most dimensions, outlined
below
First, transpose to move reduction dims, if needed
Convert reduction from n-d to 2-d canonical form
Then, for outer reductions, we emit the appropriate op
(add/mul/min/max/or/and/xor) and combine the results.
Differential Revision: https://reviews.llvm.org/D107675
Add in-source documentation on how CanonicalLoopInfo is intended to be used. In particular, clarify what parts of a CanonicalLoopInfo is considered part of the loop, that those parts must be side-effect free, and that InsertPoints to instructions outside those parts can be expected to be preserved after method calls implementing loop-associated directives.
CanonicalLoopInfo are now invalidated after it does not describe canonical loop anymore and asserts when trying to use it afterwards.
In addition, rename `createXYZWorkshareLoop` to `applyXYZWorkshareLoop` and remove the update location to avoid that the impression that they insert something from scratch at that location where in reality its InsertPoint is ignored. createStaticWorkshareLoop does not return a CanonicalLoopInfo anymore. First, it was not a canonical loop in the clarified sense (containing side-effects in form of calls to the OpenMP runtime). Second, it is ambiguous which of the two possible canonical loops it should actually return. It will not be needed before a feature expected to be introduced in OpenMP 6.0
Also see discussion in D105706.
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D107540
Move StaticVerifierFunctionEmitter to CodeGenHelper.h so that it can be
used for both ODS and DRR.
Reviewed By: jpienaar
Differential Revision: https://reviews.llvm.org/D106636
Using the python API to easily set up sparse kernels, this test
exhaustively builds, compilers, and runs SpMM for all annotations
on a sparse tensor, making sure every version generates the correct
result. This test also illustrates using the python API to set up
a sparse kernel and sparse compilation.
Reviewed By: bixia
Differential Revision: https://reviews.llvm.org/D107943
This reverts the revert 28c04794df.
The failing MLIR test that caused the revert should be fixed in this
version.
Also includes a PPC test fix previously in 1f87c7c478.
This can be useful when one needs to know which unrolled iteration an Op belongs to, for example, conveying noalias information among memory-affecting ops in parallel-access loops.
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D107789
Existing linalg.conv2d is not well optimized for performance. Changed to a
version that is more aligned for optimziation. Include the corresponding
transposes to use this optimized version.
This also splits the conv and depthwise conv into separate implementations
to avoid overly complex lowerings.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D107504
This is a bit cleaner and removes issues with 2d vectors. It also has a
big impact on constant folding, hence the test changes.
Differential Revision: https://reviews.llvm.org/D107896
The conversion is a straightforward one-to-one mapping with optional unrolling
for nD vectors, similarly to other cast operations.
Depends On D107889
Reviewed By: cota, akuegel
Differential Revision: https://reviews.llvm.org/D107891
The constraint was checking that the type is not an LLVM structure or array
type, but was not checking that it is an LLVM-compatible type, making it accept
incorrect types. As a result, some LLVM dialect ops could process values that
are not compatible with the LLVM dialect leading to further issues with
conversions and translations that assume all values are LLVM-compatible. Make
LLVM_AnyNonAggregate only accept LLVM-compatible types.
Reviewed By: cota, akuegel
Differential Revision: https://reviews.llvm.org/D107889
Reimplement this function in terms of `composeMatchingMap`.
Also fix a bug in `composeMatchingMap` where local dims of `this` could be missing in `localCst`.
Differential Revision: https://reviews.llvm.org/D107813
This function overload is similar to the existing `FlatAffineConstraints::addLowerOrUpperBound`. It constrains a dimension based on an affine map. However, in contrast to the other overloading, it does not attempt to align dimensions/symbols of the affine map with the dimensions/symbols of the constraint set. Instead, dimensions/symbols are expected to already be aligned.
Differential Revision: https://reviews.llvm.org/D107727
This function aligns an affine map (and operands) with given dims and syms SSA values.
This is useful in conjunction with `FlatAffineConstraints::addLowerOrUpperBound`, which requires the `boundMap` to be aligned with the constraint set's dims and syms.
