Commit Graph

561 Commits (532d297c46b13066da851af6eca53b91ff7a13c0)

Author SHA1 Message Date
Abhishek Varma 5337944ddb [MLIR][TORCH] Add e2e support for aten.randint
-- This commit adds e2e support for aten.randint by decomposing it into
   an aten.randint.low by setting low=0.

Signed-off-by: Abhishek Varma <abhishek@nod-labs.com>
2023-04-07 00:13:56 +05:30
Vivek Khandelwal 2213ce0855 [TorchDynamo] Add aten.squeeze op to the decomposition list
Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2023-04-06 22:21:25 +05:30
Vivek Khandelwal 5e9582b055 [MLIR][TORCH] Add e2e support aten.movedim.int op
Signed-Off By: Vivek Khandelwal<vivek@nod-labs.com>
2023-04-04 17:53:27 +05:30
Vivek Khandelwal 82fb9c7fb8 [MLIR][TORCH] Add decomposition for prims::squeeze op
This commit adds the decomposition for the prims.squeeze op.

Signed-Off By: Vivek Khandelwal<vivek@nod-labs.com>
2023-04-01 21:45:58 +05:30
Ramiro Leal-Cavazos e0f301c890
Add `extra_library` kwarg to `torch_mlir.compile` (#1986)
This commit adds the ability to specify extra abstract interpretation
functions in `torch_mlir.compile` to use during type refinement. This
allows users to easily add custom ops without having to interact with
MLIR or C++ directly.
2023-03-30 09:20:19 -07:00
Ramiro Leal-Cavazos 42d780dde0
Remove convolution_overrideable, convolution_backward_overrideable (#1984)
The ops `aten.convolution_overrideable` and
`aten.convolution_backward_overrideable` are currently not e2e tested
in Torch-MLIR. Moreover, there is no way to add e2e tests for them
because the ops cannot be called using the CPU backend (this also
prevents adding tested dtype functions for these ops). Since these two
ops are not expected to ever appear in PyTorch traces obtained through
standard means (https://github.com/pytorch/pytorch/issues/97481),
Torch-MLIR should not have to worry about them.
2023-03-29 15:05:56 -07:00
Maksim Levental 953ea39cb5
handles 2,3,4 from https://github.com/llvm/torch-mlir/issues/1963 (#1964) 2023-03-24 21:50:01 -05:00
Ramiro Leal-Cavazos a7449785ec
Use upstream shape functions when available (#1952)
There are several ops that have their shape function upstream and had
not been updated in Torch-MLIR to use the upstream version. This
commit updates those shape function. In addition, TODOs have been
added for shape functions that should be upstream but are not.
2023-03-24 09:13:43 -07:00
Ramiro Leal-Cavazos eae3ff7f1c
Change dtype functions interface to take ints tuple for each tensor (#1965)
The original design for the dtype functions outlined in
https://github.com/llvm/torch-mlir/issues/1462 was unable to properly
handle ops that take optional tensors as an input when the optional
tensor has a value of None. By the time the op gets imported into
torch-mlir, if an optional value is None, all information about the
original type is lost from the op type signature, preventing
torch-mlir from knowing if a value of None was from an optional tensor
or not, which was crucial in the original design since each tensor
argument must be turned into two separate arguments for the dtype
function.

This commit changes the interface to dtype functions such that each
tensor turns into a tuple of two ints, the first representing the rank
of the tensor and the second the dtype of the tensor. Since now there
is a one-to-one correspondence between the operands of an op and the
operands of its dtype function, there is no ambiguity about which
operand of the op corresponds with which operand of the dtype
function.

To test the implementation, this commit defines dtype function for
convolution op, which takes one optional tensor as an argument.
2023-03-23 11:05:39 -07:00
Jiahao Li 4912c3937d
Support aten.stack op and decompose it into unsqueeze & cat (#1747) 2023-03-11 09:25:25 +08:00
Eric Kunze 4c7e7ec116
Update LLVM tag to 21f4b84c (#1918)
Update PassManager C interface to use mlirPassManagerRunOnOp
Update python calls to PassManager to also use operation instead of
module
2023-03-06 22:53:26 -08:00
Zhekun Zhang 1d3a7419c5
[Torch Dialect] add RSub, ScalarImplicit canonicalize (#1899)
* add rsub, scalarimplit canonicalizer

* reformat

* address comments

* fix bug

* fix test

* Update elementwise.py

* resolve merge conflict

* change to 3

* change to 3

* real fix

* fix name

* add torchdynamo fail test

---------

Co-authored-by: zhekun.zhang <zhekun.zhang@bytedance.com>
2023-03-06 17:38:27 -08:00
Priya Savithiri c2ef5f4165
Add HardtanhBackward TOSA and LINALG support (#1721) 2023-03-06 10:16:37 -08:00
Yuanqiang Liu 7a8304f935
[Torch Dialect] add folder for aten.sub.float (#1871) 2023-03-02 09:07:33 -08:00
Yuanqiang Liu fc1e091d6a
[Torch Dialect] add aten.pow.int_float op and it's folder (#1872) 2023-02-28 09:36:05 -08:00
Vivek Khandelwal a32840ffd7 build: manually update PyTorch version
Set PyTorch and TorchVision version to nightly release 2023-02-27.
This commit also adds the lowering for aten.add and aten.Float.Scalar op.

Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2023-02-28 22:43:39 +05:30
Prateek Gupta 207229297e
[TORCH-MLIR] Add ODS for aten.clamp.Tensor op. (#1894)
This commit adds the ODS definition for the aten.clamp.Tensor op.

Signed-off-by: Prateek Gupta <prateek.gupta2@cerebras.net>
2023-02-24 09:18:24 -08:00
Vivek Khandelwal 6a3438f672 build: manually update PyTorch version
Set PyTorch and TorchVision version to nightly release 2023-02-20.

Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2023-02-23 11:33:41 +05:30
Zachary Cetinic e7111d473b
[Torch Dialect] Scatter reduce lowering (#1884)
- Lowers the torch.scatter_reduce to linalg_on_tensors dialect.
- Includes support for "sum", "prod", "amax", "amin" and "mean".
2023-02-21 23:05:55 +00:00
Yuanqiang Liu eb74014dd8
[Torch] decompose aten.norm.ScalarOpt_dim to aten.linalg_vector_norm (#1849) 2023-02-20 20:08:29 -08:00
Vivek Khandelwal b17d4d4f08
[MLIR][TORCH] Add decomposition for aten.bernoulli.p op (#1882)
Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2023-02-15 22:36:29 +05:30
Yuanqiang Liu 6ab990e1e8
[Torch Dialect] add folder for aten.Int.float (#1863) 2023-02-10 13:59:03 -08:00
Yuanqiang Liu 2f6fdb7f0b
[Torch Dialect] add folder for prim.min.int (#1864) 2023-02-10 13:58:15 -08:00
Zachary Cetinic 2a4a61f98f
Add aten.scatter_reduce op definition (#1846) 2023-02-07 21:59:07 +00:00
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
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
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
Gleb Kazantaev aa3a88c8d9
Fix JIT schema matching for when ListType is used (#1826) 2023-01-23 21:43:18 -05:00
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 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
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
Jiahao Li e2698433db
Fix empty tensor when select -1 (#1787) 2023-01-17 10:14:14 -08:00
Jiahao Li 4f94831fed
[LINALG][TOSA][MHLO] Add e2e support for aten bitwise ops (#1753) 2023-01-11 14:40:03 -08:00
Vivek Khandelwal fd236b2c89 [MLIR][TORCH] Add decomposition for prims.var and prims.sqrt op
Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2023-01-11 17:39:10 +05:30
Gleb Kazantaev c8b867b876
Added support for aten::norm.ScalarOpt_dim (#1774)
* Added support for aten::norm.ScalarOpt_dim

* Disable NormalizeModule_basic for linalg
2023-01-10 13:08:25 -05:00
Jiahao Li 8dc5d985eb
Add e2e support for aten logical or/and/xor/not ops (#1761) 2023-01-03 18:11:25 -08:00
Ramiro Leal-Cavazos 273664ded6
[custom op] Replace `tanh` dtype function with `expm1` (#1769)
This commit replaces the `tanh` dtype function, which was being used
to test the implementation of dtype functions in
a710237437, with a dtype function for
`expm1`. The dtype function for `expm1` is identical to the `tanh`
one, so the same level of testing is maintained.

