For Traceable FSDP2, the most common use case is to have `fullgraph=False` for forward pass (to allow user-level graph breaks), and `fullgraph=True` for compiled autograd backward pass (required for queue_callback support).
With `torch._dynamo.compiled_autograd=True`, previously we are not able to set different `fullgraph` config value for forward vs. backward pass, since `rebuild_ctx` just reuses the forward compile config as-is. This PR adds `torch._dynamo.config.compiled_autograd_kwargs_override` config to allow forcing `fullgraph=True` for CA Dynamo tracing.
With this PR, we can remove standalone compiled autograd ctx manager usage in Traceable FSDP2 unit tests, and consolidate on using `torch._dynamo.compiled_autograd=True`.
Test commands:
- `pytest -rA test/distributed/_composable/fsdp/test_fully_shard_compile.py::TestFullyShardCompile::test_transformer_backend_inductor_fullgraph_True`
Pull Request resolved: https://github.com/pytorch/pytorch/pull/136967
Approved by: https://github.com/xmfan
Summary: Previously is_fbcode just checked whether the checkout was git or not. This is extremely error prone. Lets make it fool-proof.
Test Plan: unit tests
Reviewed By: masnesral
Differential Revision: D63545169
Pull Request resolved: https://github.com/pytorch/pytorch/pull/136871
Approved by: https://github.com/masnesral
PYTHONPATH=$(pwd) python benchmarks/update_hint_benchmark.py out
as of this diff, compile_time_instruction_count counts the number of instruction from within
convert_frame.compile_inner
```
update_hint_regression,compile_time_instruction_count,10522459165
```
will add result from CI once populated.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/133834
Approved by: https://github.com/aorenste
Setting `torch._dynamo.config.skip_fsdp_hooks = True` is required for graph-break compiled FSDP2, thus setting it to default will make this adoption easier. If users want to use Traceable FSDP2, they can set this to False manually (which will allow FSDP2 hooks to be traced through).
Pull Request resolved: https://github.com/pytorch/pytorch/pull/133531
Approved by: https://github.com/awgu
ghstack dependencies: #133532
This PR implements an opt-in configuration option for synchronizing compilation across all ranks at the end of Dynamo tracing (and potentially, other places in the future). There are two pieces to this PR:
1. Implementing infrastructure for compiler collectives (DistributedState/LocalState, the actual collective)
2. Using this infrastructure to synchronize automatic dynamic choices across all ranks
The infrastructure in part one can be used for other purposes, just add more (serializable) fields to LocalState.
Here is how automatic dynamic synchronization works:
1. Preflight in "torch/_dynamo/variables/builder.py": On the first Dynamo trace run, we trace without automatic dynamic at all; we assume all Tensor inputs that are not otherwise marked are static. This run is purely to collect all Tensor input sizes in the program.
2. torch/_dynamo/output_graph.py: At the end of the first Dynamo trace run, we perform a compiler collective to distribute all Tensor input sizes to all ranks. Then, we restart Dynamo
3. Apply the updates in "torch/_dynamo/variables/builder.py": Now that we have all sizes for every rank, we now update frame state with the observed sizes for all ranks, in rank order. Under the assumption that frame state is consistent on all ranks, this series of updates will preserve consistency.
For future work, it would be safer if we force a consistent hint on all ranks; this is more involved as we have to interpose in fakification.
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/130935
Approved by: https://github.com/jansel
With the current state of export's dynamic shapes, we struggle with guards and constraints that are beyond the current dynamic shapes language, expressed with dims and derived dims. While we can compile and guarantee correctness for guards within the current language (e.g. min/max ranges, linear relationships, integer divisibility) we struggle to dynamically compile guards which extend beyond that.
For these "complex" guards, we typically do either of the following: 1) raise a constraint violation error, along the lines of "not all values of <symbol> in the specified range satisfy <guard>", with or without suggested fixes, 2) specialize to the provided static values and suggest removing dynamism, or 3) fail compilation due to some arbitrary unsupported case. Previous [work](https://github.com/pytorch/pytorch/pull/124949) went towards resolving this by disabling forced specializations, instead allowing the user to fail at runtime with incorrect inputs.