Differential Revision: https://reviews.llvm.org/D107728
Some folding cases are trivial to fold away, specifically no-op cases where
an operation's input and output are the same. Canonicalizing these away
removes unneeded operations.
The current version includes tensor cast operations to resolve shape
discreprencies that occur when an operation's result type differs from the
input type. These are resolved during a tosa shape propagation pass.
Reviewed By: NatashaKnk
Differential Revision: https://reviews.llvm.org/D107321
Dilation only requires increasing the padding on the left/right side of the
input, and including dilation in the convolution. This implementation still
lacks support for strided convolutions.
Reviewed By: NatashaKnk
Differential Revision: https://reviews.llvm.org/D107680
When using an attribute where a value is expected previously this would fail
complaining about unbound symbol. Instead make error clear and mention common
failure reason.
This patch enables normalizing memrefs with MemRef_ReinterpretCastOp by
adding MemRefsNormalizable trait in the Op definition.
Signed-off-by: Haruki Imai <imaihal@jp.ibm.com>
Reviewed By: bondhugula
Differential Revision: https://reviews.llvm.org/D107425
This enables querying shapes/values as shapes without mutating the IR
directly (e.g., towards enabling doing inference in analysis &
application steps, inferring function shape with constant from callsite,
...). Add a new ShapeAdaptor that abstracts over whether shape is from
Type or ShapedTypeComponents or DenseIntElementsAttribute. This adds new
accessors to ValueShapeRange to get Shape and value as shape, but
doesn't restrict or remove the previous way of accessing Type via the
Value for now, that does mean a less refined shape could be accidentally
queried and will be restricted in follow up.
Currently restricted Value query to what can be represented as Shape. So
only supports cases where constant subgraph evaluation's output is a
shape. I had considered making it more general, but without TBD extern
attribute concept or some such a user cannot today uniformly avoid
overhead.
Update TOSA ops and also the shape inference pass.
Differential Revision: https://reviews.llvm.org/D107768
Replace some code snippets With scf::ForOp methods. Additionally,
share a listener at one more point (although this pattern is still
not safe to roll back currently)
Differential Revision: https://reviews.llvm.org/D107754
The following constructor call (and others) used to be ambiguous:
```
FlatAffineConstraints constraints(0, 0, 0);
```
Differential Revision: https://reviews.llvm.org/D107726
Looks "under the hood" of the sparse stogage schemes.
Users should typically not be interested in these details
(hey, that is why we have "sparse compilers"!) but this
test makes sure the compact contents are as expected.
Reviewed By: ThomasRaoux, bixia
Differential Revision: https://reviews.llvm.org/D107683
Implements lowering dense to sparse conversion, for static tensor types only.
First step towards general sparse_tensor.convert support.
Reviewed By: ThomasRaoux
Differential Revision: https://reviews.llvm.org/D107681
These ops were not ported to the nD vector conversion when it was introduced
and nobody needed them so far.
Reviewed By: gysit
Differential Revision: https://reviews.llvm.org/D107750
Instead, include `<cstdlib>` which is the canonical header containing
the declaration of `alloca()`.
Reviewed By: bondhugula
Differential Revision: https://reviews.llvm.org/D107699
Perform scalar constant propagation for FPTruncOp only if the resulting value can be represented without precision loss or rounding.
Example:
%cst = constant 1.000000e+00 : f32
%0 = fptrunc %cst : f32 to bf16
-->
%cst = constant 1.000000e+00 : bf16
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D107518
Use new return type for `OpAsmDialectInterface::getAlias`:
* `AliasResult::NoAlias` if an alias was not provided.
* `AliasResult::OverridableAlias` if an alias was provided, but it might be overriden by other hook.
* `AliasResult::FinalAlias` if an alias was provided and it should be used (no other hooks will be checked).
In that case `AsmPrinter` will use either the first alias with `FinalAlias` result or
the last alias with `OverridableAlias` result (it depends on dialect array order).
Used `OverridableAlias` result for `BuiltinOpAsmDialectInterface`.
Use case: provide more informative alias for built-in attributes like `AffineMapAttr`
instead of generic "map<N>".