Currently, there are ops getting dtype information from the
`RefineTypes` pass and ops getting dtype information from the
`TorchDtypeRefinementPipeline`. Since each pass can only propagete
dtype information for the ops it knows how to handle, some models with
many ops handled in both passes require the two dtype propagation
passes to execute many times, reaching the iteration limit set in the
`LowerToBackendContractPass`. To temporarily avoid this issue while
the migration to `TorchDtypeRefinementPipeline` is finished, this
commit switches `tanh` to `expm1`, since the latter is used a lot less
in large models.
2023-01-03 14:18:26 -08:00
Srirammaswamy a88e3766e8
Add E2E support for LeakyRelu and LeakyReluBackward ops (#1733)
Co-authored-by: srirammaswamy <srirammaswamy@gmail.com>
2023-01-03 08:30:16 -08:00
Ashay Rane ac780529b4
Revert e2e support for aten logical or/and/xor/not ops (#1757)
This reverts commit eaab9be207, since it
is causing the post-merge CI tests to fail, causing subsequent PRs to be
blocked.  Specifically, the tests
`ElementwiseAtenLogicalAndOpPromoteBroadcastModule_basic` and
`ElementwiseAtenLogicalXorOpPromoteBroadcastModule_basic` fail because
the oracle does not match the computed result.  This patch reverts the
commit to make the post-merge builds green again.
2022-12-29 21:01:06 -06:00
Shivam Gupta 2f45959f0d
Prelu lowering to linalg (#1712)
Prelu lowering to linalg
2022-12-28 08:51:33 +05:30
Jiahao Li eaab9be207
Add e2e support for aten logical or/and/xor/not ops (#1752) 2022-12-26 10:23:38 +08:00
Ramiro Leal-Cavazos 3260a1ea6e
Allow passing traced `torch.nn.Module`s into `torch_mlir.compile` (#1743)
This commit adds support for passing to `torch_mlir.compile` the
result of running `torch.jit.trace` on a model by relaxing the
condition that checks if the model is already in JIT IR to allow any
`torch.jit.ScriptModule`.

Fixes https://github.com/llvm/torch-mlir/issues/1739
2022-12-22 08:39:55 -08:00
Jiahao Li 60a139271d
Add aten.std.correction op and its decomposition (#1731) 2022-12-21 21:02:40 -08:00
Jiahao Li 15b249777b
[Torch][MHLO] Decompose aten.copy op. Lower aten.rsqrt & sigmoid to mhlo. (#1734) 2022-12-22 10:13:59 +08:00
Jae Hoon (Antonio) Kim 1d695239ff
Unrevert #1724 (#1737)
* Unrevert #1724

* Update pytorch requirements.txt
2022-12-20 11:17:21 -05:00
Ashay Rane dd1cf578a6
build: fix LTC code after upstream PyTorch change (#1727)
pytorch/pytorch@140a3139 reverted a change from yesterday, causing the
RollPyTorch action to break.  This patch reverts the corresponding
change in the torch-mlir LTC code.

This patch also re-enables tests that were previously marked as XFAIL.
2022-12-16 13:07:38 -06:00
ataheridezfouli-groq 17ee643aeb
[TORCH] Add Complex Number support (#1673)
Add Complex number dtype support to torch tensors. Add
aten.fft_fft op to test complex numbers.
2022-12-15 21:40:01 +00:00
Jae Hoon (Antonio) Kim a2a93891ea
Replace asIntArrayRefSlow with macro (#1724)
* Replace asIntArrayRefSlow with macro

* Update pytorch requirements.txt
2022-12-15 11:52:41 -05:00
Sean Silva af9e8a5e63 [torchdynamo] Move to aot_autograd instead of raw make_fx
As [@ezyang suggested](https://github.com/pytorch/pytorch/issues/90276#issuecomment-1339791275),
use `torch._dynamo.optimizations.training.aot_autograd` instead of raw
`make_fx`. This is more future proof and gives us the backward pass and
functionalization. We don't currently get functionalization because of
https://github.com/pytorch/pytorch/issues/90759

This also incidentally fixes the source location handling, which makes
`lockstep_basic.py` give an accurate source location!
2022-12-15 01:55:50 -08:00
Ahmed S. Taei b1f6832849
Add aten.slice.Tensor & aten.cat folders (#1691) 2022-12-13 13:02:47 -08:00
Ramiro Leal-Cavazos a710237437
[custom op] Generalize shape library logic to work with dtypes (#1594)
* [custom op] Generalize shape library logic to work with dtypes

This commit generalizes the shape library logic, so that dtype rules
for ops can also be expressed using the same mechanism. In other
words, each op can now have a shape function and a dtype function
specified in Python that is imported during lowering to calculate the
shapes and dtypes throught a program. For more information about how
to specify a dtype function, see the updated
`docs/adding_a_shape_and_dtype_function.md`.

For those not familiar with how the shape library works, the file
`docs/calculations_lib.md` provides an overview.
2022-12-13 08:25:41 -08:00
Ashay Rane 430737b820
[cleanup] fix naming of private variable according to the style guide (#1704) 2022-12-12 09:04:46 -06:00
Vivek Khandelwal d4862ec611 [MLIR][TORCH] Add e2e support for aten.var_mean op
Signed-Off By: Vivek Khandelwal<vivek@nod-labs.com>
2022-12-12 15:46:54 +05:30
Sean Silva 7731211d02 Remove eager_mode
This was an experimental attempt at rolling out own op-by-op executor
with `__torch_dispatch__`, but it proved difficult to make it robust.
Op-by-op execution is very easy to implement robustly now with the
PyTorch 2.0 stack, so we don't need eager_mode.

Downstream users were using eager_mode to implement lockstep numerical
accuracy debuggers. We implemented the same functionality with
TorchDynamo in https://github.com/llvm/torch-mlir/pull/1681 so now there
is not much reason to continue maintaining it.
2022-12-09 03:50:00 -08:00
Gleb Kazantaev 804f9f1f8f
Extended TorchMLIRLoweringContext with virtual CreateComputation method (#1699)
* Extended TorchMLIRLoweringContext with virtual CreateComputation method

* Fix device_data_cast return value
2022-12-08 15:57:07 -05:00
Vivek Khandelwal 3e4bb2bd8e [MLIR][TORCH] Add E2E support for randn and randn.generator op
Signed-Off By: Vivek Khandelwal<vivek@nod-labs.com>
2022-12-06 22:41:24 +05:30
Vivek Khandelwal ef39b9ebb4 build: manually update PyTorch version
Set PyTorch and TorchVision version to nightly release 2022-12-05.

Signed-Off By: Vivek Khandelwal<vivek@nod-labs.com>
2022-12-05 22:44:32 +05:30
Sean Silva 88db99946b [torchdynamo] Use decompositions to support a few ops 2022-12-01 11:25:20 -08:00
Abhishek Varma c27c1791f1 [MLIR][TORCH] Add e2e support for `aten.amax` op
-- This commit adds e2e support for `atend.amax` op.

