when lowering the linalg conv operation into scf, then bufferized, there
is a cast operation on an allocated buffer to the type of a
memref with different strides (casued by the slice). This incompatible
cast makes the lowering fails.
Commit `45577fb79e7bef9d6874906971f485ca6ed111b8` introduced a
regression causing garbage to be generated for IR involving
`tensor.extract_slice` and `tensor.collapse` in some cases. This
commit adds two tests checking for the absence of the regression.
Co-authored-by: Umut Sahin <umut.sahin@zama.ai>
Rebase to llvm-project at 3f81841474fe with a pending upstream patch
for arbitrary element types in linalg named operations.
Co-authored-by: Ayoub Benaissa <ayoub.benaissa@zama.ai>
The function `tensorDataFromMemRef` reading elements from a result
memref and copying them to an instance of `TensorData` fails to handle
memrefs with non-zero offsets as it adds the offset twice when
calculating indexes. This may result in out-of-bounds memory accesses
and incorrect results.
This patch fixes the calculation of indexes by removing the second
addition of the offset.
Pass all arguments starting with `command` passed to
`add_concretelang_doc` to `add_mlir_doc`. This allows for the
specification of additional command line arguments to be passed to
`mlir-tblgen`.
This commit is introduced because python bindings for `tensor.from_elements` are not generated automatically. Previously, we overcame this with string manipulation, but with the latest version of the compiler, it became a problem. This commit should be reverted eventually. See https://discourse.llvm.org/t/cannot-create-tensor-from-elements-operation-from-python-bindings/4768 for the discussion in LLVM forums.
This commit rebases the compiler onto commit f69328049e9e from
llvm-project.
Changes:
* Use of the one-shot bufferizer for improved memory management
* A new pass `OneShotBufferizeDPSWrapper` that converts functions
returning tensors to destination-passing-style as required by the
one-shot bufferizer
* A new pass `LinalgGenericOpWithTensorsToLoopsPass` that converts
`linalg.generic` operations with value semantics to loop nests
* Rebase onto a fork of llvm-project at f69328049e9e with local
modifications to enable bufferization of `linalg.generic` operations
with value semantics
* Workaround for the absence of type propagation after type conversion
via extra patterns in all dialect conversion passes
* Printer, parser and verifier definitions moved from inline
declarations in ODS to the respective source files as required by
upstream changes
* New tests for functions with a large number of inputs
* Increase the number of allowed task inputs as required by new tests
* Use upstream function `mlir_configure_python_dev_packages()` to
locate Python development files for compatibility with various CMake
versions
Co-authored-by: Quentin Bourgerie <quentin.bourgerie@zama.ai>
Co-authored-by: Ayoub Benaissa <ayoub.benaissa@zama.ai>
Co-authored-by: Antoniu Pop <antoniu.pop@zama.ai>