Reviewed By: rriddle
Differential Revision: https://reviews.llvm.org/D107437
Avoiding absolute imports allows the code to be relocatable (which is used for out of tree integrations).
Differential Revision: https://reviews.llvm.org/D107617
CastOp::areCastCompatible does not check whether casts are definitely compatible.
When going from dynamic to static offset or stride, the canonicalization cannot
know whether it is really cast compatible. In that case, it can only canonicalize
to an alloc plus copy.
Differential Revision: https://reviews.llvm.org/D107545
Tested with gcc-10. Other compilers may generate additional warnings. This does not fix all warnings. There are a few extra ones in LLVMCore and MLIR.
* `OpEmitter::getAttrNameIndex`: -Wunused-function (function is private and not used anywhere)
* `PrintOpPass` copy constructor: -Wextra ("Base class should be explicitly initialized in the copy constructor")
* `LegalizeForLLVMExport.cpp`: -Woverflow (overflow is expected, silence warning by making the cast explicit)
Differential Revision: https://reviews.llvm.org/D107525
The file in which `Region::viewGraph` is defined has changed. This should have been updated with D106342.
Differential Revision: https://reviews.llvm.org/D107517
There is a case in EmitDiagnostics where the filter check is bypassed (when locationStack is empty). Filter might also be bypassed when loc instead of showableLoc is added to the locationStack.
Reviewed By: rriddle
Differential Revision: https://reviews.llvm.org/D106522
A text file may be comprised of many different "chunks", when
the input file contains the `// -----` split markers. We don't
need to use a unique MLIRContext per chunk, as having
separate contexts is intended to allow for easy unloading of
unused data and all chunks have the same lifetime (tied to the
input file). This commit uses one context for the entire file,
greatly reducing memory consumption in certain situations (up
to 70%).
Differential Revision: https://reviews.llvm.org/D107488
'mlir::DiagnosticEngine::DiagnosticEngine(const mlir::DiagnosticEngine&)' is implicitly deleted because the default definition would be ill-formed.
Reviewed By: rdzhabarov
Differential Revision: https://reviews.llvm.org/D107287
Making sure the AMX dialect webpage reads better with a short introduction on the purpose of this dialect.
Reviewed By: grosul1, bondhugula
Differential Revision: https://reviews.llvm.org/D107419
There are many downstream users of llvm::dbgs, which is defined in Debug.h. Before D106342, many users included that dependency transitively via the now deleted ViewRegionGraph.h. Adding it back to Transforms/Passes.h for convenience.
Differential Revision: https://reviews.llvm.org/D107451
* Add new pass option `print-data-flow-edges`, default value `true`.
* Add new pass option `print-control-flow-edges`, default value `false`.
* Remove `PrintCFGPass`. Same functionality now provided by
`PrintOpPass`.
Differential Revision: https://reviews.llvm.org/D106342
The existing vector transforms reduce the dimension of transfer_read
ops. However, beyond a certain point, the vector op actually has
to be reduced to a scalar load, since we can't load a zero-dimension
vector. This handles this case.
Note that in the longer term, it may be preferaby to support
zero-dimension vectors. see
https://llvm.discourse.group/t/should-we-have-0-d-vectors/3097.
Differential Revision: https://reviews.llvm.org/D103432
This test ensures that an error is generated from the Python side when running a module pass on a function. The test used to instantiate ViewOpGraph, however, this pass was changed into a general "any op" pass in D106253. Therefore, a different pass must be used in this test.
Differential Revision: https://reviews.llvm.org/D107424
* New pass option `max-label-len`: Truncate attributes/result types that have more #chars.
* New pass option `print-attrs`: Activate/deactivate rendering of attributes.
* New pass option `printResultTypes`: Activate/deactivate rendering of result types.
Differential Revision: https://reviews.llvm.org/D106337
* Visualize blocks and regions as subgraphs.
* Generate DOT file directly instead of using `GraphTraits`. `GraphTraits` does not support subgraphs.