Signed-off-by: Abhishek Varma <abhishek@nod-labs.com>
2022-11-30 17:54:37 +05:30
Tanyo Kwok bbcdb38d99
Revert "Decompose torch.slice_scatter (#1622)" (#1659)
This reverts commit f3f2f10030.
2022-11-30 12:47:13 +08:00
Daniel Ellis e2de20575f
Automatically strip overloads for FX-based models. 2022-11-29 22:19:09 -05:00
Vivek Khandelwal d9cbf01d1e Revert "build: update llvm tag to 147fe9de"
This reverts commit e45ad313d4.
2022-11-25 12:41:56 +05:30
Sean Silva 28957adaac [torchdynamo] Initial TorchDynamo support
This adds a basic e2e Config for TorchDynamo using
Linalg-on-Tensors/RefBackend.
But TorchDynamo is pretty orthogonal to
various other pieces, so it should compose nicely with variations like:
- Switching out all the backends (Linalg-on-Tensors, TOSA, MHLO)
- PyTorch functionalization and decompositions
- Taking the example inputs and compiling with all dynamic or all static
  shapes without duplicating tests.

This adds it to the CI, but there are still a lot of XFAIL's.

This also adds a helper `from torch_mlir.dynamo import
make_simple_dynamo_backend` which simplifies some of the steps for
making a Torch-MLIR-based TorchDynamo backend. We include "simple" in
the name because we are going to be exploring various things next from
the long-term roadmap.

The next steps are:
- Burn down all the XFAIL's.
- Start working on the pieces from the [long-term roadmap](https://github.com/llvm/torch-mlir/blob/main/docs/long_term_roadmap.md).
  - Add functionalization/decompositions into the TorchDynamo flow and
    remove reliance on the current Torch-MLIR "frontend".
  - Write a pure-Python direct FX->MLIR importer.
  - Hook up the new PyTorch symbolic shape stuff.
  - Explore PrimTorch decompositions for simplifying backends.
2022-11-24 04:10:25 -08:00
Vivek Khandelwal e45ad313d4 build: update llvm tag to 147fe9de
Summary of changes:
- Update call to `hasNoEffect` utility
- `KDynamicSize` value changed to
  `std::numeric_limits<int64_t>::min()` from `-1`
- Update tags
  llvm: 147fe9de29dc13c14835127b35280c4d95c8e8ba
  mhlo: 1944b5fa6062ec4c065d726c9c5d64f1487ee8c5

Signed-Off By: Vivek Khandelwal<vivek@nod-labs.com>
2022-11-24 12:44:43 +05:30
Tanyo Kwok 14f1260ac4
Add more mhlo basic converters (#1628)
* Add more mhlo basic converters

* remove unused pinnedMemory constraints

* refine naming
2022-11-24 14:28:34 +08:00
Maksim Levental bfcfd60d55
[MLIR][TORCH] Refix differentiable view (#1639)
* `BatchMlpLayerModule_basic` passes

* Fix https://github.com/llvm/torch-mlir/issues/1618 by stripping `requires_grad` from results of view ops.
2022-11-23 15:35:39 -06:00
Tanyo Kwok f3f2f10030
Decompose torch.slice_scatter (#1622)
* Decompose torch.slice_scatter

* fix compilation error

* update file check

* fix ci

* fix i64 torch.tensor dtype
2022-11-23 18:14:12 +08:00
Vivek Khandelwal 68f568b704 [MLIR][TORCH] Add E2E support for prims.convert_element_type op
Signed-Off By: Vivek Khandelwal<vivek@nod-labs.com>
2022-11-22 09:36:36 +05:30
Maksim Levental ed901094c1
Fix https://github.com/llvm/torch-mlir/issues/1618 by stripping `requires_grad` from results of view ops. (#1624) 2022-11-21 19:15:53 -06:00
Vivek Khandelwal 4cbd3927d7 [MLIR][TORCH] Add aten.sort.int op
Signed-Off By: Vivek Khandelwal<vivek@nod-labs.com>
2022-11-20 19:00:41 +05:30
Abhishek Varma 1d949f3ac2 [MLIR][TORCH] Fix aten.upsample_nearest2d op
-- aten.upsample_nearest2d.vec op is not present
   owing to https://github.com/pytorch/pytorch/pull/85638
-- So this commit adds a lowering on aten.upsample_nearest2d.

Signed-off-by: Abhishek Varma <abhishek@nod-labs.com>
2022-11-18 13:41:47 +05:30
Vivek Khandelwal 5f7177da35 [MLIR][TORCH] Add decomposition for aten.var_mean.correction op
Signed-Off By: Vivek Khandelwal<vivek@nod-labs.com>
2022-11-17 13:00:09 +05:30
Sean Silva 3695ca83e6 [torch_mlir.compile] Handle the case of already-scripted models better
Closes #1582
2022-11-16 10:47:13 -08:00
Vivek Khandelwal a1d3afdba9 [MLIR][TORCH] Add E2E support for aten.randint.low op
Signed-Off By: Vivek Khandelwal<vivek@nod-labs.com>
2022-11-16 09:54:18 +05:30
George Petterson 92f385bd9f [MLIR][TORCH] Add E2E support aten.convolution_backward op
This commit adds the decomposition for the `aten.convolution_backward`
and `aten.convolution_backward_overrideable` op.
2022-11-15 07:38:26 +05:30
Gleb Kazantaev 6909eaf7fc
Update TorchMlirBackendImpl Methods (#1580)
* Fix LTC build

* Remove passing test from xfail set
2022-11-14 00:37:49 -05:00
Vivek Khandelwal a558034c1a [MLIR][TORCH] Fix aten.upsample_nearest2d_backward op
Signed-Off By: Vivek Khandelwal<vivek@nod-labs.com>
2022-11-12 00:05:36 +05:30
Vivek Khandelwal d571d050fd [torch_mlir.compile] Fixes issue with the https://github.com/llvm/torch-mlir/issues/1557
Signed-Off By: Vivek Khandelwal<vivek@nod-labs.com>
2022-11-11 18:05:15 +05:30
Sean Silva cc468d2d16 [cleanup] Be consistent about apostrophe 2022-11-10 07:42:15 -08:00
Xiafei Qiu 4f173c6e0f
update llvm tag to a2620e00. (#1567)
- also update MHLO to 57ba12a2(branch greencommit/2022-11-07-a2620e00)
- change -pass-pipeline format to make tests pass.
2022-11-10 18:39:28 +08:00
Sean Silva 64914603fa [torch_mlir.compile] Add support for multiple exported methods
For AoT deployments models often have multiple exported methods.
This patch enables something like this:

```
class TwoMethodsModule(torch.nn.Module):
    def sin(self, x):
        return torch.ops.aten.sin(x)

    def cos(self, x):
        return torch.ops.aten.cos(x)

example_args = torch_mlir.ExampleArgs()
example_args.add_method("sin", torch.ones(2, 3))
example_args.add_method("cos", torch.ones(2, 4))
print(torch_mlir.compile(TwoMethodsModule(), example_args))
```

In the
[long-term](https://github.com/llvm/torch-mlir/blob/main/docs/long_term_roadmap.md#tools-for-advanced-aot-deployments)
we will need to reconcile this with our story for stateful models and the
backend contract being purely functional. For now, this provides some basic
infra that seems harmless. Arguably, we could tighten up the backend contract
even more to only allow a single compiled function which would prohibit this or
require building out a layer above.

Fixes #1557
2022-11-10 02:10:22 -08:00
Jae Hoon (Antonio) Kim 2ec4b06bbb
Remove MakeView from IR Builder (#1552)
* Remove MakeView from IR Builder

* Update PyTorch requirements
2022-11-09 13:46:34 -05:00
Ashay Rane d99b2ddb1b
importer: fix usage after PyTorch update (#1555)
Unless requested otherwise, PyTorch no longer installs most of the
header files under the caffe2 directory (see
https://github.com/pytorch/pytorch/pull/87986).  This breaks our
importer code since we need to use the `MakeGuard()` function to execute
statements in the event of exceptions.

