Summary:
Previously we were touching up unbacked bindings between Dynamo and AOTAutograd in strict export, but the logic had a bug: if an unbacked symint gets substituted by a backed symint, we would put the backed symint in the unbacked bindings (the check `is_symbol` was not enough here).
This PR fixes this logic, and moreover, moves it into the serializer instead, because we don't need this adjustment outside serde.
Test Plan: added test
D68880766
Pull Request resolved: https://github.com/pytorch/pytorch/pull/146115
Approved by: https://github.com/pianpwk
Previously, in non-strict path, we always error when trying to inplace update a constant tensor because those constant tensors are not actually wrapped by functional tensors. This is correct behaviour in torch.compile, because dynamo makes all constant tensors into buffers and AOTDispatcher just lifts them and wraps them in functional tensors. However, in non-strict, there is no such step that registers constants as buffers so AOTDispatcher panics when it sees these dangling constant tensors when functioanalizing.
Due to recent change in the IR, this is no longer an issue in non-strict path because we don't call AOTDispatcher at training IR level, but now it is a problem for both strict and non-strict when we lower to inference. (lowering to inference is very similar to non-strict tracing) As a result, we have at least one external (https://github.com/pytorch/pytorch/issues/141336) and internal issues reported due to this difference.
To fix this, there are two ways:
1. Make functionalization be aware of constant tensors and map them to functional tensors on the fly. This makes functionalization invariant uglier and could potentially open up a gate for more nasty bugs.
2. Special handle this in export. This seems more aligned with what dynamo does today so i think we should do it this way. I think the current state could benefit from more refactors to make the run_deocmpositions to be more similar to strict export (because both of them now handle this constant registerinig logic) but it is bit complicated to do it now because strict export version of this logic is also not complete because it doesn't take into account of export graph renaming pass etc). I will follow up with more refactors after this PR (T213466691) to unblock users faster.
For future reference:
Why are we not doing "turning constants into non-persistent buffers and never de-register"? The reason is because in some internal models, they rely on module.to to reliably work to move params/buffers to correct device. As a result, buffers are moved while constants are not. In composibility meeting, we agreed that export won't do device agnostic tracing going forward (it will provide a way to specify FakeTensor in CPU that can be configured to be run on GPU), so after that is done, we can always turn constants into non-persistent buffers which will simplify export's constant handling.
Differential Revision: [D68610739](https://our.internmc.facebook.com/intern/diff/D68610739)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/145593
Approved by: https://github.com/avikchaudhuri
Summary:
Previously we were touching up unbacked bindings between Dynamo and AOTAutograd in strict export, but the logic had a bug: if an unbacked symint gets substituted by a backed symint, we would put the backed symint in the unbacked bindings (the check `is_symbol` was not enough here).
This PR fixes this logic, and moreover, moves it into the serializer instead, because we don't need this adjustment outside serde.
Test Plan: added test
Differential Revision: D68880766
Pull Request resolved: https://github.com/pytorch/pytorch/pull/146115
Approved by: https://github.com/pianpwk
Adds unbacked bindings during deserialization. These are carried by a node's metadata, and map pending fresh unbacked symbols to paths to such symbols inside the corresponding example value carried by the node's metadata.
Since it is awkward to serialize paths, we only serialize the names of these symbols and reconstruct the paths on deserialization, using a shape env util. We also need to bump counters for unbacked symbols here, because the shape env util we use to create these symbols (when deserializing example values) don't do so, and not doing so makes later passes (like `run_decompositions`) crash because new unbacked symbols don't get new names.
This is enough for non-strict. For strict, the unbacked bindings and example values in node metadata can get out of sync, because of running AOTAutograd as an additional step after Dynamo. So we have to sync those back.
Differential Revision: [D68232274](https://our.internmc.facebook.com/intern/diff/D68232274/)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/144894
Approved by: https://github.com/pianpwk
Summary:
Add experimental support for torch.nn.Module as input types.
Before this change, we don't support module inputs but recently we saw some interesting use cases like gpt-fast https://github.com/pytorch-labs/gpt-fast/blob/main/generate.py#L68 where we directly pass in a module input for different variants of the same models.
Since we don't really care about non-param or non-buffer states in non strict mode, we don't care about those either and pretend they are like plain constants during tracing. We treat any module input like a nested container of tensor, and each time we will automatically register a pytree handler for these module types to flatten its state dict into a group of tensors. We will just inline any module method call during tracing like we did for `self` module in export_for_training. This will make input modules' behavior very similar to the training module in typical case, except that we don't record the inputs as parameter or buffers but rather just plain user inputs.