In this PR, relying on [hybrid backed-unbacked symints](https://github.com/pytorch/pytorch/issues/121749), [deferred runtime asserts](https://github.com/pytorch/pytorch/blob/main/torch/fx/passes/runtime_assert.py), and the function [_is_supported_equivalence()](d7de4c9d80/torch/fx/experimental/symbolic_shapes.py (L1824)), we add a flag `_allow_complex_guards_as_runtime_asserts` which allows the user to compile exported programs containing these guards and maintain dynamism, while adding correctness checks as runtime assertions in the graph.
Hybrid backed-unbacked symints allow us to easily bypass "implicit" guards emitted from computation - guards that we ~expect to be true. Popular examples revolve around reshapes:
```
# reshape
def forward(self, x, y): # x: [s0, s1], y: [s2]
return x.reshape([-1]) + y # guard s0 * s1 = s2
This leads to the following exported program
class GraphModule(torch.nn.Module):
def forward(self, x: "f32[s0, s1]", y: "f32[s2]"):
sym_size_int: "Sym(s2)" = torch.ops.aten.sym_size.int(y, 0)
mul: "Sym(-s2)" = -1 * sym_size_int; sym_size_int = None
sym_size_int_1: "Sym(s0)" = torch.ops.aten.sym_size.int(x, 0)
sym_size_int_2: "Sym(s1)" = torch.ops.aten.sym_size.int(x, 1)
mul_1: "Sym(s0*s1)" = sym_size_int_1 * sym_size_int_2; sym_size_int_1 = sym_size_int_2 = None
add: "Sym(s0*s1 - s2)" = mul + mul_1; mul = mul_1 = None
eq: "Sym(Eq(s0*s1 - s2, 0))" = add == 0; add = None
_assert_scalar = torch.ops.aten._assert_scalar.default(eq, "Runtime assertion failed for expression Eq(s0*s1 - s2, 0) on node 'eq'"); eq = None
view: "f32[s0*s1]" = torch.ops.aten.view.default(x, [-1]); x = None
add_1: "f32[s0*s1]" = torch.ops.aten.add.Tensor(view, y); view = y = None
return (add_1,)
```
Another case is symbol divisibility:
```
def forward(self, x): # x: [s0, s1]
return x.reshape([-1, x.shape[0] - 1]) # Eq(Mod(s0 * s1, s0 - 1), 0)
```
Applying deferred runtime asserts also helps dynamic compilation for "explicit" complex guards that typically cause problems for export. For example we can generate runtime asserts for not-equal guards, and complex conditions like the following:
```
class Foo(torch.nn.Module):
def forward(self, x, y):
# check that negation of first guard also shows up as runtime assertion
if x.shape[0] == y.shape[0]: # False
return x + y
elif x.shape[0] == y.shape[0] ** 3: # False
return x + 2, y + 3
elif x.shape[0] ** 2 == y.shape[0] * 3: # True
return x * 2.0, y * 3.0
```
For the above graph we will generate 3 runtime assertions: the negation of the first 2, and the 3rd condition as a guard.
One additional benefit here over the current state of exported programs is that this adds further correctness guarantees - previously with explicit complex guards, if compilation succeeded, the guards would be ignored at runtime, treated as given.
As shown above, the runtime asserts appear as math ops in the graph, generated by the sympy interpreter, resulting in an _assert_scalar call. There is an option to avoid adding these asserts into the graph, by setting `TORCH_DYNAMO_DO_NOT_EMIT_RUNTIME_ASSERTS=1`. This results in the "original" computation graph, with dynamism, and any incorrect inputs will fail on ops during runtime. Further work could go into prettifying the printer, so the majority of the graph isn't guard-related.
Ideally this PR would subsume and remove the recently added [_disable_forced_specializations](https://github.com/pytorch/pytorch/pull/124949) flag, but that flag still handles one additional case of specialization: single-variable equalities where the symbol is solvable for a concrete value: see this [PR](https://github.com/pytorch/pytorch/pull/126925)
This PR doesn't change any behavior around data-dependent errors/unbacked symints yet, that could be further work.