To fix this issue, this patch implements a rudimentary version of
PyTorch's ScopeGuard, where once the class variable goes out of scope,
it executes a predefined method.
2022-11-04 15:02:23 -05:00
Vivek Khandelwal fedf8c0640 [MLIR][TORCH] Add E2E support for aten.upsample_nearest2d_backward.vec op
Signed-Off By: Vivek Khandelwal<vivek@nod-labs.com>
2022-11-04 22:10:07 +05:30
Jae Hoon (Antonio) Kim 0701464c47
Remove view ops from IR builder (#1534)
* Remove view ops from IR builder

* Update PyTorch requirements
2022-10-30 21:42:44 -04:00
Vivek Khandelwal c86177730d [MLIR][TORCH] Add E2E support for aten.fill.Tensor op
This commit adds the decomposition for `aten.fill.Tensor` op.

Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2022-10-30 18:40:47 +05:30
Ramiro Leal-Cavazos b723186983
Remove all but one of valsem ops + move fill.Scalar to elementwise (#1531)
This commit removes almost all of the valsem ops, since the value
semantics version of the ops now exist in PyTorch. The only op missing
is `aten.bernoulli_.float`. In addition, this commit also simplifies
the implementation of `aten.fill.Scalar` by moving it to the pattern
that converts elementwise ops.
2022-10-28 15:06:11 +00:00
Vivek Khandelwal ea602127b6 [MLIR][TORCH] Add E2E support for aten.addcmul_ and aten.addcdiv_ op
Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2022-10-28 16:07:50 +05:30
Vivek Khandelwal ca87033d2f [MLIR][TORCH] Add E2E support for aten.mse_loss op
This commit adds decomposition for the `aten.mse_loss` op.

Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2022-10-25 21:06:58 +05:30
Jae Hoon (Antonio) Kim 2f300935bf
Reference lazy graph executor (#1507)
* Add LazyGraphExecutor registration

* Update PyTorch version to 1.14.0.dev20221024

Co-authored-by: Roll PyTorch Action <torch-mlir@users.noreply.github.com>
2022-10-24 17:15:11 -04:00
Ramiro Leal-Cavazos 82a3860e25
build: update llvm tag to 4546397e (#1502)
This commit makes the following changes needed to update bump LLVM:

- Replace `linalg.init_tensor` with `tensor.empty` (see:
https://reviews.llvm.org/D135129)
- Replace `NoSideEffect` with `Pure` (see
https://reviews.llvm.org/D135505)
- Replace `body` region accessor for `ReduceOp` and `ReduceWindowOp`
with `getBody`
- Fix incorrect use of `tosa::ReduceSumOp` in `AtenNativeLayerNormOp`
conversion pattern. The result type of `tosa::ReduceSumOp` must have
the same rank as the input type. (see:
https://www.mlplatform.org/tosa/tosa_spec.html#_reduce_sum)

Co-authored-by: Ashay Rane <ashay@users.noreply.github.com>

Co-authored-by: Ashay Rane <ashay@users.noreply.github.com>
2022-10-18 04:22:53 +00:00
Gleb Kazantaev bdb5083d33
New ops support & enhancements (#1494)
* New ops support & enhancements

* Enabled xfail ltc tests
2022-10-14 10:28:21 -04:00
Prashant Kumar 3a2cd23380 [LINALG] Add lowering for aten::round op.
-- Added the lowering for aten::round op.
-- Added the folding for integer cases.
2022-10-13 02:41:26 +05:30
Sean Silva 6403c0e56f torch_mlir.compile: allow custom backend_legal_ops set
Allow customizing `backend_legal_ops` for "torch" output type, since we
don't know which backend will be used (it might be a custom backend).
We don't allow customizing the `backend_legal_ops` for the other output
types (Linalg, TOSA, MHLO) since those backends control their set of
legal ops directly.

Fixes #1418
2022-10-12 04:21:22 -07:00
Abhishek Varma 61db1b5c4d
[MLIR][TORCH] Add e2e support for `aten.Mish` op (#1470)
-- This commit adds e2e support for `aten.Mish` op.
-- `aten.Mish` op is decomposed as following :-
    Mish(x) = x * Tanh(Softplus(x))

Signed-off-by: Abhishek Varma <avarma094@gmail.com>

Signed-off-by: Abhishek Varma <avarma094@gmail.com>
2022-10-11 14:03:10 -07:00
Jae Hoon (Antonio) Kim 3e08f5a779
Fix `fromIntArrayRef` call (#1479)
* Fix fromSymint call

* Update PyTorch requirement

* Re-enable LTC
2022-10-11 13:29:07 -04:00
Gaurav Shukla da90a25f90 [MLIR][TORCH] Add E2E support for `aten.[div.int|bitwise_or.Tensor]` ops
This commit adds lowering of `aten.div.int` and `aten.bitwise_or.Tensor`
ops. Both these ops are required in order to support bloom_560m model.

Signed-Off-by: Gaurav Shukla <gaurav@nod-labs.com>
2022-10-10 22:28:51 +05:30
Daniel Ellis e7b2b84a66 Update torch-mlir-opt error message. 2022-10-05 15:02:10 -04:00
Jae Hoon (Antonio) Kim c57d801260
Fix functionalize_aten_op calls for symint ops (#1459)
* Fix functionalize_aten_op calls for symint ops

* Update PyTorch version
2022-10-05 10:23:48 -04:00
Gleb Kazantaev 708fa346a6
Fix Base Lazy Backend Type Conversion (#1412)
* Fix c10::prim::Constant conversion; Added CAPI for passes; Added passes to base lazy backend

* Update ivalue_importer to use ImportOptions; Added tests for non-value/value tensor types

* Added tests for scalar Constant import; Updated MB::importFunction to use ImportOptions

* Test updates

* Move back module variable name

* Remove RefineTypes from TorchMlirLoweringContext::Build()

* Rename pass; Remove passes from base lazy backend

* Rename pass to VerifyBackendContractPass

* Aligned cmd pass name; Fixed TorchConversion passes registration
2022-10-04 15:53:28 -07:00
Daniel Ellis 2ba71af651 Add support for mv decomposition. 2022-10-04 11:34:45 -04:00
Prashant Kumar 6777a9484d [LINALG] Add lowering for the aten.upsample_nearest2d op. 2022-10-04 17:20:29 +05:30
Daniel Ellis 4d47f1671a Reject dictionary inputs when tracing.
The underlying error message was misleading.  See https://github.com/llvm/torch-mlir/issues/1425
2022-09-30 16:02:35 -04:00
AmosLewis 940959589b [MLIR][TORCH] Add Byte and Char Dtype support 2022-09-30 13:19:31 +05:30
Ashay Rane 0b46462528
Miscellaneous fixes for Windows builds (#1376)
* test: allow spaces in path to Python executable

On Windows, the path to the Python binary may contain spaces, so this
patch adds quotes around the path to the python executable.

Thanks to @sstamenova for suggesting the fix!

* python: remove header file that causes Windows build failures

Similar to https://reviews.llvm.org/D125284, we can safely remove this
header file without affecting the build on either Linux.  It is
necessary to remove this header file on Windows builds since otherwise
it causes build errors.

* python: drop `TORCH_API` from function defined in Torch-MLIR

`TORCH_API` should apply to functions that are either exported by
libtorch.so or ones that are imported from libtorch.so by its downstream
consumers (like Torch-MLIR).  Neither case applies to the
`importJitFunctionAsFuncOp()` function, since it is defined in
Torch-MLIR (and thus outside libtorch.so).  This patch fixes the problem
by dropping `TORCH_API` from that function's declaration.

