Commit Graph

1115 Commits (163fa57cde44dd6a8f14e422d341d8deb4a40c62)
 

Author SHA1 Message Date
Ashay Rane 163fa57cde
torch: allow torch dialect ops after running drop-shape pass (#979)
In the `pyhpc_turbulent_kinetic_energy` TorchBench benchmark, the shape
calculation occurs inside loops, but because `DropShapeCalculationsPass`
does not explicitly mark the Torch dialect as legal, the pass execution
fails.

This patch adds Torch to the list of legal dialects, and adds a test to
validate the translation.
2022-06-25 07:27:47 -07:00
Gaurav Shukla 1be604bfd3 [LINALG] Lower `aten.Matmul` to `linalg.BatchMatmul`
This commit lowers `aten.matmul` to `linalg.BatchMatmul` under the
following conditions:
1. The result of matrix multiplication must have batch dimensions,
   i.e., rank greater than 2.
2. The resultant matrix must have at most 1 dynamic batch dimension.

It also handles broadcasting of batch dimensions when batch dimensions
of the matrices are broadcastable.

Signed-Off-by: Gaurav Shukla <gaurav@nod-labs.com>
2022-06-25 10:58:06 +05:30
Sean Silva f774e63abd development.md: updating the LLVM submodule
We recently noticed that it is necessary to update the shape lib while
doing an LLVM bump: https://github.com/llvm/torch-mlir/pull/973
We didn't have a place to document that this was a necessary step, so
add a basic section to hold that and the basic instructions for how to
do a bump.
2022-06-24 16:33:15 -07:00
Sambhav Jain 3bb2d548a5
Include Bazel BUILD steps to development.md (#954)
* Add bazel build steps

* move to development.md

* reorder
2022-06-24 11:27:22 -07:00
Ramiro Leal-Cavazos 400fecc1e5
[LINALG] Fix shape function of index.Tensor + support N-rank inputs (#972)
This commit fixes the shape function for `index.Tensor`, adding
support for multiple index tensors and `None`s in the indices
list. This commit also adds support for input tensors of rank greater
than 1. The lowering for `index.Tensor` still has the the limitation
that only a single index tensor along the first dimension of the input
tensor is supported.
2022-06-24 09:45:44 -07:00
Ashay Rane 234fc7fe0c
linalg: lower `aten.triu` op to `linalg.generic` (#965)
Prior to this patch, the torch dialect included `AtenTriuOp` for
computing the upper triangular part of the input matrix, but there was
no code for lowering the op to the linalg dialect.

This patch adds code to generate a `linalg.generic` operation that
compares indices (computed using `linalg.index`) to choose between zero
or the original value (using `arith.select`).  The lowering fails if the
number of dimensions are less than two.  This patch also adds a few
end-to-end tests.
2022-06-23 22:45:48 -07:00
Tanyo Kwok 143a7bcb76
[MLIR][TORCH] Add folder for torch_c.from_i64 & torch_c.to_i64 (#933)
* [MLIR][TORCH] Add folder for torch_c.from_i64 & torch_c.to_i64

* add unit tests for each individual fold

* fix failure of NumelZeroRankModule & TestMultipleTensorAndPrimitiveTypesReturn
2022-06-24 09:34:39 +08:00
Ramiro Leal-Cavazos 189afa82c5
Update shape library with LLVM bump changes (#973) 2022-06-23 18:13:03 -07:00
gpetters94 38650812fd
Bump LLVM (#970)
Adds my grouped conv2d op for #858
2022-06-23 14:45:57 -07:00
erman-gurses 5cff40c88a Add canonicalization for aten.add.tensor op 2022-06-23 17:24:59 -04:00
Maksim Levental e143a34948
Bump LLVM (#966) 2022-06-22 22:50:39 -05:00
Maksim Levental 829717c96e
Bump LLVM (#958) 2022-06-22 22:23:46 -05:00
Ramiro Leal-Cavazos 8b94759303
[LINALG] Fix handling of size-1 dims in `aten.view` (#962)
This commit adds support for several size-1 dims in a row in an
expansion using `aten.view`.
2022-06-22 15:58:40 -07:00
Maksim Levental a34dad2e07
Fix `verifyLinalgCompatibleTypes` which currently doesn't successfully catch `torch.tensor`. (#947) 2022-06-15 18:21:36 -05:00
Vivek Khandelwal 77ab31641f [MLIR][TORCH] Add decomposition of aten.numpy_T op
This commit adds the decomposition of `aten.numpy_T` op into
`aten.t` or `aten.permute` op.

Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2022-06-16 00:01:22 +05:30
Vivek Khandelwal 4605dc9c99 [MLIR][TORCH] Add support for bool type in convertScalarToDtype utility
Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2022-06-16 00:00:47 +05:30
Albert Sandru 708a51ae2e Add E2E support for aten.is_floating_point 2022-06-15 11:54:00 -05:00
Ramiro Leal-Cavazos 246c2df65a
[LINALG] Fix typo in conversion pattern of `aten.embedding` (#942) 2022-06-15 09:45:10 -07:00
Sean Silva 2a0c5de363 Update shape lib docs. 2022-06-15 16:29:16 +00:00
Vivek Khandelwal aed5517fda [MLIR][TORCH] Add integer dtype support for aten.rsub.Scalar op
Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2022-06-15 16:46:28 +05:30
Bob Adolf b90837ee24
Temporarily revert support for custom op extensions. (#944)
The MacOS builders are having linking trouble with the extension library.
Until it's fixed, all support for op extensions is disabled. It should be
easy to restore once the issue is resolved.
2022-06-14 18:24:40 -07:00
powderluv 8fd084377d
Update CMakeLists.txt 2022-06-14 14:46:52 -07:00
powderluv dfc6f7c547
Update CMakeLists.txt
Emergency fix to unblock the nightly Release builder
2022-06-14 14:38:35 -07:00
Ramiro Leal-Cavazos 93f6d8e776
[LINALG] Add 0-rank case for `aten.permute` (#940)
The function `AffineMap::inferFromExprList` does not work if the first
vector of expressions is empty, because it uses these expressions to
obtain the context. This prevented `aten.permute` from working for
inputs of 0-rank. This commit adds support for 0-rank inputs.
2022-06-14 12:50:46 -07:00
Vivek Khandelwal 33fa8e7761 [MLIR][TORCH] Add decomposition of aten.floor_divide op
This commit adds the decomposition of `aten.floor_divide` op into
`aten.div.Tensor_mode` op.

Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2022-06-14 08:56:25 +05:30
Tanyo Kwok 0d4445eaf9
Fix: 0 sizes tensor being regarded as unknown rank (#923) 2022-06-14 09:58:50 +08:00
Bob Adolf 0a7ba62438
Allow torch-mlir to support PyTorch extensions. (#895)
PyTorch allows new operators to be registered dynamically in modules.
Torch-mlir already makes it fairly straightforward to add support for
new operators, and this commit just extends that support to allow new
PyTorch ops to come from a external module.

This does *not* allow ops to be dynamically loaded into torch-mlir.
Torch-mlir must still be compiled with support built-in.

Add a `_torch_mlir_custom_op_example` subpackage to `torch_mlir` which
registers an demonstration op. It will not be imported by default when
importing torch_mlir. It's strictly for testing and documentation.

Adds an end-to-end test for the `torch_mlir_custom_op_example::identity` op.

With all these changes, we should now be actively testing PyTorch extension
support with all future patches.
2022-06-13 14:51:30 -07:00
powderluv 4bf34523d7
Update README.md 2022-06-10 20:00:43 -07:00
powderluv 1adc0f1661
Revert requirements.txt (#930)
https://download.pytorch.org/whl/nightly/cpu/torch_nightly.html