Test Plan: buck run mode/opt caffe2/test:test_export -- -r test_module_input
Differential Revision: D67680827
Pull Request resolved: https://github.com/pytorch/pytorch/pull/143925
Approved by: https://github.com/tugsbayasgalan
Summary:
A reland of https://github.com/pytorch/pytorch/pull/142426.
Copying the description over here:
For torch.export (strict and non-strict), we don't do functional decomposition. Instead, we preserve the custom triton ops as custom ops. This is because we want the exported program to be high-level and serializable.
The alternative:
If we decompose the custom op to a functional hop and make it a node in exported program, we need to figure out ways of serializing the hop and its arguments, which can be triton.jited python functions and triton dtypes. This is undesireble because:
it can be tedious to maintain layer that serialize the jited function (e.g. with a string) and dtypes.
changes to triton or the serialization logic for triton arguments can be BC breaking
exported program will expose the implementation detail (i.e. triton source code) for a specific backend (GPU) to users, which mixes levels of abstraction.
Future plans:
After this PR, in the short term, we expect users to have a seperate aot_compile stage that compiles the exported program into a Cubin file on the same machine that users call export, which does autotuning and removes triton dependency and serve the model with Cubin. This guarantees that triton changes won't break BC.
In the long term, we may export multiple cubins for the triton op directly.
Test Plan: see new tests.
Differential Revision: D67879685
Pull Request resolved: https://github.com/pytorch/pytorch/pull/144284
Approved by: https://github.com/zou3519
We added an is_export flag under torch.compiler.is_exporting. This comes handy when we try to do some special logic in user-level and system-level (e.g. in upper of the stack).
In increasing-scope:
- `_is_fx_tracing` is set to True when we use under symbolic_trace or make_fx.
- `is_exporting` is set to True when we're doing strict or non-strict export, which internally has a step that calls make_fx and set _is_fx_tracing to be True.
- `is_compiling` is set to True when we're either doing strict, non-strict export or torch.compile.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/142425
Approved by: https://github.com/avikchaudhuri
For better tracking, we need to make maybe aliasing/mutating ops with proper tag. We need to special case native_batch_norm because it is not a CIA but has a wrong schema. I guess native_batch_norm will be removed at some point, so until then we just keep it around.
D60347117
Pull Request resolved: https://github.com/pytorch/pytorch/pull/131990
Approved by: https://github.com/bdhirsh
Differential Revision: [D65362160](https://our.internmc.facebook.com/intern/diff/D65362160)
State after this IR:
1. For the tests that require inference IR, they are replaced with ep.run_decomp({}) so export_for_training_run_decomp is sort of redundant but i guess it is still nice that multiple round of retracing still working. In general, we need some auditing to reduce our redundant testing coverages.
2. After this PR landed and not get reverted for a week or so, i will replace the export_for_training calls with export as they are the same thing now.
3. Added more tests to also cover now "deprecated" old IR by patching export to use old export. For reviewers, please look at the internal version.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/139511
Approved by: https://github.com/ydwu4, https://github.com/angelayi, https://github.com/avikchaudhuri
Summary: The gm_torch_level can be a _LazyGraphModule(GraphModule) instead of a GraphModule. When we call .recompile(), GraphModule populates the self._out_spec, but _LazyGraphModule(GraphModule).recompile() doesn't populate it.
Test Plan: CI
Differential Revision: D65902135
Pull Request resolved: https://github.com/pytorch/pytorch/pull/140608
Approved by: https://github.com/tugsbayasgalan
In this diff, i make test_torchbind.py tests to handle training IR. Today in the training IR, we don't see the effect token and HOP because this happens at the FunctionalTensorMode. Maybe in the future, we should move this logic up to the training IR so that writing passes etc on training Ir is safer. But for the migration purposes, i think it is ok for now. I also fixed two bugs:
1. ep.module() doesn't register all aliased constants in the module.
2. When we retrace, we need to fakify the original Torchbind object.
3. We don't run any DCE on training IR so we need to add some more torch ops to verifier.
Differential Revision: [D64853530](https://our.internmc.facebook.com/intern/diff/D64853530)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/138658
Approved by: https://github.com/ydwu4, https://github.com/zhxchen17
Summary:
`torch.fx.Interpreter.run()` only takes args as input. Currently we pass kwargs as well which causes errors during retracing.
Flatten the kwargs and concat them with args will solve the issue.
Several previously failing tests under `_retraceability_non_strict` now passes.
Test Plan:
```
buck2 test @//mode/dev-nosan //caffe2/test:test_export -- -r _retraceability_non_strict
```
Differential Revision: D64980053
Pull Request resolved: https://github.com/pytorch/pytorch/pull/138927
Approved by: https://github.com/angelayi
Previously we would error when trying to preserve the call signature for a module when it was called multiple times. This PR can now do this without erroring. The fix is to propagate call indices in a few more places.