* python: make output of class anotations deterministic

The `class-annotator-repr.py` test checks for class annotations in a
specific order, but prior to this patch, the order was
non-deterministic, since the code iterated on an _unordered_ map.

This patch makes the iteration order deterministic through two changes:
1. using a sorted map
2. using the class qualified name instead of the address of the class in
memory

* test: use Python3_EXECUTABLE as interpreter path for consistency

This ensures that tests use the Python3 version that was detected using
CMake, instead of whichever python version that happens to be in the
PATH variable when invoking the test.

* test: fix RUN string

The parenthesis syntax does not run on Windows (the shell interprets the
`(` character as part of the path).  Moreover, the ODR violation in the
comment no longer seems to apply.

* python: port parallel test framework to Windows

Since Windows does not support `fork` natively, Python's
`multiprocessing` module needs to use `spawn` on Windows.  However, to
use `spawn`, the multiprocessing module serializes (or pickles) the
worker function and its arguments.  Sadly, the multiprocessing module
(both the default one in Python and the one that is extended in PyTorch)
is unable to serialize lambda functions (see
https://stackoverflow.com/a/19985580) for detals.

Unfortunately, given how our tests are structured, we require that the
function under test is passed as an argument to another function, so we
cannot sidestep our use of lambda functions.

To resolve this problem, this patch makes use of the `multiprocess` and
`dill` Python modules, which together offers a multiprocessing mechanism
that can serialize lambda functions.  The multiprocess module also
offers a process pool, which simplifies the code for our parallel
testing framework.
2022-09-29 12:07:43 -05:00
Jae Hoon (Antonio) Kim fa5a8e21a3
Propagate parameter names to TorchMlirComputation (#1420)
* Propagate parameter name to MLIR

* Add TorchMlirNode Constructor Hook

* Make func_op mutable

- Purpose of this is to allow modification of func_op by subclass
  backend

* Clean up unnecessary changes

* Remove unnecessary attribute case

* Address PR comments
2022-09-29 11:43:39 -04:00
武家伟 c03aa63325
[MLIR] Add canonicalizer for aten.slice.t op (#1413)
* [MLIR] Add canonicalizer for aten.slice.t op

* Add mlir tests and strength the canonicalizer

* rename variable

Co-authored-by: Vremold <xremold@gamil.com>
2022-09-26 14:35:50 -07:00
Jae Hoon (Antonio) Kim 3e27aa2be3
Fix as_strided/slice symint (#1401)
* Fix as_strided symint

* Re-enable LTC tests

* Re-enable LTC

* Add hardtanh shape inference function

* Fix slice symint
2022-09-26 12:16:49 -04:00
Tanyo Kwok 72e422b589
Add relu6 and binary broadcasts (#1408)
* Add relu6 and binary broadcasts
2022-09-23 20:39:15 +08:00
武家伟 0e2e94d542
Add torch-to-mhlo e2e support for AtenArangeStartStepOp (#1385)
Co-authored-by: Vremold <xremold@gamil.com>
2022-09-20 22:31:24 +08:00
Jae Hoon (Antonio) Kim 8967463980
Fix symint ops and blacklist `lift_fresh_copy` (#1373)
* Add symint to native functions yaml

* Re-enable LTC

* Fix new_empty_strided and narrow_copy
2022-09-20 10:16:04 -04:00
武家伟 4f3cd236dd
Strength the shape inference for aten.arange-like op (#1367)
Strength the shape inference for aten.arange-like op by
1. registering aten.sub and aten.ceil.Scalar op and design folders for them.
2. register a new constant-like op: Torch::ConstantNumberOp and design canonicalizer for it.
2022-09-20 12:40:19 +08:00
Ashay Rane 1895b581c4
shape-lib: generate string as multiple lines to work with MSVC (#1370)
As @oroppas identified, literal strings that are over 16,380 characters
cause the MSVC compiler to throw an error (C2026), eventually causing
the Windows build of Torch-MLIR to fail because the length of the
generated MLIR for the shape library crosses the allowed threshold.

This patch fixes the problem by making the Python script generate one
literal string per line to satisfy the MSVC compiler.

Thanks to @oroppas for the bulk of the effort required to resolve this!
2022-09-16 15:16:01 -05:00
gpetters94 48418b9c22
Fold away type_as (#1358) 2022-09-12 18:59:12 -04:00
Vivek Khandelwal 71b1f0dd7a [MLIR][TORCH] Add E2E support for aten.index.Tensor_hacked_twin op
This commit adds lowering of `index.Tensor_hacked_twin` op.

Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2022-09-12 21:47:18 +05:30
George Petterson a12b9c4492 Add lowering for aten::cumsum 2022-09-12 09:28:07 +05:30
Vivek Khandelwal 326f21229e [MLIR][TORCH] Fix shape calculation for aten::pow.Tensor_Tensor op
Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2022-09-08 21:14:12 +05:30
Vivek Khandelwal e35741fb1d [MLIR][TORCH] Add E2E support for aten.bitwise_not op
This commit adds lowering of `aten.bitwise_not` op.

Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2022-09-08 17:52:12 +05:30
Vivek Khandelwal 7dfadc2498 [MLIR][TORCH] Add E2E support for aten.lift_fresh_copy op
This commit adds lowering of `aten.lift_fresh_copy` op.

Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2022-09-08 12:32:16 +05:30
Vivek Khandelwal c19fccfca2 [MLIR][TORCH] Add E2E support for aten.pow.Tensor_Tensor op
This commit adds lowering of `aten.pow.Tensor_Tensor` op.

Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com>
2022-09-08 10:01:42 +05:30
武家伟 6a1893a517
[MLIR][MHLO] Add AtenFrobeniusNormDimOp and add its conversion pattern to MHLO and linalg (#1306)
* Add aten.frobenius_norm.dim op and init its conversion pattern to linalg and MHLO, 
* run symbolic-shape-optimization before hlo-legalize-to-linalg to fit more mhlo e2e tests.
2022-09-08 10:15:36 +08:00
Gaurav Shukla 99093d0623 [TORCH] Add decomposition of `aten.linear` op
This commit adds decomposition of `aten.linear` op. Due to limited
support at tosa backend in case of dynamic dimensions, this
decomposition is currently disabled for tosa backend.

Signed-Off-by: Gaurav Shukla <gaurav@nod-labs.com>
2022-09-07 16:58:27 +05:30
Tanyo Kwok 512f2d9c23
Add decomposition to aten.native_layer_norm (#1332)
* Add decomposition to aten.native_layer_norm

* fix ci error
2022-09-02 09:29:22 +08:00
Sean Silva e16b43e20b Remove "torchscript" association from the e2e framework.
We use it for more than TorchScript testing now. This is a purely
mechanical change to adjust some file paths to remove "torchscript".

