prims.expand expects the start and end dimensions to be strictly less
than the rank of the tensor.

Implemented ONNX.Range. The spec says the data type for start, limit,
delta are 0-D can be doub...

Co-authored-by: Lisa Liu <[email protected]>

Handles the multiple cases of `onnx` constant values and converts them
to `torch` literal tenso...

- Add fixes for
https://github.com/llvm/llvm-project/commit/af78e5daf0791135485dbd7972ffedb9277...

This property is used for `ninja -j<jobs>` and given that we manage the
CPU usage of our machin...

* Add find compiler-rt libs cmake module for windows

This patch adds find compiler-rt libs fu...

This patch makes some minor adjustments to top level CMakeLists.txt
and CopyDir, Testsuite modu...

Add Tutorial Development with Clang-Repl(GSoC'23 blog)

Handle both `torch.dequantize` and `torch.quantize_per_tensor` including
the op based quantizat...

libunwind is now required to build libcxx and libcxxabi by default.

See llvm-project:8f90e6937a...

Add Maxpool ONNX op support.
Add Utils.h/cpp files to create a constant int list for ONNX.

This should fix the build which started failing after [1],
see for example https://lab.llvm.org...

Add support for representing an optional layer in each probe type. This
only handles storage an...

Several functions had to be pessimized when converting to module type. Start moving them back to...

libc++ use it by default.

llvm/llvm-project#77689 is not a fix.

This reverts commit 8202ffdd47b2197057cfa14c8f2153a20d3...

Spelling out latency should make it easier to understand what this
attribute means.

Looks like llvm/llvm-project#77689 fixed the issue.

This reverts commit 23cd70e66807ccdb8e75512...

llvm/llvm-project#77687 enabled libunwind, but
not all tests pass with sanitizers.

So disabl...

This patch makes minor changes to ignore windows line endings when
Fortran tests are run on win...

Logos for the MLIR Sparsifier project, combining the MLIR logo with the
concept of a sparse ten...

See https://github.com/llvm/llvm-project/pull/77687 for more details.

When the conversion from FSM to SV was added to the CAPI, an include
of the circt/Conversion/Pa...

This PR updates the torch-to-tosa conversion with following changes:

- Support torch.none as ...

* [tools] Fix not utility to work on iOS

This addresses issue #77137. The implementation uses...

Closes nod-ai/SHARK-Turbine#290

This patch implements timeit timing tool found at tools/timeit.c for Windows.
It is required fo...

This simplifies the `arc.state` operation by always requiring latency > 0 and uses `arc.call` op...

Layer Normalization using the torch.aten.native_layer_norm

https://github.com/nod-ai/SHARK-T...

Change a test to be called "layers.fir" and not "groups.fir". Change it
internally to use layer...

This change seems to make the test problem go away.

In Python3.8, must use `typing.Dict` to subscript types.

Worked until now, but the upcoming llvm bump doesn't like it.

maps onnx.expand to torch aten broadcast_to, three tests added

---------

Co-authored-by: K...

We are running out of space in PyPI, so this removes support for
versions Python 3.9 and Python ...

Signed-off-by: Schuyler Eldridge <[email protected]>

Set PyTorch and TorchVision version to nightly release 2024-01-09.

Signed-Off By: Vivek Khandelw...

Signed-Off By: Vivek Khandelwal <[email protected]>

Signed-Off By: Vivek Khandelwal <[email protected]>

Signed-Off By: Vivek Khandelwal <[email protected]>

This adds a sanity-check bot to the AMDGPU OpenMP Offloading bot
infrastructure. It is a copy o...

Adaptive pooling ops can only be decomposed into their non-adaptive
counterparts in trivial cas...

Dummy internal paths must be added to expanded ports.

We intended to keep the InstanceGraph up to date, but when we
completely erase FIRRTL ClassOps,...

Floating-point literals should use the correct type specifier.

Also add "sparsifier" for "sparse compiler" and
changes layout in open projects to make it more...

The lowering for `aten.clamp` did not support integer types. Added
support for integer types in...

This enables an actual 8-way parallel testing of libc for GPU on AMDGPU.
We limit the number of...

https://github.com/llvm/llvm-project/issues/77137

This bot adds clang and flang frontends to the OpenMP runtime on
AMDGPUs. Given the current sta...

Add a utility in FIRRTLOps.cpp for getting an optional attribute from an
array of named attribut...

Since
https://github.com/llvm/llvm-project/commit/4983432f17eb4b445e161c5f8278c6ea4d5d1241,
ty...

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

Use a unique build directory for each build to avoid incremental build
correctness issues while...

The annotated builder sets BUILDBOT_CLOBBER to empty string when clobber
isn't requested so we ...

Signed-off-by: Schuyler Eldridge <[email protected]>

Fix an issue in the `LowerSignatures` pass where instances would discard
their location and inh...

This is workaround for reverts like this
https://github.com/llvm/llvm-project/commit/0c23163184...

Match what we use in CI.