Note that while this works in the presence of params, buffers, and tensor constants, preserving call signatures for multiple calls to a module when buffers are mutated is not supported yet. This is future work. The main problem is that we do not have enough metadata to `copy_` mutated buffers at the end of each call to a module, so the next call can read those buffers at the beginning. Making this work will likely need some explicit tracking of intermediate values of mutated buffers when collecting metadata during functionalization in export.
Note also that we stop short of creating a single graph out of multiple graphs: that is still future work. So the unflattened module will still have different targets `n`, `n@1`, `n@2`, etc. for each call when we ask the module call signature of `n` to be preserved. However it is way easier to swap all of these targets with a replacement that behaves similar to the original, because all of these calls will respect the original module call signature. (In particular, any constant inputs will be carried by the calls.)
Differential Revision: D64406945
Pull Request resolved: https://github.com/pytorch/pytorch/pull/137999
Approved by: https://github.com/tugsbayasgalan
Summary:
Tensor constants can show up through wrapped methods, so that they may not always be found in constant attributes. They need to be fakified and their meta vals need to be found to create graph signatures nevertheless. Otherwise non-strict barfs.
Longer term maybe we should pull this fakification up in non-strict.
Test Plan: added test
Differential Revision: D64480272
Pull Request resolved: https://github.com/pytorch/pytorch/pull/138091
Approved by: https://github.com/tugsbayasgalan
Resolve#137540
Summary:
We might get different state_dict and named_parameters result when the module has registered custom state_dict_hooks.
For exported_program's state_dict, we want the state_dict to reflect the actual module hierarchy at runtime, and it might be different from the model's state_dict() output if the model has state_dict hooks.
To do weight swapping, one needs to either re-export or turn-off the hooks when saving model's state_dict().
Previously, ExportedProgram uses nn.Module's state_dict() method to populate its own state_dict, but it doesn't work for some models (e.g. llama3_3_vision) because ExportedProgram's state_dict and an nn.Module's state_dict have some subtle differences semantically.
nn.Module's state_dict is about how the state should be serialized, and it reflects the structure of the original user model code. In contrast, export specializes on a “run” of a model, and its state_dict needs to reflect the runtime module hierarchy.
One example where these two are different is TorchTune's Llama3_2_vision text decoder. Here, a FusionLayer is added as a local optimization and it is not part of the "static model definition". In runtime, we have mod.layers[3].layer.sa_norm.scale.
But in nn.Module's state_dict, the authors of the model added a state_dict hook to remove the "layer" in mod.state_dict() to reflect the static model definition, so we have mod.state_dict()["layers.3.sa_norm.scale"].
In this Diff, we change ExportedProgram to populate its state_dict using named_parameters() and named_buffers() instead. So in ExportedProgram's state_dict, we have "layers.3.layer.sa_norm.scale", which reflects the runtime module hierarchy.
Now one problem this presents is weight swapping. Since ExportedProgram's state and the model's state is not the same anymore, weight swapping procedure also needs to change slightly.
In internal Ads and RecSys models deployment, weight swapping is where they have one model that is currently being being deployed and serving traffic, and they want to swap out the weights with newly trained model weights without having to redo the whole exporting/lowering process and create a new artifact. So they would move the deployed model’s pointer to the state dict over to the new state dict. Because of this, it’s previously a requirement that the FQNs are matching between the exported and the eager model’s state dict.
The new ExportedProgram's state dict still supports weight swapping, but the state_dict to be swapped needs to be obtained from torch.export.exported_program instead of model.state_dict() if the model has state_dict hooks.
The new requirement is that the FQNs are matching between the exported’s state dict and the state_dict obtained from `_disabled_load_state_dict_hooks(M)` context manager. One benefit of having this new API is that we are now in full control within export of gathering and updating the model state.
If a model doesn't have any state_dict hooks, one can still use model.state_dict() for weight swapping, so it's BC.
Test Plan:
```
buck2 run 'fbcode//mode/dev-nosan' fbcode//caffe2/test:test_export -- -r test_export_for_training_with_state_dict_hooks
```
Differential Revision: D64080561
Pull Request resolved: https://github.com/pytorch/pytorch/pull/137609
Approved by: https://github.com/angelayi, https://github.com/pianpwk
Summary: Fixes a couple of problems: constants didn't have metadata before creating graph signatures, and graph signatures weren't updated when lifting constants.