Differential Revision: https://reviews.llvm.org/D106253
Also makes style consistent with the "surrounding"
text that appears on one webpage in MLIR doc
Reviewed By: grosul1
Differential Revision: https://reviews.llvm.org/D107418
We can propagate the shape from tosa.cond_if operands into the true/false
regions then through the connected blocks. Then, using the tosa.yield ops
we can determine what all possible return types are.
Reviewed By: jpienaar
Differential Revision: https://reviews.llvm.org/D105940
Handles shape inference for identity, cast, and rescale. These were missed
during the initialy elementwise work. This includes resize shape propagation
which includes both attribute and input type based propagation.
Reviewed By: jpienaar
Differential Revision: https://reviews.llvm.org/D105845
This prevents an explosion of threads, given that each file gets its own context and thus its own thread pool. We don't really need a thread pool for the LSP contexts anyways, so it's better to just disable threading.
mlir/test/transforms/loop-fusion.mlir is too big and is split into mlir/test/transforms/loop-fusion.mlir, mlir/test/transforms/loop-fusion-2.mlir, mlir/test/transforms/loop-fusion-3.mlir
and mlir/test/transforms/loop-fusion-4.mlir. Further tests can be added in mlir/test/transforms/loop-fusion-4.mlir
Reviewed By: bondhugula
Differential Revision: https://reviews.llvm.org/D106473
This patch adds the critical construct to the OpenMP dialect. The
implementation models the definition in 2.17.1 of the OpenMP 5 standard.
A name and hint can be specified. The name is a global entity or has
external linkage, it is modelled as a FlatSymbolRefAttr. Hint is
modelled as an integer enum attribute.
Also lowering to LLVM IR using the OpenMP IRBuilder.
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D107135
Store both interfaceID and objectID as key for interface registration callback.
Otherwise the implementation allows to register only one external model per one object in the single dialect.
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D107274
Silence clang-tidy warning in AffineOps.cpp due to the inability to see
through the typeswitch. NFC.
Differential Revision: https://reviews.llvm.org/D106125
There are cases in which it is not desirable to fully compose the bound map with the operands when adding lower/upper bounds to a `FlatAffineConstraints`.
E.g., this is the case when bounds should be expressed in terms of the operands only (and not the operands' dependencies). This also makes `addLowerOrUpperBound` useable together with operands that are defined through semi-affine expressions.
Differential Revision: https://reviews.llvm.org/D107221
Bounds such as `dim_{pos} <= c_1 * dim_x + ...` where `x == pos` are invalid. `addLowerOrUpperBound` previously added an incorrect inequality to the set. Such cases are now explicitly rejected.
Differential Revision: https://reviews.llvm.org/D107220
Add ForLoopBoundSpecialization pass, which specializes scf.for loops into a "main loop" where `step` divides the iteration space evenly and into an scf.if that handles the last iteration.
This transformation is useful for vectorization and loop tiling. E.g., when vectorizing loads/stores, programs will spend most of their time in the main loop, in which only unmasked loads/stores are used. Only the in the last iteration (scf.if), slower masked loads/stores are used.
Subsequent commits will apply this transformation in the SparseDialect and in Linalg's loop tiling.
Differential Revision: https://reviews.llvm.org/D105804
There was a slightly mismatch between the double COO and actual numerical
type in the final sparse tensor storage (due to external formats always
using double). This minor revision removes that inconsistency by using a
properly typed COO and casting during the "add" method instead. This also
prepares alternative ways of initializing the COO object.
Reviewed By: gussmith23
Differential Revision: https://reviews.llvm.org/D107310
We had a [bad bug](69655864ee) over in CIRCT
caused by accidentally passing around PatternRewriter
by value. There is no reason to support copy/assignment
of the pattern rewriter, so disable it.
Differential Revision: https://reviews.llvm.org/D107232
This patch fixes a bug in the existing implementation of detectAsFloorDiv,
where floordivs with numerator with non-zero constant term and floordivs with
numerator only consisting of a constant term were not being detected.
