The name of the method was confusing. It is bufferizesToMemoryWrite, but from the perspective of OpResults.
`bufferizesToMemoryWrite(OpResult)` now supports ops with regions that do not have aliasing OpOperands (such as `scf.if`). These ops no longer need to implement `isMemoryWrite`.
Differential Revision: https://reviews.llvm.org/D141684
Currently, all the non-stable sorting algorithms are implemented via the
straightforward quick sort. This will be fixed in the following PR.
Reviewed By: aartbik
Differential Revision: https://reviews.llvm.org/D142678
Previously, we choose the value at (lo + hi)/2 as a pivot for partitioning the
data in [lo, hi). We now choose the median for the three values at lo, (lo +
hi)/2, and (hi-1) as a pivot to match the std::qsort implementation.
Reviewed By: aartbik
Differential Revision: https://reviews.llvm.org/D142679
This patch moves some utils into CodegenEnv class, it should make the code easier to follow and it eliminates several indirect value assignment that use `ptr**`.
Reviewed By: aartbik
Differential Revision: https://reviews.llvm.org/D142040
The handling of unknown ops will be tightened in a subsequent change. All sparse_tensor ops should implement BufferizableOpInterface, otherwise, they are treated as "unknown" and additional buffer allocs/copies may be inserted around them.
Differential Revision: https://reviews.llvm.org/D142005
The bulk of D142074 seems to have gotten overwritten due to some sort of merge conflict (afaict there's no record of it having been reverted intentionally). So this commit redoes those changes. In addition to the original changes, this commit also:
* moves the definition of `getRankedTensorType` (from `Transforms/CodegenUtils.h` to `IR/SparseTensor.h`), so that it can be used by `IR/SparseTensorDialect.cpp`.
* adds `getMemRefType` as another abbreviation.
Reviewed By: aartbik
Differential Revision: https://reviews.llvm.org/D142503
Use SparseTensorDescriptor whenever not calling setters, to avoid needing to create a temporal buffer for simple query purposes.
Reviewed By: bixia, wrengr
Differential Revision: https://reviews.llvm.org/D141953
Previously, we rely on InsertOp to add values to the result, in the same way we
add values to a sparse tensor with compressed dimensions. We now direct store
values to the values buffer.
Reviewed By: Peiming
Differential Revision: https://reviews.llvm.org/D141517
The patch adds operations to `BlockAndValueMapping` and renames it to `IRMapping`. When operations are cloned, old operations are mapped to the cloned operations. This allows mapping from an operation to a cloned operation. Example:
```
Operation *opWithRegion = ...
Operation *opInsideRegion = &opWithRegion->front().front();
IRMapping map
Operation *newOpWithRegion = opWithRegion->clone(map);
Operation *newOpInsideRegion = map.lookupOrNull(opInsideRegion);
```
Migration instructions:
All includes to `mlir/IR/BlockAndValueMapping.h` should be replaced with `mlir/IR/IRMapping.h`. All uses of `BlockAndValueMapping` need to be renamed to `IRMapping`.
Reviewed By: rriddle, mehdi_amini
Differential Revision: https://reviews.llvm.org/D139665
Previously, we rely on the InsertOp to gradually increase the size of the
storage for all sparse tensors. We now allocate the full size values buffer
for annotated all dense tensors when we first allocate the tensor. This avoids
the cost of gradually increasing the buffer and allows accessing the values
buffer as if it were a dense tensor.
Reviewed By: Peiming
Differential Revision: https://reviews.llvm.org/D141516
Move the functionality to codegen utils for sharing with the codegen path.
Reviewed By: Peiming
Differential Revision: https://reviews.llvm.org/D141514
Previously, we generate AOS subviews for indices buffers when constructing an
immutable sparse tensor descriptor. We now only generate such subviews when
getIdxMemRefOrView is requested.
Reviewed By: Peiming
Differential Revision: https://reviews.llvm.org/D141325
Previously, we use a temporary tensor with identity ordering. We now use a
temporary tensor with the destination dimension ordering, to enable the use of
sort_coo for sorting the tensor.
Reviewed By: Peiming
Differential Revision: https://reviews.llvm.org/D141295
The template parameter is no longer needed after MutSparseTensorDescriptor
is implemented as a subclass of SparseTensorDescriptorImpl. The only purpose
for it was to enable SFINAE.
Reviewed By: bixia
Differential Revision: https://reviews.llvm.org/D141303
This patch fixes build failure due to -Wsign-compare in sparse2SparseRewrite(...) after https://reviews.llvm.org/D140871.
```
llvm-project/mlir/lib/Dialect/SparseTensor/Transforms/SparseTensorRewriting.cpp:842:32: error: comparison of integers of different signs: 'uint64_t' (aka 'unsigned long') and 'int64_t' (aka 'long') [-Werror,-Wsign-compare]
for (uint64_t i = 0; i < rank; i++) {
~ ^ ~~~~
1 error generated.
```
Reviewed By: MaskRay
Differential Revision: https://reviews.llvm.org/D141104
Use an array of structures to represent the indices for the tailing COO region
of a sparse tensor.
Reviewed By: aartbik
Differential Revision: https://reviews.llvm.org/D140870
All members are now private and access is through delegate
or convenience methods only (except the loop emitter, which
is still under refactoring).
Reviewed By: Peiming
Differential Revision: https://reviews.llvm.org/D140519
Also, as a proof of concept, all functionality related to reductions
has been refactored into private fields and a clean public API. As a
result, some dead code was found as well. This approach also simplifies
asserting on a proper environment state for each call.
NOTE: making all other fields private and migrating more methods into
this new class is still TBD in yes another next revision!
Reviewed By: Peiming
Differential Revision: https://reviews.llvm.org/D140443
This cleans up a lot of parameter passing. It also prepares adding
proper "delegate" functions to the new environment and moving this
out into its own class with a better OO design.
Reviewed By: Peiming
Differential Revision: https://reviews.llvm.org/D140257
std::optional::value() has undesired exception checking semantics and is
unavailable in older Xcode (see _LIBCPP_AVAILABILITY_BAD_OPTIONAL_ACCESS). The
call sites block std::optional migration.
std::optional::value() has undesired exception checking semantics and is
unavailable in older Xcode (see _LIBCPP_AVAILABILITY_BAD_OPTIONAL_ACCESS). The
call sites block std::optional migration.
This is part of an effort to migrate from llvm::Optional to
std::optional. This patch changes the way mlir-tblgen generates .inc
files, and modifies tests and documentation appropriately. It is a "no
compromises" patch, and doesn't leave the user with an unpleasant mix of
llvm::Optional and std::optional.
A non-trivial change has been made to ControlFlowInterfaces to split one
constructor into two, relating to a build failure on Windows.
See also: https://discourse.llvm.org/t/deprecating-llvm-optional-x-hasvalue-getvalue-getvalueor/63716
Signed-off-by: Ramkumar Ramachandra <r@artagnon.com>
Differential Revision: https://reviews.llvm.org/D138934
When concat along dim 0, and all inputs/outputs are ordered with identity dimension ordering,
the concatenated coordinates will be yield in lexOrder, thus no need to sort.
Reviewed By: aartbik
Differential Revision: https://reviews.llvm.org/D140228