Test Plan: fixed test
Differential Revision: D64081786
Pull Request resolved: https://github.com/pytorch/pytorch/pull/137547
Approved by: https://github.com/tugsbayasgalan
Summary: We had attribute assignment detection and handling of registered buffer assignments when using `aot_autograd`, but not when using just `make_fx`. Fixed.
Test Plan: expanded coverage of `test_state_tensors` to use `export` instead of `torch.export.export`
Differential Revision: D63802576
Pull Request resolved: https://github.com/pytorch/pytorch/pull/137240
Approved by: https://github.com/tugsbayasgalan
Removing `_transform_shapes_for_default_dynamic` and `assume_static_by_default=False` as added in https://github.com/pytorch/pytorch/pull/133620.
This reverts back to `assume_static_by_default=True` with the use of dynamo decorators (e.g. `maybe_mark_dynamic, mark_static`, instead) for handling Dim.AUTO & Dim.STATIC instead. This is easier to maintain, as it doesn't requiring reasoning about "inverting" the dynamic_shapes specs, and also opens up usage of other decorators (`mark_dynamic, mark_unbacked`).
On the user side this change has no effect, but internally this means dynamic behavior is determined only by the `dynamic_shapes` specs (ignoring user-side input decorators following https://github.com/pytorch/pytorch/pull/135536), but transferring this information for _DimHints via decorators, for Dynamo/non-strict to create symbolic_contexts accordingly, e.g. 7c6d543a5b/torch/_dynamo/variables/builder.py (L2646-L2666)
One caveat is we don't raise errors for dynamic decorators on the user side, since we don't know if they're from user markings, or from re-exporting with inputs we've previously marked.
Differential Revision: D63358628
Pull Request resolved: https://github.com/pytorch/pytorch/pull/136591
Approved by: https://github.com/avikchaudhuri
Summary:
A bit of refactoring to prepare to remove `None` as a way to specify static dimensions in dynamic shapes, given we already have `Dim.STATIC` for the same purpose. We will now warn whenever this happens. However no tests were modified because problematic uses of `None` still need to behave as they do today, until we are ready to remove support. It should be easy to port tests by replacing the warning function to raise instead.
Note that other uses of `None`, such as for entire values (tensor or non-tensor) remain as is. Moving forward this should be the only purpose of `None` (at least externally).
Finally, there's a bit of confusion in our representation now because `AUTO` also internally transforms to `None`. Renamed dynamic_shapes to transformed_dynamic_shapes where this happens. Overall the two forms (pre and post transformation) have different properties so should probably not be represented in the same format in the future.
Test Plan: existing
Differential Revision: D62040729
Pull Request resolved: https://github.com/pytorch/pytorch/pull/134877
Approved by: https://github.com/pianpwk
Fixes#133252
In strict mode, we have this routine for mapping traced parameters to their FQNs using tensor ids. Currently we assume there's at least 1 unique FQN for each traced parameter, but this seems to break with parameter reuse when call_module nodes are present. Adding a test case where this breaks.
Fixes this by assigning the same FQN to all traced parameters with the same tensor id. This is fine because we return the original state_dict for the EP, and the unflattener has its own routine of handling aliasing: https://github.com/pytorch/pytorch/pull/125758
Pull Request resolved: https://github.com/pytorch/pytorch/pull/134500
Approved by: https://github.com/angelayi
Summary:
Today there is no good mechanism to detect progress of non-strict export line-by-line in user code. This caused some pain recently in trying to find the exact line of user code that was triggering a bug where the process appeared stuck because deep down something was calling some symbolic shapes code that was suffering some exponential blowup.
This PR adds a environment variable for extended debugging that will log the line of user code corresponding to every torch function call. It only works in non-strict export for now. Prefix setting this environment variable with `TORCH_LOGS` enabled for `export` logs at `DEBUG` level (i.e., with a `+` prefix), i.e.,.:
```
TORCHEXPORT_EXTENDED_DEBUG_CURRENT_LOC=1 TORCH_LOGS="+export" ...
```
This will show logs with something like:
```
...
prim::device called at .../example.py:4284 in foo
TensorBase.item called at .../example.py:4277 in bar
...
```
We already have an existing place to intercept torch functions where we process data-dependent errors in non-strict, so parking the logging there. An alternative place we could be doing this is where we add `stack_trace` metadata when generating code, but unfortunately at least the example that motivated this gets stuck before generating code, so that would be too late.
Test Plan: ran it on some sample commands
Differential Revision: D61692156
Pull Request resolved: https://github.com/pytorch/pytorch/pull/134298
Approved by: https://github.com/angelayi
Refactors construction of ExportGraphSignature object for export & training IR, explicitly creating AOTAutograd signature for training IR. This will be helpful for upcoming refactors for placeholder naming & runtime asserts prettifying.
