This commit lowers `aten.matmul` to `linalg.BatchMatmul` under the
following conditions:
1. The ...

We recently noticed that it is necessary to update the shape lib while
doing an LLVM bump: https...

* Add bazel build steps

* move to development.md

* reorder

This commit fixes the shape function for `index.Tensor`, adding
support for multiple index tens...

Prior to this patch, the torch dialect included `AtenTriuOp` for
computing the upper triangular...

* [MLIR][TORCH] Add folder for torch_c.from_i64 & torch_c.to_i64

* add unit tests for each in...

Adds my grouped conv2d op for #858

This commit adds support for several size-1 dims in a row in an
expansion using `aten.view`.

This commit adds the decomposition of `aten.numpy_T` op into
`aten.t` or `aten.permute` op.

Sig...

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

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

The MacOS builders are having linking trouble with the extension library.
Until it's fixed, all...

Emergency fix to unblock the nightly Release builder

The function `AffineMap::inferFromExprList` does not work if the first
vector of expressions is...

This commit adds the decomposition of `aten.floor_divide` op into
`aten.div.Tensor_mode` op.

Si...

PyTorch allows new operators to be registered dynamically in modules.
Torch-mlir already makes ...

https://download.pytorch.org/whl/nightly/cpu/torch_nightly.html

is going to be deprecated via...

Also update the PyTorch nightly URL

Looks like they renamed it in location

merging to unblock builders

Seems to have moved in the latest nightly

* Generate underscore variant of functional ops

* Do not apply `IsTrailingUnderscoreInplaceVa...

* Remove pybind deps from importer and annotator
* Rename files to class_annotator_pybind.cpp/.h

Now that upstream exposes them nicely, we can use them.

I noticed that we had added stuff into ...

The flag for printing the IR after each pass is now prefixed with
"mlir". This commit updates t...

This commit adds lowering of `aten.div.Tensor_mode` op.
This commit also fixes formatting for th...

This commit decomposes `aten.baddbmm` op into `aten.bmm`,
`aten.mul.Scalar`, and `aten.add.Tenso...

This commit decomposes `aten.clamp_min` and `aten.clamp_max` op
into `aten.clamp` op.

Signed-Of...

use_tracing=True was behaving unexpectedly because the handling of
single arguments was happenin...

No need to be shipping two releases a day, our supported packages and binaries have grown.

The patch bumped up the LLVM tag made manual fixes to the code in
`ShapeLibrary.cpp`. However,...

This commit adds the decomposition of `aten.adaptive_avg_pool2d` op into
`aten.avg_pool2d` op. T...

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

The op `aten.unsqueeze` has a mutating variant. This commit adds
support for that variant.

When compiling without assertions (i.e. in `NDEBUG` mode), a handful of
statements turn to NOPs...

This commit adds lowering of `aten.Bool.float` and `aten.Bool.int` op.

Signed-Off By: Vivek Kha...

This commit adds lowering of `aten.any.bool` op.

Signed-Off By: Vivek Khandelwal <vivek@nod-lab...

This commit adds lowering of `aten.sqrt.int` op.

Signed-Off By: Vivek Khandelwal <vivek@nod-lab...

There is more detail to be added here, but this is a start.

This patch adds support for the torch.linalg.vector_norm op to the torch
dialect, including the...

In cases where a refinement/derefinement was needed, we didn't fold.

Fixes https://github.com/l...

This can help with development and reporting bugs.

1. With the help of `make_fx` we are able to get the full training graph
with weight updates....

We do this by inroducing a TensorPlaceholder class, which can be used to
specify dynamic sizes. ...

In addition to updating the llvm-project submodule, this patch also:

1. updates shape functio...

GH runner times out when building 3.9 and 3.10 on macOS.

Previously only system default versions were built. Now we build
binaries for both 3.9 and 3.10

The op `aten.rand_like` was missing a shape function, unit tests, and
the `dtype` argument was ...

Fix up wheels per python version on OSX

A user might want to avoid the extra layer of multiprocessing libary for
debugging purpose. In s...

This commit adds support for aten.max_pool2d, aten.max_pool2d_with_indices,
and aten.avg_pool2d ...

This commit adds lowering of `aten.ScalarImplicit` and `aten.Int.Scalar` op.

Signed-Off By: Viv...

`aten.detach` op is folded and returns the first operand since it's an
identity function(kind of...

The preserve memory specifies that `If any of the input tensors is in channels_last format,
oper...

typo

Fix the type promotion code for scalar only operation to return
TorchType which is the type trac...

Signed-Off By: Vivek Khandelwal [email protected]

This change makes the e2e testing parallel using the multiprocessing
python module.

This commit adds lowering of `aten.ge.float`, `aten.ge.float_int`,
`aten.ne.float_int`, `aten.gt...

The main changes are:
- Added `ValueKnowledge.scalarType` to track scalar type information.
- Ad...

Set default OSX SDK to 11.0 not 11:0

- This commit adds support for `aten.mean.dim` op.
- It also adds a new test script `stats.py` f...

It would change the result type.

This also has a fix for the adjustment of types of TupleConstruct
inputs, which I found when usi...

This commit decomposes `aten.to.dtype_layout` op into `aten.to.dtype` op.
This commit also fixes...

This commit adds lowering of `aten.masked_fill.Scalar` op.
This commit also fixes the formatting...

It belongs in the development docs.

Useful when we are only building for one architecture.