Commit Graph

756 Commits (0a0a1b4476ffe1ab59d7b39cf91650c4f68c6caa)
 

Author SHA1 Message Date
Wang Kangyu b33543af85 Add lowering of aten.floor op 2021-11-06 17:31:44 -04:00
nodlabs 5ff823ace9 lowerd Sqrt to linalg
reused clang-format, as changes got deleted
2021-11-06 11:29:46 -04:00
Gaurav Shukla 2ce47dc8e4 [TORCH][MLIR] Add E2E support for aten.expand
This commit adds decomposition of `aten.Expand` to `aten.BroadcastTo`
op.

Signed-Off-by: Gaurav Shukla <gaurav@nod-labs.com>
2021-11-03 23:58:59 +05:30
Prashant Kumar ef897dbb19 Add lowering of `aten.log_softmax` op.
The `aten.log_softmax` is decomposed into `aten.softmax` and
`aten.log` op.
2021-11-03 22:10:05 +05:30
Prashant Kumar 127c7d8e27 Add lowering of `torch.log` op
The lowering of `torch.log` op has been added.

Signed-off-by: Prashant Kumar <prashant@nod-labs.com>
2021-11-02 21:18:00 +05:30
George Petterson 6dde5b347e Add rsub 2021-11-02 09:56:48 -04:00
Prashant Kumar 53b4275ef5 Add lowering of `aten.Int.Tensor` op.
The lowering of `aten.Int.Tensor` op has been added.
The changes has been made as a part of `convert-torch-to-linalg` pass.

Signed-off-by: Prashant Kumar <prashant@nod-labs.com>
2021-11-01 21:58:08 +05:30
Gaurav Shukla 69eaf9a154 [MLIR][TORCH] Add E2E support for `torch.aten.view`
- This commit adds lowering of `aten.View` to `linalg.TensorExpandShape`.
- This lowering will be successful only when one or more static
  dimensions are expanded.
- It also fixes a typo in `ConvertAtenFlattenUsingIntsOp` conversion
  pattern.

Signed-Off-by: Gaurav Shukla <gaurav@nod-labs.com>
2021-10-29 22:33:10 +05:30
Yi Zhang 752abc8d01 Add type promotion code to refine types.
The types have different levels of categories: where
complex > floating > integral > boolean (> means left hand
side has higher category).

The operands have different levels of priorities where:
dimensioned tensor > 0-dim tensor > scalar == wrapped 0-dim tensor.
This is represented by the `ResultTypeState.dimResult`,
`ResultTypeState.zeroResult` and `ResultTypeState..wrappedResult` in
the source code.

For operands of the same priorities, the result type should be the
highest categories with sufficient width to hold all operands.

By default, only the highest priority operands participate in the type
promotion logic. Lower priority operands participate if they are in
a higher category than any higher priority operands.

For example, <[],f32> (lower priority) and <[1], si64> tensor would
result in <[?],f32> tensor because floating > integeral. Another example
<[],f64> (lower priority) and <[1], f32> tensor would result in
<[?], f32> tensor because f32 and f64 are the same category.

The ScalarType enum definition, type promotion table, ResultTypeState
struct definition and some helpers are copied from
aten/src/ATen/native/TypeProperties.*
Other references:
- https://pytorch.org/docs/stable/tensor_attributes.html#type-promotion-doc
- https://github.com/pytorch/pytorch/issues/9515

Other minor changes:
1. Fix `visitExpandLikeOp` to consider cases where the given sizes list
size is larger than the input rank.
2. Add back the somehow deleted `torch.aten.softmax.int` tests in
decompose-complex-ops.mlir.
2021-10-29 11:17:39 -04:00
George Petterson 2ea2ab518b Add contiguous 2021-10-29 11:11:50 -04:00
Sean Silva eb6996d557 Update llvm-project to 6f9c25167d16acff3ff8e4f54a8c14a2a175fc59
- Changes to dialect conversion that result in no-op materializations
  not being created.
2021-10-28 17:43:04 -07:00
Sean Silva c46d48f9f5 Make error reporting a bit better.
- Split out TOSA in the CI.
- Add summary of unexpected test outcomes. This works better when there
  are many XFAIL'ing tests, as it only prints out the error_str on
  FAIL, not on XFAIL. Example here:
  https://gist.github.com/silvasean/c7886ec7b3d35c21563cb09f7c3407da
2021-10-28 13:20:16 -07:00
Sean Silva b02b65cf6e Fix for upstream Torch change.
After https://github.com/pytorch/pytorch/pull/65967 the `graph()` method
is only available on `torch::jit::GraphFunction` now.

Fixes https://github.com/llvm/torch-mlir/issues/388
2021-10-28 11:12:05 -07:00
Suraj Sudhir 7e4ef74774
[tosa] Add Torch.sigmoid fp32 to TOSA (#386)
* [tosa] Add Torch.sigmoid fp32 to TOSA

Signed-off-by: Suraj Sudhir <suraj.sudhir@arm.com>
2021-10-28 10:09:12 -07:00
Prateek Gupta c33a2ca952 [TORCH][MLIR] Add E2E support for aten.permute.
This commit adds lowering of aten.permute to linalg.generic operation.