NOTE: will take naming change suggestions for the flag :)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/127129
Approved by: https://github.com/avikchaudhuri
tlparse prints failure description like this
> dynamic shape operator: aten._unique2.default; to enable, set torch._dynamo.config.capture_dynamic_output_shape_ops = True
adding os env var to set it easier for testing
Pull Request resolved: https://github.com/pytorch/pytorch/pull/127017
Approved by: https://github.com/jackiexu1992
The big idea is that floats are treated as Tensors on input/output to the FX graph, but on the inside, we immediately call item() on the synthetic Tensor and record regular float operations on it. Canonicalization to Tensor operations will happen in a standalone FX pass. This behavior is controlled by `specialize_float` config variable when set to False.
The generated graph looks like this for the test `test_unspec_float_output`:
```
def forward(self, L_x_: "f32[3]", L_y_: "f32[]"):
l_x_ = L_x_
l_y_ = L_y_
# File: /data/users/ezyang/a/pytorch/test/dynamo/test_unspec.py:511 in f, code: return x + 1, y * 2
add: "f32[3]" = l_x_ + 1; l_x_ = None
item: "Sym(zf0)" = l_y_.item(); l_y_ = None
mul: "Sym(2*zf0)" = item * 2; item = None
scalar_tensor: "f32[]" = torch.scalar_tensor(mul); mul = None
return (add, scalar_tensor)
```
The ingredients:
* **torch/_dynamo/variables/builder.py** When `specialize_float` is False, we wrap float literals with `wrap_symfloat`. This is an unholy mashup of `wrap_symint` and `wrap_unspecialized_primitive`. The overall strategy is that we first generate a tensor argument (because that's what we want to show up into the FX graph), but then immediately call item() on the tensor argument to get a SymNodeVariable, which we will do the rest of the tracing with. Importantly, this SymNodeVariable is backed with the source of the original float: this means we can guard on the resulting value (something we could NOT do with UnspecializedPythonVariable). This has to be done manually, because if you literally call item() on the tensor, you will end up with an unbacked float. There is a bit of copy paste from wrap_symint and wrap_unspecialized_primitive which we can try to factor out, but this really is its own thing and you should review every line of code in the function.
* **torch/fx/experimental/symbolic_shapes.py** We now can generate guards on float inputs, and these guards are handled inside of ShapeEnv. So we need to be able to allocate (backed!) float symbols, and produce guards for them. Fairly straightforward generalization.
* **torch/_dynamo/codegen.py** I also need to maintain the invariant that there are no float outputs to the FX graph. I chose to do this at codegen time. When we detect a SymNodeVariable on the return stack for a float, we on the fly convert it (via `as_tensor`) to a TensorVariable, which is the true output. We then special case the output bytecode to call item() on it again. The tensor conversion is memoized on SymNodeVariable since we typically run the code generation process twice.
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/125325
Approved by: https://github.com/lezcano, https://github.com/jansel
Turning on guard_nn_modules adds large number of guards, so we are bound to take a perf hit. But the perf hit is small. These are the numbers
First we observe that compared to Python guards, C++ guards give around 6x speedup. This reduces the total time spent in guards. This is shown in the last column (cpp_guards/inductor_optimized_latency). The worst model is around 1.61%, with most of the models below 1%. I think this is good enough signal to turn the config on.
One might also wonder how much guard slowdown occurs with `guard_nn_modules=True`. This is the table
For most models, the guard overhead with nn module guards is under 2x. There are a few outliers, where the slowdown is really high and for those models we spend 1%-2% time in C++ guards as shown in first table.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/125202
Approved by: https://github.com/ezyang
Summary: [#123231](https://github.com/pytorch/pytorch/pull/123231) adds cudagraph supports for more types of functions (i.e., cudagraph managed input mutation). These newly supported functions may have mutated static inputs, leading to assertion errors in some workload which skip cudagraph previously. This diff adds a config to opt in the new feature.
Test Plan: ci
Differential Revision: D56481353
Pull Request resolved: https://github.com/pytorch/pytorch/pull/124754
Approved by: https://github.com/eellison
When minifying, the after-aot minifier ignores non-floating values by
default but does check them when running the the initial graph dump step.
This means we may capture a graph that doesn't fail the tester and doesn't have
any meaningful divergence.