The most perceptible change here is that now e2e tests are run with

```
./tools/e2e_test.sh
instead of:
./tools/torchscript_e2e_test.sh
```
2022-08-29 14:10:03 -07:00
Jae Hoon (Antonio) Kim 8e880a2d00
Fix symint related functionalization ops (#1289)
* Fix symint related functionalization ops

* Remove zeros xfail from LTC tests
2022-08-26 16:13:28 -04:00
Henry Tu e869e68559
Fix LTC lib_torch_mlir_ltc.so import error (#1283)
* Build LTC to _mlir_libs directory

* Update CMakeLists.txt
2022-08-25 18:25:01 -04:00
Henry Tu a1ace0657d
Revert updating mlir_native_functions.cpp signature (#1281)
* Revert updating mlir_native_functions.cpp signature, due to a7edf71360

* Restored NewZeros to LTC XFAIL set
2022-08-25 13:00:33 -04:00
Henry Tu e2f862cb85
Fix LTC build warnings (#1272)
* Resolved Wunused-variable

* Fix Wunneeded-internal-declaration

* Address review comment

* Update autogen_ltc_backend.py

* Update mlir_native_functions.cpp to work with updated PyTorch

* Remove NewZeros from LTC XFAIL set
2022-08-24 15:04:28 -04:00
gpetters94 f012279fa2
Add transposed case for at::convolution (#917)
Also adds a decomposition for aten::conv_transposed2d.input
2022-08-24 12:19:35 -04:00
Sean Silva d7d67979b2 [cleanup] Change OutputType enum values to strings
The use of numbers was arbitrary and was preventing the enum values from
being put in the natural order.
2022-08-23 17:59:39 -07:00
Tanyo Kwok 3d0e18bbe7
Add decomposition for aten.roll (#1170)
* Add decomposition for aten.roll

* add e2e unittest

* refine type of torch.roll

* fix aten::cat output type
2022-08-24 08:36:05 +08:00
Tanyo Kwok 2374098d71
[MHLO] Init end to end unit tests (#1223) 2022-08-23 16:47:21 +08:00
Tanyo Kwok 9176b5ed29
Add decomposition for aten.flatten.using_ints (#1161) 2022-08-23 11:52:54 +08:00
Sean Silva 01290d134a Add a way for backends to control which ops are legal for them.
We were already hitting many cases where backends different in terms of
the legal ops that they wanted. This caused unnecessary coupling between
the backends. Examples:
- https://github.com/llvm/torch-mlir/pull/1161
- https://github.com/llvm/torch-mlir/pull/862

This PR centralizes all compilation to go through `torch_mlir.compile`
so that we can keep the logic centralized there. We should move these
lists closer to each backend. Especially cases like
https://github.com/llvm/torch-mlir/pull/862 where blocking a
decomposition is necessary to avoid a crash emphasize that the set of
decompositions is tightly coupled to the backend, and should be
"controlled by the backend" and not something arbitrarily tweakable.

Also:
- Fix a small bug in the way we passed through the backendLegalOps
  option.
- Add better error messages in `torch_mlir.compile` for import errors.
2022-08-22 14:16:13 -07:00
Alex Tsao c38308f3ef
Add lowering for _convolution.deprecated (#1259)
* Add lowering for _convolution.deprecated
2022-08-22 11:17:36 +08:00
Henry Tu ba17a4d6c0
Reenable LTC in out-of-tree build (for real this time) (#1205)
* Fix OOT LTC CI build failure

* Disable LTC during macOS package gen

* Add more details about static TorchMLIRJITIRImporter library
2022-08-19 15:25:00 -04:00
Ramiro Leal-Cavazos f07f7d20f9
Clean up shape functions that use `sum_mean_dim` (#1217)
I recently fixed the handling of the `dim` argument in
`sum_mean_dim` (59fccab857). Therefore,
the checks that the `dim` input is `None` or `[]` are no longer needed.
2022-08-18 08:23:43 -07:00
Quinn Dawkins 85f383ce0b
Bump the shape lib to match the upstream functions currently in PyTorch (#1236)
Bumps the shape library:
 - Updates the function signature for aten.arange.start_step
 - upstream_shape_functions.mean_dim -> upstream_shape_functions.sum_mean_dim
2022-08-17 00:11:04 -04:00
nithinsubbiah fde390c766 Re-enable custom op support 2022-08-16 22:49:08 +05:30
Jae Hoon (Antonio) Kim 0af55781ae
Propagate device data names (#1157)
* Propagate device data names

* Address PR comment

* Add example usage

* Add test for device data names

* Make TorchMlirComputation fields protected

* Add lazy backend device data name unit tests

* Disable lazy backend tests if LTC is disabled

* Add comments
2022-08-16 09:30:22 -04:00
武家伟 3b3cb99ef8
Generalize canonicalization pattern for more aten.sub/div/mul/add op (#1209)
Generalize canonicalization pattern for more sub/div/mul/add op, but for AtenDivTensorModeOp in 'trunc' rounding mode, we try to fold it.
2022-08-16 13:24:08 +08:00
Sambhav Jain 41aa562fb4
s/external/externals/g (#1222)
Fix remaining instances of `external/llvm-project`.
2022-08-13 07:13:56 -07:00
Prashant Kumar b1a506624c Add decomposition of `aten.masked.tensor` op.
`aten.masked.tensor` op has been decomposed to `aten.masked.scalar` op.
2022-08-11 07:48:04 +05:30
Vidush Singhal dd2da5a038
E2E support for AtenRemainderScalarOp (#1200) 2022-08-10 20:02:06 -04:00
gpetters94 79b9cf9468
Add lowering for aten.to.device (#1107) 2022-08-10 19:24:02 -04:00
powderluv e55fc4deb5
Revert "E2E support for AtenRemainderScalarOp (#1119)" (#1190)
This reverts commit 34e207eeb5.
2022-08-08 22:59:57 -07:00
Henry Tu 3e97a33c80
Revert "Reenable LTC in out-of-tree build (#1177)" (#1183)
This reverts commit f85ae9c685.
2022-08-08 18:58:35 -07:00
Vidush Singhal 34e207eeb5
E2E support for AtenRemainderScalarOp (#1119)
* E2E support for AtenRemainderScalarOp
2022-08-08 20:02:52 -04:00
Vidush Singhal b70548edff
Add decomposition and E2E support for Aten_EmbeddingBag (#1137)
* Add decomposition and E2E support for Aten_EmbeddingBag
2022-08-08 18:56:49 -04:00
Henry Tu f85ae9c685
Reenable LTC in out-of-tree build (#1177) 2022-08-08 17:35:22 -04:00
Tanyo Kwok 290d7755fb
importer: add initial support for loading Float16 tensors (#1169)
follow up #761:

    This patch updates the `torch_mlir::convertTensorToMlirElementsAttr()`
    method to enable the creation of tensors whose base type is Float16.
    This patch also adds a test to validate the IR generation, and it
    updates the test for importing tensors of various types.
2022-08-08 12:37:31 +08:00
Henry Tu e322f6a878
Update LTC CMake hack documentation (#1155)
* Update CMakeLists.txt

* Update CMakeLists.txt

* Update CMakeLists.txt

* Update CMakeLists.txt

* Update buildAndTest.yml

* Update setup.py

* Address review comments
2022-08-05 14:12:20 -04:00
Vivek Khandelwal c129a6de93 [MLIR][TORCH] Add support for dim=None to Aten[Var|Std]DimOp
PyTorch recently added support for `dim=None` in the `torch.var`
(5ca9b2b6fa)
and `torch.std`op (eb0e30e0bc).
This commit adds the corresponding support in torch-mlir.

Signed-Off By: Vivek Khandelwal<vivek@nod-labs.com>
2022-08-05 20:28:56 +05:30
Sean Silva 31727f81d8 torch_mlir.compile: Allow ignoring traced shapes
In some cases, users know that a traced graph is valid for a wider set
of shapes than they originally traced it with. Provide an option for
users to ignore the shapes in the traced graph when they know it is
legal.