Reviewed By: vinayaka-polymage
Differential Revision: https://reviews.llvm.org/D107214
The `DataLayout` class currently contains the member `layoutStack` which is hidden behind a preprocessor region dependant on the NDEBUG macro. Code wise this makes a lot of sense, as the `layoutStack` is used for extra assertions that users will want when compiling a debug build.
It however has the uncomfortable consequence of leading to a different ABI in Debug and Release builds. This I think is a bit annoying for downstream projects and others as they may want to build against a stable Release of MLIR in Release mode, but be able to debug their own project depending on MLIR.
This patch changes the related uses of NDEBUG to LLVM_ENABLE_ABI_BREAKING_CHECKS. As the macro is computed at configure time of LLVM, it may not change based on compiler settings of a downstream projects like NDEBUG would.
Differential Revision: https://reviews.llvm.org/D107227
Introduces a conversion from one (sparse) tensor type to another
(sparse) tensor type. See the operation doc for details. Actual
codegen for all cases is still TBD.
Reviewed By: ThomasRaoux
Differential Revision: https://reviews.llvm.org/D107205
NFC. Clean up stale doc comments on memref replacement utility and some
variable renaming in it to avoid confusion.
Differential Revision: https://reviews.llvm.org/D107144
This allows for reusing the same output channel when the extension reloads after updating the server. Currently, whenever the extension restarts a new output channel is created (which can lead to a large number of seemingly dead output channels).
Quite a few things were out-of-date, or just not
organized well. This revision updates the extension
name, repo, icon, and many other components in
preperation for publishing the extension to the
marketplace.
If the source value to load is bool, and we have native storage
capability support for the source bitwidth, we still cannot directly
rewrite uses; we need to perform casting to bool first.
Reviewed By: hanchung
Differential Revision: https://reviews.llvm.org/D107119
If the source value to store is bool, and we have native storage
capability support for the target bitwidth, we still cannot directly
store; we need to perform casting to match the target memref
element's bitwidth.
Reviewed By: hanchung
Differential Revision: https://reviews.llvm.org/D107114
Rationale:
External file formats always store the values as doubles, so this was
hard coded in the memory resident COO scheme used to pass data into the
final sparse storage scheme during setup. However, with alternative methods
on the horizon of setting up these temporary COO schemes, it is time to
properly template this data structure.
Reviewed By: gussmith23
Differential Revision: https://reviews.llvm.org/D107001
The effect name is used by tablegen when generating the getEffects method of the SideEffectInterfaces. It is currently unqualified even though the class is contained within the mlir namespace, leading to compiler errors when using namespace mlir; isn't used before including the generated cpp file.
This patch fixes that by simply fully qualifying the class name.
Differential Revision: https://reviews.llvm.org/D107171
The presence of AffineIfOp inside AffineFor prevents fusion of the other loops to happen. For example:
```
affine.for %i0 = 0 to 10 {
affine.store %cf7, %a[%i0] : memref<10xf32>
}
affine.for %i1 = 0 to 10 {
%v0 = affine.load %a[%i1] : memref<10xf32>
affine.store %v0, %b[%i1] : memref<10xf32>
}
affine.for %i2 = 0 to 10 {
affine.if #set(%i2) {
%v0 = affine.load %b[%i2] : memref<10xf32>
}
}
```
The first two loops were not be fused because of `affine.if` inside the last `affine.for`.
The issue seems to come from a conservative constraint that does not allow fusion if there are ops whose number of regions != 0 (affine.if is one of them).
This patch just removes such a constraint when`affine.if` is inside `affine.for`. The existing `canFuseLoops` method is able to handle `affine.if` correctly.
Reviewed By: bondhugula, vinayaka-polymage
Differential Revision: https://reviews.llvm.org/D105963
When we vectorize a scalar constant, the vector constant is inserted before its
first user if the scalar constant is defined outside the loops to be vectorized.
It is possible that the vector constant does not dominate all its users. To fix
the problem, we find the innermost vectorized loop that encloses that first user
and insert the vector constant at the top of the loop body.
Reviewed By: nicolasvasilache
Differential Revision: https://reviews.llvm.org/D106609
Make broadcastable needs the output shape to determine whether the operation
includes additional broadcasting. Include some canonicalizations for TOSA
to remove unneeded reshape.