For example, the derivative of `elu(x)` depends on `x > 0` so this value is
saved for backwards and so becomes a graph output. However, the difference
between `FLT_MIN` and `0` in `x` is now enough to trigger an accuracy failure.
I fix this by adding a config variable and environment variable to ignore these
non floating point values.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/123006
Approved by: https://github.com/ezyang
ghstack dependencies: #123005
Currently when there is a print/warning in the graph, dynamo graph breaks causing export to fail. However export would like to just skip over these print/warning calls: https://github.com/pytorch/pytorch/issues/113792.
Additionally there's a torch.compile feature request to "reorder prints" so that instead of graph breaking when hitting prints/logging, we can skip over these prints to create larger compiled graphs, and then print the results out after those compiled graphs: https://github.com/pytorch/pytorch/issues/93739. This PR also adds the `reorderable_logging_functions` config for users to register logging functions to be reordered (like `print` or a custom logging function). Printout of the bytecode after reordering the prints looks like the following: P914736600
There are some limitations to the printing right now:
* You can only register logging functions, not methods
* Inputs to the logging functions can only be tensors, constants, and format strings
* Inputs to the logging functions which will later be mutated in-place will not be printed correctly
TODO: Add the following tests
* print function with argument of nested data structure;
* print function with argument of nested data structure being updated inside of compile region (this would test if we handle side effect correctly);
* custom defined logging functions with nn.Module or nn.Module attribute arguments;
* custom defined logging functions with submodule input/output as arguments (we need to handle the mapping and fused-out value);
* custom defined logging functions with tensor argument and mutation inside of the function (TBD: this may increase memory usage);
Pull Request resolved: https://github.com/pytorch/pytorch/pull/116106
Approved by: https://github.com/yanboliang
Currently when there is a print/warning in the graph, dynamo graph breaks causing export to fail. However export would like to just skip over these print/warning calls: https://github.com/pytorch/pytorch/issues/113792.
Additionally there's a torch.compile feature request to "reorder prints" so that instead of graph breaking when hitting prints/logging, we can skip over these prints to create larger compiled graphs, and then print the results out after those compiled graphs: https://github.com/pytorch/pytorch/issues/93739. This PR also adds the `reorderable_logging_functions` config for users to register logging functions to be reordered (like `print` or a custom logging function). Printout of the bytecode after reordering the prints looks like the following: P914736600
There are some limitations to the printing right now:
* You can only register logging functions, not methods
* Inputs to the logging functions can only be tensors, constants, and format strings
* Inputs to the logging functions which will later be mutated in-place will not be printed correctly
TODO: Add the following tests
* print function with argument of nested data structure;
* print function with argument of nested data structure being updated inside of compile region (this would test if we handle side effect correctly);
* custom defined logging functions with nn.Module or nn.Module attribute arguments;
* custom defined logging functions with submodule input/output as arguments (we need to handle the mapping and fused-out value);
* custom defined logging functions with tensor argument and mutation inside of the function (TBD: this may increase memory usage);
Pull Request resolved: https://github.com/pytorch/pytorch/pull/116106
Approved by: https://github.com/yanboliang
Tacotron2 causes massive loop unrolling resulting in very large graphs (26k nodes) which was causing inductor (and tracing itself) to choke.
The unrolling size is controlled by the environment variable TORCHDYNAMO_MAX_LOOP_UNROLL_NODES which defaults to the arbitrary value 5000.
This updates the tacotron2 timings as follows:
eager timing: 3m:23s -> 35s
aot_eager timing: 4m:12s -> 39s
inductor timing: 22m:24s ->1m
For reference the big loop in tacotron2 was this one (model.py[405]):
```
while len(mel_outputs) < decoder_inputs.size(0) - 1:
decoder_input = decoder_inputs[len(mel_outputs)]
mel_output, gate_output, attention_weights = self.decode(decoder_input)
mel_outputs += [mel_output.squeeze(1)]
gate_outputs += [gate_output.squeeze(1)]
alignments += [attention_weights]
```
which gets unrolled and inlined adding about 36 nodes to the graph per iteration.