Signed-Off-By: Prateek Gupta <prateek@nod-labs.com>
2021-10-28 10:25:26 -04:00
stephenneuendorffer 614b889dc6
Enable python extensions when building out of tree (#363) 2021-10-27 17:04:12 -07:00
Sean Silva 30df2ec71b Add min/max/clamp support.
Part of #380

Also
- BoolType is not considered as Scalar
- e2e framework fixes for nan handling
- `tu.rand(..., low=, high=)` support
- delete unused variable (fix warning)
- Add IouOfModule from #380 to e2e test suite (this is a common
  calculation in vision models)

 Your branch is ahead of 'origin/main' by 1 commit.
2021-10-27 13:29:21 -07:00
Ramiro Leal-Cavazos 029c30c060
Fix typo in lazytensor example passes (#385) 2021-10-27 12:40:40 -07:00
Prashant Kumar 5009cbf55c Add lowering of aten.matmul op.
Lowering of `aten.matmul` op is added from torch to linalg dialect.
The different cases correspond to
https://pytorch.org/docs/stable/generated/torch.matmul.html.
TODO: Broadcasting in case of batch-matmul is yet to be taken care of.

Signed-off-by: Prashant Kumar <prashant@nod-labs.com>
2021-10-26 12:45:09 -04:00
Boian Petkantchin e276dbbaa6
Add aten::gelu lowering (#374)
* Print more exception info on error during test execution

* Fix formatting

* Add aten::gelu lowering

Co-authored-by: Boian Petkantchin <boian@nod-labs.com>
2021-10-25 16:16:01 -07:00
Sean Silva a6943ef90c Rename `tosa-to-linalg-on-tensors` to `tosa-to-linalg`
The pass name changed upstream.
2021-10-25 20:43:54 +00:00
Stella Laurenzo 47209539a8 Bump llvm-project to f1b922188ead5ca492c8d8edd47921b013a22ae0.
Includes a fix to use `add_mlir_public_c_api_library` for Torch-MLIR's CAPI library, which is now required (note: upstream sample has it the right way).

Disabled a TOSA test per discussion: https://github.com/llvm/torch-mlir/issues/379
2021-10-25 13:22:07 -07:00
Ramiro Leal-Cavazos 8bfb819d35 Fix bug with transpose of negative dims
Summary:
This commit fixes an off-by-one error in how negative dimensiosn were
being handled in the lowering of transpose. This commit also adds
tests to transpose and unsqueeze to test negative dimensions.
2021-10-25 15:50:55 -04:00
Stella Laurenzo a23d77100b Set some wheel building optimization options.
* Also adds a requirements.txt and updates docs to reference it versus stringy pip install.
* Adds doc with instructions on creating a wheel.

Fixes #370
2021-10-25 18:30:53 +00:00
Stella Laurenzo 92ae692387
Filter checks to only run on push to main branch. (#372)
Keeps redundent pull request and push workflows from running when pushing to branches in the main repo.
2021-10-21 21:23:21 -07:00
Stella Laurenzo fe69bb339c
Bump llvm-project to 3d92722f74993969243d1400bc3257ca3d03902f. (#369)
* Picks up Python configure changes (was pinned to a bad intermediate commit).
* Uses the new mlir_configure_python_dev_packages() to ensure CMake python is found consistently.
* Fixes the JIT importer to build as a MODULE vs SHARED (needed for linking to Python as a module, per config changes).
* Adds some notes to the README to help folks build a smaller set focused just on this project.
2021-10-21 21:09:00 -07:00
George Petterson 22aeb967c5 Add ones 2021-10-21 14:46:59 -04:00
Yi Zhang abfaf8c577 Add aten.ne.bool to make CI pass 2021-10-21 14:45:41 -04:00
George Petterson 7c47b9a0c8 Formatting fix 2021-10-19 13:33:31 -04:00
George Petterson 8853dfbc74 Add broadcast 2021-10-19 13:33:31 -04:00
Yi Zhang a459e09ab7 E2e support for aten.softmax.int and aten.embedding
- Added a DecomposeComplexOps pass to decompose complex torchOps.
- Refactored `visitAtenArgmaxOp` and `visitAtenAnyDimOp` to
`visitReductionAlongDimIntOp`.
- Moved some helper functions into
torch-mlir/Dialect/Torch/Utils/Utils.h to be shared by multiple files.
- Added support for f64 tensor as argument and return types.
2021-10-18 17:57:45 -04:00
Yi Zhang 0902438882 Update llvm-project to a54f4eae0e1d0ef5adccdcf9f6c2b518dc1101aa
This brings in https://reviews.llvm.org/D110797. PRs that are in
progress will need to use scripts provided by
https://llvm.discourse.group/t/psa-removed-arithmetic-ops-from-standard/4455.
2021-10-18 13:36:42 -04:00
Boian Petkantchin 393bd211c7 In readme build instructions force CMake to use Python virtual env
By default CMake is quite happy to use any Python that it finds
to fit the version requirements.
2021-10-14 10:27:56 -07:00
dan 7750d2173a add argmax lowering
Add argmax lowering from torch to linalg
2021-10-13 14:31:16 -04:00
Sean Silva 19e9fc4ef1 Bring some more order to the e2e error reporting situation.
- Move `run_pipeline_with_repro_report` to a more common place, and use it
  consistently
- Attach a `torch.debug_module_name` to the enclosing `builtin.module`
  op to allow for self-contained error reporting (not needing to pass
  the names around.
- Remove redundant error reporting in linalg_on_tensors_backend.py and
  tosa_backend.py (their respective backend abstract base classes now
  take care of the error reports themselves)
- Save off original value of sys.stderr, rather than always resetting to
  `sys.__stderr__`. This is just more hygienic, and allows nesting if
  desired.
2021-10-08 13:00:12 -07:00
Stephen Neuendorffer fc5d03c081 post-commit review cleanup for #356
see https://github.com/llvm/torch-mlir/pull/356
2021-10-08 11:28:40 -07:00
Sean Silva 0c5c84d63d Add a basic TOSA E2E backend.
We lower through linalg-on-tensors and use RefBackend to run it.
This adds enough support for a "tanh" op. Adding more ops should be
fairly mechanical now that things are wired up. Run with:
```
./tools/torchscript_e2e_test.sh -c tosa
```

