Commit Graph

1770 Commits (cc819e73dd5a48162dcfc099bbdf1b54ca6831da)
 

Author SHA1 Message Date
Chi_Liu cc819e73dd
[TOSA] Fix broadcast_to input and output different shape support (#1855) 2023-02-09 09:15:14 -08:00
Roll PyTorch Action 83534370c3 update PyTorch version to 2.0.0.dev20230209 2023-02-09 15:23:33 +00:00
Roll PyTorch Action eb7bac7ca4 update PyTorch version to 2.0.0.dev20230208 2023-02-08 16:14:29 +00:00
Vivek Khandelwal 282c4e027b build: manually update PyTorch version
Set PyTorch and TorchVision version to nightly release 2023-02-07.
Fixes https://github.com/llvm/torch-mlir/issues/1792.

Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2023-02-08 11:41:58 +05:30
Zachary Cetinic 2a4a61f98f
Add aten.scatter_reduce op definition (#1846) 2023-02-07 21:59:07 +00:00
Tanyo Kwok 3ebe5a5a67
build: update llvm tag to ba8b8a73f (#1856)
* build: update llvm tag to ba8b8a73f

* skip ConvolutionModule2DGroups_basic
2023-02-07 23:25:59 +08:00
Vivek Khandelwal 3e60e6021f build: manually update PyTorch version
Set PyTorch and TorchVision version to nightly release 2023-02-06.

Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2023-02-07 12:02:18 +05:30
Yuanqiang Liu 089018b658
[MHLO] move AtenTanhOp to ConvertAtenUnaryFPOnlyPatten and add sin/cos/ceil/floor pattern (#1847) 2023-02-06 11:14:26 -08:00
Vivek Khandelwal c957cebd03 build: manually update PyTorch version
Set PyTorch and TorchVision version to nightly release 2023-02-05.

Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2023-02-06 13:23:28 +05:30
Zachary Cetinic 2c2009a13d
Add in-place variant of torch.scatter_add (#1836) 2023-02-03 17:54:28 +00:00
Jiahao Li f58ba19448
Add aten.bucketize op and its decomposition (#1834) 2023-02-03 10:20:47 +08:00
Roll PyTorch Action 5d55390111 update PyTorch version to 2.0.0.dev20230202 2023-02-02 15:06:52 +00:00
Ashay Rane 711646d095
mhlo: migrate conversion to stablehlo (#1840)
This patch replaces all MHLO operations with their StableHLO
counterparts and adds a validation pass to ensure that no MHLO operations
remain before translating all Stablehlo operations to the MHLO dialect
for further lowering to the Linalg dialect.

This patch also updates all lit tests so that they refer to the
`convert-torch-to-stablehlo` pass and so that they check for StableHLO
operations.
2023-02-02 07:29:47 -06:00
Vivek Khandelwal ed9d8d1fb7 [MLIR][TORCH] Add support for clone op with channels last memory format
Fixes https://github.com/llvm/torch-mlir/issues/1829

Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2023-02-02 16:04:42 +05:30
Ashay Rane 95361747c2
Revert "update PyTorch version to 2.0.0.dev20230201" (#1842)
This reverts commit ca400e02ce, since the
(upstream) Windows build of torchvision has an incorrect dependency.
2023-02-01 13:43:53 -06:00
Roll PyTorch Action ca400e02ce update PyTorch version to 2.0.0.dev20230201 2023-02-01 16:27:37 +00:00
Roll PyTorch Action 5057e88715 update PyTorch version to 2.0.0.dev20230131 2023-01-31 14:40:23 +00:00
Sean Silva 72fbf316b4 Update LLVM and MHLO submodules.
Week of 01/30/2023:

Green LLVM commit: e31ee6417c33a6e2f0e8440b1a86d5365279ad68
Green MHLO commit: c2a6f4064d426567b9ef7b0d29d5ab86dc7b2b02 (branch greencommit/2023-01-30-e31ee641)
2023-01-31 06:08:21 -08:00
Roll PyTorch Action c622f59300 update PyTorch version to 2.0.0.dev20230130 2023-01-30 14:33:24 +00:00
Jiahao Li f5b689e12f
[MHLO] Support aten.cumsum op in mhlo backend (#1825) 2023-01-29 21:38:27 -08:00
Roll PyTorch Action 6040d7ce00 update PyTorch version to 2.0.0.dev20230129 2023-01-29 14:32:31 +00:00
Roll PyTorch Action ef4316e046 update PyTorch version to 2.0.0.dev20230128 2023-01-28 14:34:24 +00:00
Chi_Liu 00fc14a6e1
[TOSA] Add to.dtype i1 to i64 (#1830) 2023-01-27 09:21:06 -08:00
Roll PyTorch Action 31ffb815e4 update PyTorch version to 2.0.0.dev20230127 2023-01-27 14:25:52 +00:00
Roll PyTorch Action 37a625e737 update PyTorch version to 2.0.0.dev20230126 2023-01-26 14:56:31 +00:00
Yi Wang 27a31157b9
Update README.md and put Python 3.10 a prerequisite (#1821)
* Update README.md and put Python 3.10 a prerequisite

* Update README.md

* Follow up with comments

* Update README.md
2023-01-25 11:32:50 -08:00
Roll PyTorch Action 506aceae7f update PyTorch version to 2.0.0.dev20230125 2023-01-25 17:08:43 +00:00
Matthias Gehre adaf05f03e
[TorchToLinalg] Lower AtenRoundOp to math::RoundEvenOp (Fixes #1811) (#1823)
[TorchToLinalg] Lower AtenRoundOp to math::RoundEvenOp (Fixes #1811)
2023-01-25 08:51:29 +01:00
Gleb Kazantaev 3930588a7e
Enable VerifyBackendContract in LTC backend (#1798)
* Enable VerifyBackendContract in LTC backend