Fixes#98467
Relates to #102839 which hopefully will result in a better fix.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/120023
Approved by: https://github.com/yanboliang
**Summary**
The reducer of `DistributedDataParallel` is implemented with C++ and it is not easy to trace the allreduce launched in the reducer. This PR modifies `DistributedDataParallel` to launch one allreduce per gradient when `compiled_autograd` is enabled. The changes allow us to use `compiled_autograd` to trace the allreduce and later be optimized (fused) in the Inductor.
**Key Logic**
1. If `ddp_python_hook` is True, we assume `compiled_autograd` is used. `DistributedDataParallel` registers `compiled_accum_grad_hook` for all parameters.
2. In the first forward() call, if `DistributedDataParallel` is not compiled, all `compiled_accum_grad_hook` are deregistered. If `DistributedDataParallel` is compiled, all `compiled_accum_grad_hook` will be compiled by `compiled_autograd`.
3. `compiled_accum_grad_hook` launches an allreduce to reduce the gradient of the parameter.
**Bucketing**
The compiled backward is slow because there is no bucketing for the allreduces. We rely on Inductor to bucket the allreduces.
The bucketing is done in a separate PR.
Differential Revision: [D49428482](https://our.internmc.facebook.com/intern/diff/D49428482/)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/110662
Approved by: https://github.com/wconstab
I feel it's easier to open a new PR rather than iterating on the previous PR (https://github.com/pytorch/pytorch/pull/105257 ) since this is more like a rewrite.
In this PR, instead of changing GraphModule directly which can easily causes BC issue, I create a LazyGraphModule class as Zachary & Jason suggested in comments from the previous PR.
The difference between LazyGraphModule and GraphModule is mainly about how re-compile for the graph module happens. In GraphModule the recompilation happens 'eagerly': constructing a GraphModule will cause the recompilation. While in LazyGraphModule, we just mark the module as needing recompilation. The real recompilation only happens when absolutely required (e.g. call forward method, access the code property etc.). In a lot of cases in torch.compile, the real recompilation eventually is not triggered at all. This can save a few seconds of compilation time.
By default, GraphModule rather than LazyGraphModule is used. `use_lazy_graph_module(True)` context manager can be used to pick LazyGraphModule instead. This has been applied to the torch.compile stack.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/117911
Approved by: https://github.com/jansel
Due to not all tests in the Dynamo shard actually running in CI, we've
started to bitrot on this implementation. Since our plan is to trace
into the functorch implementations instead of construct a HOP
(which is what capture_func_transforms=True does), let's turn off this
config by default.
Test Plan:
- Tests
Pull Request resolved: https://github.com/pytorch/pytorch/pull/115267
Approved by: https://github.com/voznesenskym, https://github.com/guilhermeleobas
`install_config_module` makes a regular module into a ConfigModule with
extra methods defined on it. mypy thinks those extra methods (or module
functions) are undefined since it cannot analyze something so
dynamic. As a workaround, I've created a fake module that defines these
extra functions, which I import into the config modules during type
checking.
As part of this change, I've also added more types to config_utils.py
and enabled typechecking for torch/_dynamo/config.py.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/112130
Approved by: https://github.com/jansel
Attempt number 2 at https://github.com/pytorch/pytorch/issues/108950.
Improves debugging for guard failures/recompilations by:
- only running guard fail reason generation during recompilation, instead of when a guard fails during dynamo cache lookup (so generating guard failure reasons is not on the critical path)
- ~~always reporting all guard failures~~ Reports the first-failing guard failure for each cache entry.
We don't expect a performance hit since the guard fail reasons are only generated at recompile time rather than runtime. Perf benchmark to check this (https://hud.pytorch.org/benchmark/torchbench/inductor_with_cudagraphs?startTime=Fri,%2027%20Oct%202023%2017:42:43%20GMT&stopTime=Fri,%2003%20Nov%202023%2017:42:43%20GMT&granularity=hour&mode=training&dtype=amp&lBranch=gh/williamwen42/62/head&lCommit=f4724f5ffc6d17ceae513a42fc18627be7b85482&rBranch=main&rCommit=29f3d392bf230072e3bffae37b078e770cae1956). We may also need to verify this on benchmarks where guard fails are common.