Fixes #997
2022-08-04 10:18:34 -07:00
Ramiro Leal-Cavazos a7af1fd873
Add support for `dim=None` to `AtenMeanDimOp` (#1129)
PyTorch recently added support for `dim=None` in the `torch.mean`
op (2bfae07a79). This
commit adds the corresponding support in torch-mlir.
2022-08-02 16:08:06 +00:00
Quinn Dawkins 38d8498b21
add e2e support for aten.atan2 (#1117)
- Includes math-to-libm pass in refbackend for math::atan2 support
2022-08-02 11:39:41 -04:00
Vidush Singhal ed13ebfd8d
E2E support for AtenEmbeddingBagPaddingIdxOp SUM Mode (#1066) 2022-08-01 16:44:11 -04:00
Alec 554570f3ab Implemented a decomposition of aten::narrow 2022-08-01 18:32:14 +05:30
Henry Tu 2c3b3606d0 Resolve remaining LTC CI failures (#1110)
* Replace CHECK_EQ with TORCH_CHECK_EQ

* Check value of TORCH_MLIR_USE_INSTALLED_PYTORCH during LTC build

* Update LTC XFAIL with NewZerosModule ops

* Explicitly blacklist _like ops

* Automatically blacklist new_/_like ops

* Prune away unused Python dependencies from LTC

* Add flag to disable LTC

* Autogen dummy _REFERENCE_LAZY_BACKEND library when LTC is disabled

* Implement compute_shape_var

* Removed Var tests from XFAIL Set

* XFAIL tests using _local_scalar_dense or index.Tensor

* Add StdDim tests to XFAIL set

* Autogen aten::cat
2022-07-30 09:40:02 -04:00
Jae Hoon (Antonio) Kim 368963243e Export LTC Headers (#1108) 2022-07-30 09:40:02 -04:00
Henry Tu 70395de197 Resolve CI testing failure for Lazy Tensor Core (#1088)
* Xfail unsupported ops

* Register FuncDialect

* Include dynamic_ir in build

* Code reformat

* Enable LTC tests for macOS and Source Build
2022-07-30 09:40:02 -04:00
Jae Hoon (Antonio) Kim 0d16a91656 Add support for lift_fresh op (#1101) 2022-07-30 09:40:02 -04:00
Jae Hoon (Antonio) Kim e37891b997 Default Device Ordinal API (#1079)
* Add default device ordinal API

* Fix reference backend
2022-07-30 09:40:02 -04:00
Antonio Kim de6c135dc3 Fix LTC autogen for CI with nightly PyTorch
- Update llvm-project pin to match main
2022-07-30 09:40:02 -04:00
Henry Tu cec74b8d37 Blacklist _convolution op (#1048)
* Blacklist _convolution op in LTC

* Removed duplicate Torch_AtenSelectScatterOp instance from autogen .td

* Removed duplicate Torch_AtenSliceScatterOp instance from autogen .td
2022-07-30 09:40:02 -04:00
Henry Tu 47bb38d180 Reference Lazy Backend (#1045)
* Changed Example MLIR backend to Reference MLIR backend

* Moved reference_ltc_backend into csrc

* Merged sys_utils.h

* Renamed reference_ltc_backend to reference_lazy_backend

* Addressed review comments

* Update docs with new library name

* Removed _REFERENCE_LAZY_BACKEND from .gitignore

* Added reference_lazy_backend to the TorchMLIRPythonModules dependency list

Fixed typo in `ltc_examples.md`

Missed instance where `ltc_backend` was used instead of `lazy_backend`.
2022-07-30 09:40:02 -04:00
Henry Tu f5acad8512 Prune xfail e2e LTC tests & fix bugs from functionalization pass (#1044)
- Pruned number of xfailed e2e LTC tests from 305 to 134
  - Reviewed every failure to ensure the error genuinely warrants an xfail
- Fixed bug where non-tensor outputs of LTC computation had `.to('cpu')` called, which caused a failure and inflated the xfail count
- Fixed bug with `HBC_basic` test where a constant tensor was created in its constructor without being declared as a buffer, which prevented the device from being updated when the parent `torch.nn.Module` got moved to the `lazy` device
  - Note that this test is still xfail'd due to some unsupported ops. Left a comment about some potential issues that may arise if it gets reenabled in the future
- Updated autogen `GeneratedTorchOps.td` to reflect the latest set of supported ops
- Renamed `aten.zero.functionalization` to `aten.zero` to reflect upstream PyTorch changes
2022-07-30 09:40:02 -04:00
Henry Tu 9de06f3ebd Update Torch MLIR readme 2022-07-30 09:40:02 -04:00
Jae Hoon (Antonio) Kim fb21c9e6cb Integrate Functionalization Pass (#998)
* Fix autogen build dir issue

* Got functionalization pass to compile

* Add slice/diagonal backwards functionalization

* Fix codegen invocation in CMakeLists.txt

* Add functionalization view ops

* Fix logsumexp out functionalization

* Fix ComputationPtr

* Blacklist new_empty op

* Add op comparison

* Remove unnecessary ops

Co-authored-by: Henry Tu <henry.tu@cerebras.net>
2022-07-30 09:40:02 -04:00
Henry Tu 1510eae75d Upstream native_batch_norm and native_batch_norm_backward shape inference functions (#978)
* Removed compute_shape_native_batch_norm

* Removed compute_shape_native_batch_norm_backward
2022-07-30 09:40:02 -04:00
Jae Hoon (Antonio) Kim a62d60829c Refactor autogen (#925) 2022-07-30 09:40:02 -04:00
Henry Tu dfcc26556a Added e2e LTC tests (#916)
* Added e2e LTC Torch MLIR tests

* Fix seed for reproducability

* Check if computation is None before getting debug string

* Updated unit tests, and added numeric tests

* Print name of the model layer that fails numeric validation

* Run LTC e2e test with CI/CD

* Set seed in main function, instead of beginning of execution

* Add comment to specify number of digits of precision

* Fixed typo

* Remove tests for LTC example models

* Added LTC option to torchscript e2e

* Implement compile and run for LTC e2e test

* xfail all tests that use ops that aren't currently supported
2022-07-30 09:40:02 -04:00
Jae Hoon (Antonio) Kim 8312fa535b Refactor Node Lowering (#914) 2022-07-30 09:40:02 -04:00
Jae Hoon (Antonio) Kim d9aee0d7a7 E2E HuggingFace Bert using LTC Backend (#912)
* Update native function definitions

* Add ops to support bert lowering

- Add empty_strided and as_strided

- Restore zeros_like to op blacklist (Without this, tensors will be unintentionally created with a CPU device rather than lazy)

- Check for composite implicit ops and add device data IR

- Also fix codegen for functionalization

* Add autogen to CMakeList

* Remove PyTorch submodule

* Reduced BERT model size

* Print Mark Step status in Torch MLIR LTC debug string

* Apply fixes to work with latest upstream/main

- Pass importOptions into getMlirTypeFromTorchType during NodeImporter::importNode

  Without this, the tensor type created may have a mismatched type as ImportOptions may cause vtensor to be used instead of tensor

* Update shape inference functions

- Fixed compute_shape_native_batch_norm when mean and var are uninitialized

  Previously, the number of shapes returned would be <3 if either mean or val was didn't exist. Instead, we now initialize them with a vector matching the number of channels.