* Update VerifyBackendContract pass

* Move convert_scalar_implicit to jit_utils

* Rename VerifyBackendContract to VerifyBackendContractNoDecompositions

* Update verify-backend-contract-error.mlir test
2023-01-24 22:14:17 -05:00
Ramiro Leal-Cavazos 6c86bec04f
build: update llvm tag to 9acc2f37 (#1828)
This commit makes the following changes:

- Update dialects to use fold API `kEmitFoldAdaptorFolder` and update
signature of `fold` methods (see PSA
https://discourse.llvm.org/t/psa-new-improved-fold-method-signature-has-landed-please-update-your-downstream-projects/67618)
- Replace `makeArrayRef` with `ArrayRef` (see
https://reviews.llvm.org/D140896)
- Remove `TypeRange{}` arg from `b.create<scf::IfOp>` since builder no
longer takes that argument
- Make `func`s in `Torch/invalid.mlir` private, since symbol
declarations cannot be public. (see https://discourse.llvm.org/t/rfc-symbol-definition-declaration-x-visibility-checks/2140)
2023-01-25 01:29:42 +00:00
Roll PyTorch Action 8ce2fffca5 update PyTorch version to 2.0.0.dev20230124 2023-01-24 16:24:40 +00:00
Gleb Kazantaev aa3a88c8d9
Fix JIT schema matching for when ListType is used (#1826) 2023-01-23 21:43:18 -05:00
Eric Kunze 95bdfaa9bf
update llvm to d23516e9ad477527a9db4d06b1fa9566680ac67c (#1812)
Rename BlockAndValueMapping to IRMapping
Moved PrimTupleConstructOp type validation to its own verifier as the
tablegen version does not work for a combination of variadic input and
non-variadic output.
2023-01-23 16:34:22 -08:00
Ashay Rane aefa7be6fa
CI: build torch-mlir python package for Python v3.8 (#1827)
Previously, torchvision had not released WHL files for Python v3.8,
causing failures in torch-mlir python package builds, so we had disabled
building for Python v3.8.

Now that the WHL files are back, this patch re-enables v3.8 builds.
2023-01-23 17:58:01 -06:00
Roll PyTorch Action c29c07b29b update PyTorch version to 2.0.0.dev20230123 2023-01-23 18:10:24 +00:00
Vivek Khandelwal 23aa6903f7 [torchdynamo] Add default decomposition for ops in the dynamo backend
Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2023-01-23 13:33:50 +05:30
Vivek Khandelwal bf4c77cbc5 build: manually update PyTorch version
Set PyTorch and TorchVision version to nightly release 2023-01-22.

Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2023-01-23 13:33:50 +05:30
powderluv 556620e349
Use delvewheel to package windows DLLs (#1820) 2023-01-22 02:47:26 -08:00
Victor Guerra 1cf09a6a34
Using `ArrayRef` deduction guides. (#1809)
`llvm::makeArrayRef` is now deprecated and can be
replaced by the newly introduced `ArrayRef` deduction guides.

Fixes: #1808

Co-authored-by: Victor Guerra <vm.guerramoran@criteo.com>
2023-01-21 08:58:07 -06:00
Roll PyTorch Action 68ac2ce655 update PyTorch version to 2.0.0.dev20230121 2023-01-21 14:27:15 +00:00
Chi_Liu c5ac42a198
[TOSA] Add aten.view shape -1 support (#1815) 2023-01-20 11:56:26 -08:00
Ramiro Leal-Cavazos d849cbad14
Make `getTypeForScalarType` safer by returning `FailureOr<Type>` (#1814)
One of the potential values for a `torch_upstream::ScalarType` is
`Undefined`. This means that conversion of a `ScalarType` to another
type is a computation that can fail. To enforce handling of the
failure case, this commit makes the two helper functions that convert
`ScalarType`s into other types return `failure()` when the
`ScalarType` is `Undefined`.
2023-01-20 18:40:13 +00:00
Roll PyTorch Action d3c6183294 update PyTorch version to 2.0.0.dev20230120 2023-01-20 14:55:41 +00:00
Chi_Liu 2587b3f583
[TOSA] Add aten.Index.Tensor support (#1771) 2023-01-19 21:19:00 -08:00
Roll PyTorch Action c4c9d66f71 update PyTorch version to 2.0.0.dev20230119 2023-01-19 14:42:35 +00:00
Vivek Khandelwal abf4f207cd [MLIR][TORCH] Add canonicalizer for aten.new_empty_strided op
Signed-Off By: Vivek Khandelwal<vivek@nod-labs.com>
2023-01-19 13:37:32 +05:30
Roll PyTorch Action 158c9b5408 update PyTorch version to 2.0.0.dev20230118 2023-01-18 14:50:48 +00:00
Vivek Khandelwal f9d59eb500 [MLIR][TORCH] Add decomposition for aten.randn_like op
This commit decomposes aten.randn_like op into aten.randn.generator op.

Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2023-01-18 12:09:27 +05:30
Vivek Khandelwal 999fd9036b [torchdynamo] Add native_group_norm and split op to the decomp list
Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2023-01-18 10:40:46 +05:30
Yuanqiang Liu 0a85033780
[MHLO] simplify aten.frobenius_norm.dim's lowering (#1800) 2023-01-17 13:52:12 -08:00