Changes:
- dedups `make_argument_spec` call, moved to export/graph_signature.py
- `_sig_to_specs` wrapped into new function `_convert_to_export_graph_signature`, directly converts GraphSignature -> ExportGraphSignature
- `_make_fx_helper` explicitly creates AOTAutograd GraphSignature object
Pull Request resolved: https://github.com/pytorch/pytorch/pull/134059
Approved by: https://github.com/angelayi, https://github.com/ydwu4
Starter version of automatic dynamic shapes for export.
Creates enums `DIM.AUTO`, `DIM.STATIC`, allowing user to specify `AUTO` for dims in dynamic_shapes specs, meaning that corresponding dims are treated as dynamic, and relevant guards will do what's necessary (e.g. refine ValueRanges, set replacements based on equality, or even set static) without raising ConstraintViolationErrors. Basically allows the user to say, "a bunch of these dims can be dynamic, let export do model analysis and return the program with maximum possible dynamism, without complaining".
The usage for specifying `dynamic_shapes` is now:
```
AUTO -> dynamic by default, return whatever produce_guards() says, even if it's static
None/int/STATIC -> static
Dim/DerivedDim -> same as before - will complain if the min/max range is invalid, or if dims related to this are unspecified.
```
Caveat 1: specifying `AUTO` for a dim won't guarantee it'll be dynamic:
- specifying `AUTO` for a dim will return the maximum possible dynamism given your program and other specified constraints, but this can still mean you'll get a static program. For example, with the program below, x is specified dynamic, but it's equal to y, which is specified static, and with how we currently do things we won't promote y to dynamic, but will demote(?) x to static. So this can be surprising if you don't fully know your model, and/or missed one of your other inputs when specifying auto-dynamic shapes.
```
class Foo(torch.nn.Module):
def forward(self, x, y):
return x + y
inputs = (torch.randn(6), torch.randn(6))
export(Foo(), inputs, dynamic_shapes={"x": (DIM.AUTO,), "y": None})
```
Caveat 2: specifying `AUTO` and Dims in the same spec is still problematic:
- The way Dims/DerivedDims are currently handled is very strict. A Dim represents a symbol, and we require a user to specify the symbol for all dims governed by the symbol - that's why we've seen errors in the past like `The values of x must always be related to y by ...`, asking the user to specify the exact relation as in the program. We also require the specified min/max range to be a subset of the valid range from model analysis. All this doesn't compose well with specifying `AUTO` just yet - for example in the program below, ideal behavior could be to return a dynamic program, where `dx = x.size(0) = y.size(0)` has range (3,6). Unfortunately this crashes, and correct behavior is to specify `dx` for both inputs. So currently we raise a UserError and crash if both Dims + `AUTO` are present in the spec.
```
class Foo(torch.nn.Module):
def forward(self, x, y):
return x + y
inputs = (torch.randn(6), torch.randn(6))
export(Foo(), inputs, dynamic_shapes={"x": (DIM.AUTO,), "y": {0: Dim("dx", min=3, max=6)}}) # this doesn't work, because x & y and related
```
Implementation details:
This is done by setting `assume_static_by_default=False`, and doing a transform on the `dynamic_shapes` spec to preserve semantics. `assume_static_by_default=False` will treat unspecified dims or Nones as dynamic. This is the opposite of what `export.export()` currently does - unspecified Dims/Nones are treated as static. Historically this static-by-default behavior, where the user deals with fewer guards, has been desirable, and we would like to respect that in this implementation. So this internal spec transformation is added, `_transform_shapes_for_default_dynamic()`, does the spec conversion necessary to be compatbile with dynamic by default. Specifically, AUTOs are converted into Nones, and Nones/unspecified dims are filled in with explicitly static constraints.
For example, this would look like, for a 3-d tensor: `{0: DIM.AUTO, 1: None, 2: Dim("dx")} -> {0: None, 1: 32, 2: Dim("dx")}`
This does seem overly complicated, but it's done to preserve dynamic shapes semantics for `torch._dynamo.export()`, which already uses `assume_static_by_default=False`, and follows the same process for generating shape constraints , via `_process_dynamic_shapes`. There the semantics are:
```
None/unspecified: dynamic by default
Dim/DerivedDim: also a strict assertion
```
If we don't care about BC for `_dynamo.export(dynamic_shapes)`, then we can just modify semantics for `_process_dynamic_shapes()` and change all the relevant tests in `test/dynamo/test_export.py`.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/133620
Approved by: https://github.com/avikchaudhuri