Reviewed By: NatashaKnk
Differential Revision: https://reviews.llvm.org/D106846
`PadTensorOp` has verification logic to make sure
result dim must be static if all the padding values are static.
Cast folding might add more static information for the src operand
of `PadTensorOp` which might change a valid operation to be invalid.
Change the canonicalizing pattern to fix this.
* Adds source targets (not included in the full set that downstreams use by default) to bundle mlir-c/ headers into the mlir/_mlir_libs/include directory.
* Adds a minimal entry point to get include and library directories.
* Used by npcomp to export a full CAPI (which is then used by the Torch extension to link npcomp).
Reviewed By: mikeurbach
Differential Revision: https://reviews.llvm.org/D107090
Currently TFRT does not support top-level coroutines, so this functionality will allow to have a single blocking await at the top level until TFRT implements the necessary functionality.
Reviewed By: ezhulenev
Differential Revision: https://reviews.llvm.org/D106730
Change the formatting of the debug print outs to elide unnecessary information.
Reviewed By: nicolasvasilache
Differential Revision: https://reviews.llvm.org/D106661
This is redundant with the callback variant and untested. Also remove
the callback-less methods for adding a dynamically legal op, as they
are no longer useful.
Differential Revision: https://reviews.llvm.org/D106786
This allows to use OperationEquivalence to track structural comparison for equality
between two operations.
Differential Revision: https://reviews.llvm.org/D106422
* For python projects that don't need JIT/ExecutionEngine, cuts the number of files to compile roughly in half (with similar reduction in end binary size).
Differential Revision: https://reviews.llvm.org/D106992
By making an explicit template specialization for the TypeID provided by these classes,
the compiler will not emit an inline weak definition and rely on the linker to unique it.
Instead a single definition will be emitted in the C++ file alongside the implementation
for these classes. That will turn into a linker error what is now a hard-to-debug runtime
behavior where instances of the same class may be using a different TypeID inside of
different DSOs.
Recommit 660a56956c after fixing gcc5
build.
Differential Revision: https://reviews.llvm.org/D105903
Historically the builtin dialect has had an empty namespace. This has unfortunately created a very awkward situation, where many utilities either have to special case the empty namespace, or just don't work at all right now. This revision adds a namespace to the builtin dialect, and starts to cleanup some of the utilities to no longer handle empty namespaces. For now, the assembly form of builtin operations does not require the `builtin.` prefix. (This should likely be re-evaluated though)
Differential Revision: https://reviews.llvm.org/D105149
Interop parallelism requires needs awaiting on results. Blocking awaits are bad for performance. TFRT supports lightweight resumption on threads, and coroutines are an abstraction than can be used to lower the kernels onto TFRT threads.
Reviewed By: ezhulenev
Differential Revision: https://reviews.llvm.org/D106508
By making an explicit template specialization for the TypeID provided by these classes,
the compiler will not emit an inline weak definition and rely on the linker to unique it.
Instead a single definition will be emitted in the C++ file alongside the implementation
for these classes. That will turn into a linker error what is now a hard-to-debug runtime
behavior where instances of the same class may be using a different TypeID inside of
different DSOs.
Differential Revision: https://reviews.llvm.org/D105903
In translation from MLIR to another IR, run the MLIR verifier on the parsed
module to ensure only valid modules are given to the translation. Previously,
we would send any module that could be parsed to the translation, including
semantically invalid modules, leading to surprising errors or lack thereof down
the pipeline.
Depends On D106937
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D106938
The verifier of the llvm.call operation was not checking for mismatches between
the number of operation results and the number of results in the signature of
the callee. Furthermore, it was possible to construct an llvm.call operation
producing an SSA value of !llvm.void type, which should not exist. Add the
verification and treat !llvm.void result type as absence of call results.
Update the GPU conversions to LLVM that were mistakenly assuming that it was
fine for llvm.call to produce values of !llvm.void type and ensure these calls
do not produce results.
