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
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 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
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 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
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
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
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
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
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
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 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
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
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
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.
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
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
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
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
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
Morten Borup Petersen