is going to be deprecated via pip 22 since it is not html5.
2022-06-10 15:23:12 -07:00
powderluv 02b917f769
Change to the real PackedParams.h location (#929)
Also update the PyTorch nightly URL
2022-06-10 14:43:52 -07:00
powderluv 4cdf4e7d47
Fix new location for PackedParams.h (#928)
Looks like they renamed it in location
2022-06-10 14:30:31 -07:00
Tanyo Kwok e70d4f732d
Fix class_annotator_pybind.h header guard (#924)
merging to unblock builders
2022-06-10 11:58:26 -07:00
powderluv 6615add806
Fix the new header location (#926)
Seems to have moved in the latest nightly
2022-06-10 11:57:58 -07:00
Maksim Levental 5c85ac3100
Handle `nn.Linear(..., bias=False)` case for TorchToLinalg (#919) 2022-06-08 21:13:43 -05:00
Henry Tu 298d095acf
Use double quotes instead of single quotes (#918) 2022-06-08 15:00:56 -04:00
Henry Tu c1da9edcf0
Generate underscore variant of functional ops (#915)
* Generate underscore variant of functional ops

* Do not apply `IsTrailingUnderscoreInplaceVariant` trait to underscore variant of functional op
2022-06-08 14:27:36 -04:00
Tanyo Kwok bd53998da8
Remove pybind deps from importer and annotator (#903)
* Remove pybind deps from importer and annotator
* Rename files to class_annotator_pybind.cpp/.h
2022-06-08 10:12:05 +08:00
Sean Silva e1b38e74dd Use upstream shape functions directly.
Now that upstream exposes them nicely, we can use them.

I noticed that we had added stuff into the upstream_shape_helpers.py
file (which was supposed to stay pristine), so some more shape functions
need to be upstreamed.

Going forward, all shape functions should be upstreamed similar to
https://github.com/pytorch/pytorch/pull/76889 instead of added in this
file.
2022-06-07 11:15:03 -07:00
Ramiro Leal-Cavazos 22c0893ec6
Update debug options in compilation errors (#913)
The flag for printing the IR after each pass is now prefixed with
"mlir". This commit updates the flag in the error reporting for the
compiler.
2022-06-07 10:55:54 -07:00
Vivek Khandelwal b95b3d844d [MLIR][TORCH] Add E2E support for aten.div.Tensor_mode op
This commit adds lowering of `aten.div.Tensor_mode` op.
This commit also fixes formatting for the test file elementwise.py.

Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2022-06-07 22:26:44 +05:30
Vivek Khandelwal a11ef674a7 [MLIR][TORCH] Add E2E support for aten.baddbmm op
This commit decomposes `aten.baddbmm` op into `aten.bmm`,
`aten.mul.Scalar`, and `aten.add.Tensor` op.

Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2022-06-07 22:26:28 +05:30
Jae Hoon (Antonio) Kim fe784fd900
Add Support for aten::scatter_add (#906) 2022-06-06 15:02:45 -04:00
Jae Hoon (Antonio) Kim 8a1839a17e
Add support for aten::arange.start_out (#905) 2022-06-06 15:02:27 -04:00
Vivek Khandelwal 2718b4d838 [MLIR][TORCH] Add E2E support for aten.clamp_[min|max] op
This commit decomposes `aten.clamp_min` and `aten.clamp_max` op
into `aten.clamp` op.

Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2022-06-06 11:52:29 +05:30
Sean Silva ccc858f531 torch_mlir.compile: Fix API footgun
use_tracing=True was behaving unexpectedly because the handling of
single arguments was happening after the torch.jit.trace call.

This also fixes the check to specifically test for a torch.Tensor or
TensorPlaceholder so that both lists and tuples would be correctly
handled.
2022-06-05 18:10:07 -07:00
powderluv 2f0b1d0b08
bump macOS builds to Python 3.10 2022-06-04 22:44:32 -07:00
powderluv b14c5d619d
Build the nightly package only once a day/night
No need to be shipping two releases a day, our supported packages and binaries have grown.
2022-06-04 22:40:53 -07:00
Vidush Singhal fc419b1e7d
Add E2E support for AtenLogicalOrOp. (#883) 2022-06-03 16:21:03 -07:00
Henry Tu abf5c94a1b
Replace valsem.aten.zero with aten.zero.functional (#893) 2022-06-03 16:27:31 -04:00
Henry Tu 650f5a5008
Added support for native_dropout_backward (#892) 2022-06-03 14:08:51 -04:00