The backend structure is very similar to linalg-on-tensors based E2E
backends and is a nice parallel (see `tosa_backend.py`). Actually, this
forced a nice refactoring to the layering here. We removed
`torchscript-module-to-linalg-on-tensors-backend-pipeline` and instead
require separately running
```
torchscript-function-to-torch-backend-pipeline,torch-backend-to-linalg-on-tensors-backend-pipeline
```
This highlights the step that lowers to the "torch backend contract"
of cleaned up `torch` dialect ops is a critical step in the lowering.
Going forward, that is the key load-bearing contract of the torch-mlir
project, not the linalg-on-tensors backend contract.

Recommended review order:
- `TorchToTosa.cpp` / `TorchToTosa/basic.mlir`
- `python/torch_mlir_e2e_test/torchscript/configs/tosa_backend.py` and
  the new `utils.py` file there.
- `python/torch_mlir_e2e_test/tosa_backends/linalg_on_tensors.py` and
  `abc.py` in that directory for the TOSA backend e2e interface.
- other misc mechanical changes
2021-10-08 09:59:45 -07:00
stephenneuendorffer df12cc0c37
Add support for out-of-tree building (#356)
Currently I haven't attempted to get the python bits working.
Instead the python bits are forcibly disabled.
2021-10-07 21:44:15 -07:00
Stephen Neuendorffer 00d42ccaee Add tool substitutions to support out-of-tree builds 2021-10-07 21:16:43 -07:00
Sean Silva b74779ff8d Remove stray references to non-existent .td file.
This must be old copypasta.
2021-10-08 01:21:23 +00:00
Sean Silva 00e94c70ef Remove packaging info for now.
Need to do manylinux build for it to work across many machines.
2021-10-07 01:38:42 +00:00
Sean Silva de8ea59d52 Fix feedback from Anush. 2021-10-07 01:28:13 +00:00
Sean Silva a157d6c946 Add instructions for installing from packages. 2021-10-07 01:20:49 +00:00
Sean Silva b6628fe774 Mark releases as "published".
This allows `pip` to see them.
2021-10-06 22:48:21 +00:00
Sean Silva 4a8d05e4a5 Add torch_mlir snapshot packages.
This closely follows IREE's
[schedule_snapshot_release.yml](f2f153d394/.github/workflows/schedule_snapshot_release.yml (L1))
workflow.

The snapshot releases can be installed with:
```
python -m pip install torch_mlir -f "https://github.com/llvm/torch-mlir/releases"
```
2021-10-06 14:50:31 -07:00
Sean Silva 712445eaa8 Bring back Python packaging.
Will add a CI job that builds and uploads snapshot packages next.
2021-10-05 13:33:30 -07:00
Sean Silva dcab39146f Remove the last mentions of npcomp from torch-mlir
These snuck through.
2021-10-05 20:17:23 +00:00
dan 2e1498ad11 add i64 support to refbackend 2021-10-05 15:12:44 -04:00
Yi Zhang fadd76e9b8 E2e for MiniLM-L6-H384-uncased-sst2
Replace the original BertSequenceClassification with this new one.
The ops needed to support are identical.
2021-10-05 12:45:19 -04:00
Phoenix Meadowlark c3e0a1e1dc Add shell markdown formatting to README.md 2021-10-04 11:59:25 -07:00