Main: `RuntimeError: element 0 of tensors does not require grad and does not have a grad_fn`
This PR: graph breaks and eager applies the mutation, new tensors are tracked
Fixes https://github.com/pytorch/pytorch/issues/109505 (the original bug does not occur, but a new bug where the mutation isn't applied - because AOTAutograd is not `requires_grad` mutation aware - is mitigated)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/113163
Approved by: https://github.com/bdhirsh
Fixes https://github.com/pytorch/pytorch/issues/112446
This is a doozy of a PR, there's a few important things to keep in mind here:
1) We MUST lift all tensors accessed via attrs to inputs, getattr is a no go in the graph, it violates the aot_autograd contract. Furthermore, aot_autograd does not know how to apply in-place ops to intermediary tensors that are attributes (aka from getattr) anyway. Views from ops are fine.
2) `.grad` access handling in dynamo peeks at the underlying value, the real tensor, because re-piping FakeTensors already made with this fake_mode through builder anew is a no go.
3) We have no proper mechanism for updating the hint / grapharg.example (the real value in (2) above) midway through trace
Therefore, what we need to do is reconcile the difference in grad stashed on grapharg.example. The easiest way to do this is lazily, upon .grad access, by reading the new value off the right fake tensors. We can then make a tensor using that data as a hint to VariableBuilder to make the right VariableTracker. Note that the example value used here (torch.zeros) in the PR, is a dummy value only used as a tracing hint, it does not leak out into real runtime code.
Alternatively, we could implement accumulate_grad_ in python...
Pull Request resolved: https://github.com/pytorch/pytorch/pull/112811
Approved by: https://github.com/jansel
Fixes https://github.com/pytorch/pytorch/issues/113030
Alias information needs to be applied in eager before we can continue to trace the graph.
----
Perhaps this is too strict - couldn't we fx trace through the in-graph (pointer) aliasing, and track mutations through fake tensors instead, and still apply the aliasing mutation epilogue for further mutations outside of graph? 🤔
Regardless, it didn't seem to work too well when I tried this. Seems that `Tensor.__setattr__` doesn't work well in fx graph.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/113043
Approved by: https://github.com/ezyang, https://github.com/voznesenskym
Major change in this PR is to make torch context manager class a separate ```TorchCtxManagerClassVariable```, since we have dynamo implementation for these ctx managers.
I was thinking to wrap them as ```UserDefinedClassVariable``` and do dispatch at ```USCVariable.call_function```, but it seems almost the same amount of work and this way is more clear.
This is on the way of moving ```TorchVariable``` to ```TorchFunctionVariable``` which will only handle the functions who would be allowed in graph (e.g, ```torch.sin```) and constant folded (e.g, ```torch.is_floating_point```). All other torch functions would be go through skip/inline rules, and would be wrapped as ```UserFunctionVariable``` (for inlined) and ```SkipFilesVariable``` (for skipped).
The next steps:
* Wrap torch modules, classes, objects as regular ```PythonModuleVariable```, ```UserDefinedClassVariable``` and ```UserDefinedObjectVariable```.
* Generate the allow in graph torch functions list and wrap them as ```TorchFunctionVariable```.
* Finally merge ```skipfiles.check``` and ```is_allowed``` into one function ```allow_skip.check(fn)``` which would return a Enum of allow, skip and inline.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/111622
Approved by: https://github.com/jansel
This PR implements 2 things:
1. support the device agnostic stream and runtime APIs captured by the dynamo.
2. support the stream methods(include the event) captured by the dynamo.
Here are details for 1st.