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D106937
- Fixed symbol insertion into `symNameToModuleMap`. Insertion
needs to happen whether symbols are renamed or not.
- Added check for the VCE triple and avoid dropping it.
- Disabled function deduplication. It requires more careful
rules. Right now it can remove different functions.
- Added tests for symbol rename listener.
- And some other code/comment cleanups.
Reviewed By: ergawy
Differential Revision: https://reviews.llvm.org/D106886
Specialize the DeduplicateInputs and RemoveIdentityLinalgOps patterns for GenericOp instead of implementing them for the LinalgOp interface.
This revsion is based on https://reviews.llvm.org/D105622 that moves the logic to erase identity CopyOps in a separate pattern.
Reviewed By: nicolasvasilache
Differential Revision: https://reviews.llvm.org/D105291
Split out an EraseIdentityCopyOp from the existing RemoveIdentityLinalgOps pattern. Introduce an additional check to ensure the pattern checks the permutation maps match. This is a preparation step to specialize RemoveIdentityLinalgOps to GenericOp only.
Reviewed By: nicolasvasilache
Differential Revision: https://reviews.llvm.org/D105622
`memref.collapse_shape` has verification logic to make sure
result dim must be static if all the collapsing src dims are static.
Cast folding might add more static information for the src operand
of `memref.collapse_shape` which might change a valid collapsing
operation to be invalid. Add `CollapseShapeOpMemRefCastFolder` pattern
to fix this.
Minor changes to `convertReassociationIndicesToExprs` to use `context`
instead of `builder` to avoid extra steps to construct temporary
builders.
Reviewed By: nicolasvasilache, mravishankar
Differential Revision: https://reviews.llvm.org/D106670
By making an explicit template specialization for the TypeID provided by these classes,
the compiler will not emit an inline weak definition and rely on the linker to unique it.
Instead a single definition will be emitted in the C++ file alongside the implementation
for these classes. That will turn into a linker error what is now a hard-to-debug runtime
behavior where instances of the same class may be using a different TypeID inside of
different DSOs.
Differential Revision: https://reviews.llvm.org/D105903
RewriteEndOp was a fake terminator operation that is no longer needed now that blocks are not required to have terminators.
Differential Revision: https://reviews.llvm.org/D106911
* Implements all of the discussed features:
- Links against common CAPI libraries that are self contained.
- Stops using the 'python/' directory at the root for everything, opening the namespace up for multiple projects to embed the MLIR python API.
- Separates declaration of sources (py and C++) needed to build the extension from building, allowing external projects to build custom assemblies from core parts of the API.
- Makes the core python API relocatable (i.e. it could be embedded as something like 'npcomp.ir', 'npcomp.dialects', etc). Still a bit more to do to make it truly isolated but the main structural reset is done.
- When building statically, installed python packages are completely self contained, suitable for direct setup and upload to PyPi, et al.
- Lets external projects assemble their own CAPI common runtime library that all extensions use. No more possibilities for TypeID issues.
- Begins modularizing the API so that external projects that just include a piece pay only for what they use.
* I also rolled in a re-organization of the native libraries that matches how I was packaging these out of tree and is a better layering (i.e. all libraries go into a nested _mlir_libs package). There is some further cleanup that I resisted since it would have required source changes that I'd rather do in a followup once everything stabilizes.
* Note that I made a somewhat odd choice in choosing to recompile all extensions for each project they are included into (as opposed to compiling once and just linking). While not leveraged yet, this will let us set definitions controlling the namespacing of the extensions so that they can be made to not conflict across projects (with preprocessor definitions).
* This will be a relatively substantial breaking change for downstreams. I will handle the npcomp migration and will coordinate with the circt folks before landing. We should stage this and make sure it isn't causing problems before landing.
* Fixed a couple of absolute imports that were causing issues.
Differential Revision: https://reviews.llvm.org/D106520
The order of testing in two sparse tensor ops was incorrect,
which could cause an invalid cast (crashing the compiler instead
of reporting the error). This revision fixes that bug.
Reviewed By: gussmith23
Differential Revision: https://reviews.llvm.org/D106841
Tres Popp