- Implemented compute_shape_mul

- Fixed bug in reshape shape inference error message

* Get MLIR backend more consistent with TS backend

- Remove LazyNativeFunctions::_unsafe_view from autogen

- Blacklist ops to make JIT graph more like output of TS backend

- Print graph when SSA value has mismatch of types and results

- Remove normalize_index from LazyShapeInference

- Fix seeds for LTC example models

* Update and clean up shape inference functions

- Prune shape inference functions

- Add shape inference function for GenerateSlice

- Add shape inference function for GenerateCopy

Co-authored-by: Henry Tu <henry.tu@cerebras.net>
2022-07-30 09:40:02 -04:00
Henry Tu 0c35e607b3 Add static shape for scalar tensors (#833)
* Assume zero rank tensors are scalar

* Run RefineTypes pass on JIT Graph

* Rollback assumption that zero rank tensors are scalar

* Set numSizes to -1 for non-ranked tensors

* Rename RefineTypes to RefineTupleTypes
2022-07-30 09:40:02 -04:00
Henry Tu de5b380143 Bert example and relevant shape inference functions (#831) 2022-07-30 09:40:02 -04:00
Henry Tu 406d1e7538 Use JIT GraphExecutor for execution in example backend (#830)
* Update LazyShapeInference header

* Use JIT GraphExecutor for execution in example backend
2022-07-30 09:40:02 -04:00
Jae Hoon (Antonio) Kim 1bde00c73d Fix LTC Decoupling (#815)
* Initial changes

* Fix up native functions

* Further fix decoupling

* Remove unnecessary ops

* Formatting and copyright banners:

* Add pytorch submodule
2022-07-30 09:40:02 -04:00
Henry Tu cca9fe126e Enable support for LTC Input/Output Mapping (#764)
* Save InputOutputAliases to TorchMlirComputation

* Implement GetResultShape for TorchMlirLoweringContext

* Use optional return type for GetResultShape

* Remove support for aten::detach

With this op enabled, tensors were being copied, which resulted in incorrect aliasing.

* Add newline before printing I/O alias mapping

* Changed printout to use "Input param" as label instead of "Input"

* Remote shape inference function for aten::detach

* Moved implementation of SetUpAlias to MlirLoweringContext

As part of this change, TorchMlirComputation has been moved to the end of mlir_lowering_context.h so that it can access some new structs in TorchMlirLoweringContext

* Use updated PyTorch API

* Remove GetResultShape

Complements this upstream PyTorch PR: pytorch/pytorch#75828

This PR adds support for mapping input and output tensors which alias each other. (e.g. maps input weight tensor in parameter to the same tensor in output after a training iteration)

MLIR: 
func @graph(%arg0: !torch.vtensor<[1,5],f32>, %arg1: !torch.vtensor<[1],si64>, ..., %arg6: !torch.vtensor<[10,5],f32>, %arg7: !torch.vtensor<[10],f32>, ...) {
  ...
  return %arg0, %arg1, %17, %23, ... : !torch.vtensor<[1,5],f32>, !torch.vtensor<[1],si64>, !torch.vtensor<[10,5],f32>, !torch.vtensor<[10],f32>, ...
}

Input/Output Alias Mapping: 
Output: 0 -> Input: 0
Output: 1 -> Input: 1
Output: 2 -> Input: 6
Output: 3 -> Input: 7
The aten::detach op has also been disabled in this PR to fix the issue of tensors not aliasing properly due to copying.
2022-07-30 09:40:02 -04:00
Antonio Kim 615ff1d31c Generate MLIR with shape information via LTC frontend (#742) 2022-07-30 09:40:02 -04:00
Henry Tu a605fe279c Add example Torch MLIR LTC Backend (#725) 2022-07-30 09:40:02 -04:00
Henry Tu 3e9b1cbd36 Added JIT to MLIR lowering (#724)
* Added JIT to MLIR lowering

Lowering to JIT is performed in a way similar to how it's done in the TS LTC backend. After a jit::Graph is constructed, it gets converted to a jit::Function, which is fed into the existing utility to generate an MlirModule in torch-mlir.

* Renamed `csrc/backend` to `csrc/base_lazy_backend`
2022-07-30 09:40:02 -04:00
Jae Hoon (Antonio) Kim 65cf1465ef Fix Torch-MLIR LTC Backend based off latest PyTorch master (#723)
* Changes as a result of the LTC TS backend decoupling

* Fix bugs in BackendImpl and codegen

* Fix based on latest PyTorch master
2022-07-30 09:40:02 -04:00
Jae Hoon (Antonio) Kim c3b20e444c Got LTC working until compile (#689) 2022-07-30 09:40:02 -04:00
Jae Hoon (Antonio) Kim 58338f79a1 Torch-MLIR LTC Backend Lowering Codegen (#621)
* Codegen and build LTC lowering

* Add LazyShapeInference header
2022-07-30 09:40:02 -04:00
Jae Hoon (Antonio) Kim 2f22e2ef40 Add initial LTC backend (#610)
* Add initial LTC backend skeleton

* Disable CI build and move TorchMLIRPyTorch.cmake
2022-07-30 09:40:02 -04:00
PhaneeshB 8b5631d4c5 [MLIR][TORCH] Add decomposition for aten.std.dim Op
Signed-Off By: Phaneesh Barwaria <phaneesh@nod-labs.com>
2022-07-29 23:52:54 +05:30
Vivek Khandelwal c681c3497a [MLIR][TORCH} Fix empty dim cases for the .dim ops
This commit fixes the shape calculation for:
1.) aten.mean.dim
2.) aten.var.dim
3.) aten.sum.dim_IntList op

Also, it fixes the lowering of `aten.mean.dim` and
`aten.sum.dim_IntList` for handling the cases of empty dim list.

Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com
2022-07-29 11:08:57 +05:30
Vivek Khandelwal d386b8f9e5 [MLIR][TORCH] Add decomposition for aten.var.correction op
This commit adds the decomposition for `aten.var.correction` op.

Signed-Off By: Vivek Khandelwal <vivek@nod-labs.com
2022-07-29 11:08:57 +05:30
Quinn Dawkins 11a8901078
[MLIR][TORCH] Add support for multiple indexing tensors for aten.index.Tensor (#1097)
- Includes a canonicalizer for `aten.add.t`needed for successfully lowering the shape function
 - Only offers support for statically sized index tensors when there is more than one
 - Dynamic shape support remains for single indexing tensors
2022-07-28 19:00:02 -04:00
Quinn Dawkins 3c9addf19c Add e2e support for aten.expm1 2022-07-27 12:31:35 +05:30
Kevin Kiningham e8f327cc00 Add lowering to linalg for softplus and log1p
Follows existing conventions for unary operators.
2022-07-25 21:25:57 +05:30
Ashay Rane 72dd04cdb3
Revert "python: trim registration and loading of dialects and passes" (#1093)
This reverts commit ad283c1043, since it's
causing nightly build failures for all platforms.
2022-07-21 09:35:42 -07:00
Ashay Rane ad283c1043
python: trim registration and loading of dialects and passes (#1084)
In the interest of merging upstream LLVM quickly, a previous patch
(7f08169) updated the torch-mlir build to register all dialects and
passes through Python bindings.  This patch limits the dialects and
passes to only those that are used in torch-mlir.

Key to this change are the removal of
`MLIRPythonExtension.RegisterEverything` and the introduction of a new
Python module (`_mlir_libs/_site_initialize_0.py`), where we register
the dialects and passes used by torch-mlir.
2022-07-20 18:34:17 -07:00
Ziheng Jiang c61c99e887
[MHLO] Init MHLO integration. (#1083)
Co-authored-by: Bairen Yi <yibairen.byron@bytedance.com>
Co-authored-by: Jiawei Wu <xremold@gmail.com>
Co-authored-by: Tianyou Guo <tianyou.gty@alibaba-inc.com>
Co-authored-by: Xu Yan <yancey.yx@alibaba-inc.com>
Co-authored-by: Ziheng Jiang <ziheng.jiang@bytedance.com>
2022-07-20 16:18:16 -07:00
Kevin Kiningham 21f905afbe
Emit underscore version of aten.sqrt (#1072) 2022-07-18 23:57:47 -07:00
Ashay Rane 7f08169380
bump llvm tag to 3580daa (#1078)
This patch makes some rudimentary changes to torch-mlir's use of MLIR
Python bindings to work with the most recent LLVM code.  We can perhaps
do better by being more selective in what we link against, instead of
using `MLIRPythonExtension.RegisterEverything`.
2022-07-18 16:49:03 -07:00