Previously the stream captured in dynamo was tightly bind to CUDA. Here we implement a global singleton container named `StreamMethodContainer` for different backends to register their associated stream methods to dynamo. When import the backend’s product, the stream operations can be registered directly by calling
```
device_stream_method = {'current_stream': method_1,
'create_stream_context': method_2,
'set_stream': method_3,
'set_stream_by_id': method_4}
torch._dynamo.stream.register_stream_method(device_name, device_stream_method)
```
Stream methods need to be passed in this API according to the precise semantics represented by the dict key in `device_stream_method`. After register, these methods can be used by dynamo to capture the stream operations in users’ script, for example, get the current stream or set the specific stream. Additionally, the wrapped stream variable and the stream context variable are changed to be the device-agnostic, the proxy functions of these variables are assigned by the associated methods in the container. All of this are illustrated in the below. Below is a illustration.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/108312
Approved by: https://github.com/jansel, https://github.com/jgong5
We want to get to a point where most UserErrors link to exportdb examples. This PR makes passing case names non-optional to make this intent clearer and encourage developers who raise UserErrors to make or point to examples that make fixing such errors more obvious for users.
In addition, sometimes there are multiple examples that are relevant to an error. Thus this PR also enables passing multiple case names.
Retry of #110733 which was reverted due to a landrace.
Differential Revision: [D50087148](https://our.internmc.facebook.com/intern/diff/D50087148/)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/110878
Approved by: https://github.com/gmagogsfm, https://github.com/tugsbayasgalan
The motivation for removing this is already present in the pre-PR comments. Copying it
~~~
# NB - SuperSource is a weird one.
# it is our only source with 2 bases, so we use the objec
# as the base, rather than the type, since an invocation
# like super(Foo, foo) is represented here, the source object base is more spiritually
# aligned with the instance, rather than the type.
# This whole construction is questionable tho, and we should probably find a way to
# avoid this exception to our otherwise nice source parentage invariant.
~~~
Instead of using super(a, b), we can use `type(b).__mro__[index]`.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/110475
Approved by: https://github.com/jansel
We want to get to a point where most `UserError`s link to `exportdb` examples. This PR makes passing case names non-optional to make this intent clearer and encourage developers who raise `UserError`s to make or point to examples that make fixing such errors more obvious for users.
In addition, sometimes there are multiple examples that are relevant to an error. Thus this PR also enables passing multiple case names.
Differential Revision: [D50020465](https://our.internmc.facebook.com/intern/diff/D50020465/)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/110733
Approved by: https://github.com/zhxchen17
Ideally all `_dynamo.exc.UserError`s should have "case names", i.e., link to examples in `exportdb`.
This PR adds case names to several instances of `_dynamo.exc.UserError`. In particular, looking at coverage based on `UserErrorType`:
* `DYNAMIC_CONTROL_FLOW`, `ANTI_PATTERN`, and `STANDARD_LIBRARY` are fully covered.
* `CONSTRAINT_VIOLATION` and `DYNAMIC_DIM` have no coverage. We don't seem to have any dedicated examples of specifying dynamic shapes in `exportdb` (although they are used in some other examples without explanation, to avoid some specialization that would make such examples moot).
* `INVALID_INPUT` is only partly covered. Frankly this is tedious to cover via examples.
Differential Revision: [D49928518](https://our.internmc.facebook.com/intern/diff/D49928518/)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/110555
Approved by: https://github.com/angelayi, https://github.com/ydwu4
Fix: #107315
This PR enables dynamo to trace through the `pytree` API by inlining its functions. In
order to do so, a few details of `pytree` had to be changed.
