Add VerbosePassInstrumentation to the second pass manager of the arcilator tool to include its s...

To clearly separate semantics, define a bit-vector type and attribute instead of reusing the bui...

We have an existing check that paths are in the same owning module as
the entity they target. T...

Fix a missing CMake dependency on the Emit dialect for the build rules for
DebugInfo.

h/t @fzi-...

Fix a missing dialect registration for the DebugInfo translation. This
fixes an error if trying...

Splat operators required the output is 1-D. This was not a required
restriction and was loosene...

This commit provides dummy solution to support elmentwise operations
(mul, add) with scalar arg...

Reduce mean lowerings did not succesfully lower to `linalg` via torched.
There were two separat...

Instead of just dedup-ing the external memory module, include the
memory wrapper module in the ...

Updating to Apple Clang 1500.3.9.4 fixes the problem that was causing
test failures.

Build/t...

- Contribution from Technical Writing team at compiler-research.org

- I have tested this locall...

- renamed the article and filename

- removed GSoC references

- removed GSoC intro

- added a b...

* torch-mlir may not handle `aten.warn`. But it could be handled by
custom users' backend which...

If the broadcast shape is length-1 at a dim while `?` in the input dim
then we need to broadcas...

This folds small version of the tensor-scalar comparison operators as
they are commonly used fo...

uses version checking since attributes exist in both versions, the only
thing that changes is w...

Fix an issue in the parser of the `firrtl.constant` op, which would
truncate the sign bit of ne...

This matches what builkite is doing right now.

Also add myself to the notification for this bot.

The file was renamed by PR https://github.com/llvm/torch-mlir/pull/2842.

These targets are useful for downstream projects that want to install
all CIRCT libraries. The ...

Both the upstream MLIR IntegerAttr and OM IntegerAttr are backed by an
arbitrary precision inte...

Most interesting arithmetic on APSInt asserts that both operands are
the same bitwidth, but the...

cc #6715.

Previous gemm version required a bias vector.
This provides an alternate path to `Torch::AtenM...

* [Fortran/gfortran] Enable passing tests

These tests now pass. Some tests that only pass on ...

The current padding operation was not functional for dynamic shapes.
Updated and enabled tests ...

Spec: https://onnx.ai/onnx/operators/onnx__Clip.html

Set PyTorch and TorchVision version to nightly release 2024-03-07.
This commit also removes the...

This moves LowerIntrinsic to preprocess transforms.

Co-authored-by: wenyangwang <[email protected]>

This picks up a bugfix in DialectConversion:
https://github.com/llvm/llvm-project/commit/960665...

Since cibuildwheel doesn't yet support Python 3.13, comment it out until support gets added.

RewriterBase::replaceAllUsesWith is not currently supported by
DialectConversion. Using it may l...

We can support `onnx.Size` by requesing the size of each dimensions and
taking the product of t...

This mostly copy-pastes the reduce minimum implementation to reduce max
to improve test coverag...

The DialectConversion infrastructure now asserts that replaced ops are
not modified, including i...

The modifyOpInPlace seems to have been used to give the
PartialLoweringFunc carte blanche to mak...

Updating PowerPC bots to use the upstream CMake support for ccache.
Replace usage of LLVM_CCACH...

The addition of an e2e test is actually provided in the Shark-Testsuite.
This adds 2 test cases...

Current implementation depends on using `aten.view` which has issues
inferring tensor collapse/...

It used to fail due to the missing runtime support, but this
was not considered a failure.
It ...

Add e2d support for `aten.linalg_norm` by decompose it to
`aten.linalg_vector_norm`.

Lowerin...

It used to fail due to the missing runtime support, but this
was not considered a failure.
It ...

Convert to individiual slices and tuple together as a list.

---------

Co-authored-by: Scot...

In some OM dialect constructs, it is possible to receive
EvaluatorValue::Reference values. In t...

`getRawBuffer` expects a densely packed vector of `i1` values however
`onnx` does not densely p...

Two build-id related LLD tests are failing on the MacOS builder I am
trying to bring up apparen...

Enable passing additional lit arguments per bot to the `ninja
check-openmp` configuration.
Thi...

I am shuffling around the builder assignments for a few workers, so
needed to remove the hard c...

Add a new release builder for the AArch64 Darwin target, and a new
worker to share build respon...

* that `dataclass`'s `slots` is supported after python 3.10.

* LLVM bump to f75c6ed93e785c09884a317ce2bfd440e7f8f573

A handful of operations are commonly used in shape calculations (slice,
concat, broadcast). Add...

Alternatives to https://github.com/llvm/torch-mlir/pull/2908

Fix https://github.com/nod-ai/SH...

Existing lowering via aten.view does not work as well for dynamic shapes
as the lowering to ten...

In https://github.com/llvm/circt/pull/6582, initial support for
classes and objects was added i...

We assumed rank was atleast 1 however it can be rank-0, generating an
illegal pair of flatten /...

This change also rewrites the `verilator` macro to `VERILATOR` to keep it consistent with other ...

Macro identifiers now use a proper symbol to handle references.
The verifiers are not yet enabl...

WireDFT is no longer used within SiFive. No issues popped up while the pass could be optionally ...

Add continuous, blocking, and non-blocking assignments to the Moore
dialect. These represent th...

Remove possibly confusing language on the conditions for why a wheel was
uploaded. This can hap...

Add an input to the workflow_dispatch even of the "Upload Wheels" GitHub
Action workflow. This ...

Change two usages of private friend members to use public member
functions. This is mildly clea...

Add a member function for recording a mapping of old FIRRTL module to new
HW module. Use this t...

Fixes a bug where memories instantiated under the DUT would be moved to
the testbench directory....

Add the emit dialect to the C-API and to Python. This is both missing and
is necessary for down...

Remove use of designated initializers. This is a C++20 feature and the
Windows builds are unhap...