In summary, this PR:
- Introduces `TreeSpecVariable` for representing `TreeSpec` instances
- Specializes `<type>.__bases__` call, returning a `TupleVariable`
- Enables the call to `id` builtin function for every variable that implements
`as_python_constant` method
- Specializes `ConstantVariable.call_method` for its (un)flatten functions
- Implements `UserDefinedObjectVariable.as_python_constant`
- Modifies `pytree` by:
- Make `SUPPORTED_NODES` a map of ids (instead of types) to `NodeDef`
- Removed `functools.wraps` function, since it can't be inlined
Pull Request resolved: https://github.com/pytorch/pytorch/pull/108533
Approved by: https://github.com/ezyang, https://github.com/voznesenskym
ghstack dependencies: #109201
before the PR, for HF's ModelOutput class, we use dicts.py/DataClassVariable with our own implementation on __getItem__, __setAttr__, __setItem__. There is a risk that ModelOutput logic may change since it is a user code
after the PR, we inline __getItem__, __setAttr__, __setItem__ using dicts.py/CustomizedDictVariable so the logic always keep AA
unit test
* python test/dynamo/test_model_output.py -k test_HF_bert_model_output
test on HF benchmark
* python benchmarks/dynamo/huggingface.py -d cuda --inference --accuracy --progress --inductor --print-dataframe-summary 2>&1
* all metric are the same before/after the PR, including pass rate, unique_graphs, graph_breaks, unique_graph_breaks
* before the PR: P790393916
* after the PR: P790368991
Pull Request resolved: https://github.com/pytorch/pytorch/pull/105044
Approved by: https://github.com/jansel
RFC: https://github.com/pytorch/rfcs/pull/54
First commit is the contents of https://github.com/Quansight-Labs/numpy_pytorch_interop/
We have already been using this in core for the last few months as a external dependency. This PR pulls all these into core.
In the next commits, I do a number of things in this order
- Fix a few small issues
- Make the tests that this PR adds pass
- Bend backwards until lintrunner passes
- Remove the optional dependency on `torch_np` and simply rely on the upstreamed code
- Fix a number dynamo tests that were passing before (they were not tasting anything I think) and are not passing now.
Missing from this PR (but not blocking):
- Have a flag that deactivates tracing NumPy functions and simply breaks. There used to be one but after the merge stopped working and I removed it. @lezcano to investigate.
- https://github.com/pytorch/pytorch/pull/106431#issuecomment-1667079543. @voznesenskym to submit a fix after we merge.
All the tests in `tests/torch_np` take about 75s to run.
This was a work by @ev-br, @rgommers @honno and I. I did not create this PR via ghstack (which would have been convenient) as this is a collaboration, and ghstack doesn't allow for shared contributions.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/106211
Approved by: https://github.com/ezyang
This PR adds initial dynamo support for DTensor, in particular, it:
- allows DTensor be passed into a compiled function, and allow fakify
DTensor during dynamo tracing by turning the inner local tensor to meta
tensor.
- We use `allow_in_graph` to include `DTensor` and `DTensor.from_local` to be represented as `TorchVariable`
- The dtensor created becomes a normal `TensorVariable` and it would insert any tensor operations to the output graph just like torch.Tensor
- note that dtensor have a new instance method `redistribute` compare to plain tensor, and we currently special handle it in `TensorVariable`
`from_local` and `redistribute` both accepts some non-trival metadata as arguments (i.e. DeviceMesh, Placement) which fx.Graph does not support. In order to let these two APIs appear in the dynamo captured graph, we encoded the metadata into a new_function (like `functools.partial`) and the new function only accepts prim args (i.e. tensor), then we put `call_function` with this new_function to the graph. This is suggested by @ezyang. The underlying rationale here is that the metadata will not change across the graph invocations so it's safe to encode them.
Captured graph:
```
def forward(self, L_x_ : torch.Tensor):
l_x_ = L_x_
# File: /scratch/wanchaol/work/pytorch/test/distributed/_tensor/test_dtensor.py:685, code: dt = DTensor.from_local(x, mesh, [Shard(0)], run_check=False)
prim_from_local = torch__dynamo_variables_torch_prim_from_local(l_x_, run_check = False); l_x_ = None
# File: /scratch/wanchaol/work/pytorch/test/distributed/_tensor/test_dtensor.py:686, code: return dt.redistribute(mesh, [Replicate()]).to_local() + 2
prim_redistribute = torch__dynamo_variables_tensor_prim_redistribute(prim_from_local); prim_from_local = None
to_local = prim_redistribute.to_local(); prim_redistribute = None
add = to_local + 2; to_local = None
return (add,)
```
Pull Request resolved: https://github.com/pytorch/pytorch/pull/103146
Approved by: https://github.com/voznesenskym