Sample script:
```python
import torch
def generate_data(b):
return (
torch.randn(b, 3, 32, 32).to(torch.float32).cuda(),
torch.randint(1000, (b,)).cuda(),
)
from torchvision.models import resnet18
def init_model():
return resnet18().to(torch.float32).cuda()
model = init_model()
model_opt = torch.compile(model, dynamic=False)
for b in range(16, 32):
data = generate_data(b)
model_opt(data[0])
```
Sample logs:
```bash
(/data/users/williamwen/py310-env) [williamwen@devgpu020.odn1 /data/users/williamwen/pytorch (wwen/log-all-guards)]$ python playground5.py
/data/users/williamwen/pytorch/torch/_inductor/compile_fx.py:141: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance.
warnings.warn(
[2023-11-06 14:50:47,605] torch._dynamo.convert_frame: [WARNING] torch._dynamo hit config.cache_size_limit (8)
[2023-11-06 14:50:47,605] torch._dynamo.convert_frame: [WARNING] function: 'forward' (/data/users/williamwen/torchvision/torchvision/models/resnet.py:284)
[2023-11-06 14:50:47,605] torch._dynamo.convert_frame: [WARNING] last reason: tensor 'L['x']' size mismatch at index 0. expected 16, actual 24
[2023-11-06 14:50:47,605] torch._dynamo.convert_frame: [WARNING] To log all recompilation reasons, use TORCH_LOGS="recompiles".
[2023-11-06 14:50:47,605] torch._dynamo.convert_frame: [WARNING] To diagnose recompilation issues, see https://pytorch.org/docs/master/compile/troubleshooting.html.
(/data/users/williamwen/py310-env) [williamwen@devgpu020.odn1 /data/users/williamwen/pytorch (wwen/log-all-guards)]$ TORCH_LOGS="recompiles" python playground5.py
/data/users/williamwen/pytorch/torch/_inductor/compile_fx.py:141: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance.
warnings.warn(
[2023-11-06 14:53:31,591] torch._dynamo.guards.__recompiles: [DEBUG] Recompiling function forward in /data/users/williamwen/torchvision/torchvision/models/resnet.py:284
[2023-11-06 14:53:31,591] torch._dynamo.guards.__recompiles: [DEBUG] triggered by the following guard failure(s):
[2023-11-06 14:53:31,591] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 16, actual 17
[2023-11-06 14:53:41,333] torch._dynamo.guards.__recompiles: [DEBUG] Recompiling function forward in /data/users/williamwen/torchvision/torchvision/models/resnet.py:284
[2023-11-06 14:53:41,333] torch._dynamo.guards.__recompiles: [DEBUG] triggered by the following guard failure(s):
[2023-11-06 14:53:41,333] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 17, actual 18
[2023-11-06 14:53:41,333] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 16, actual 18
[2023-11-06 14:53:50,463] torch._dynamo.guards.__recompiles: [DEBUG] Recompiling function forward in /data/users/williamwen/torchvision/torchvision/models/resnet.py:284
[2023-11-06 14:53:50,463] torch._dynamo.guards.__recompiles: [DEBUG] triggered by the following guard failure(s):
[2023-11-06 14:53:50,463] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 18, actual 19
[2023-11-06 14:53:50,463] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 17, actual 19
[2023-11-06 14:53:50,463] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 16, actual 19
[2023-11-06 14:53:59,848] torch._dynamo.guards.__recompiles: [DEBUG] Recompiling function forward in /data/users/williamwen/torchvision/torchvision/models/resnet.py:284
[2023-11-06 14:53:59,848] torch._dynamo.guards.__recompiles: [DEBUG] triggered by the following guard failure(s):
[2023-11-06 14:53:59,848] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 19, actual 20
[2023-11-06 14:53:59,848] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 18, actual 20
[2023-11-06 14:53:59,848] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 17, actual 20
[2023-11-06 14:53:59,848] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 16, actual 20
[2023-11-06 14:54:08,549] torch._dynamo.guards.__recompiles: [DEBUG] Recompiling function forward in /data/users/williamwen/torchvision/torchvision/models/resnet.py:284
[2023-11-06 14:54:08,549] torch._dynamo.guards.__recompiles: [DEBUG] triggered by the following guard failure(s):
[2023-11-06 14:54:08,549] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 20, actual 21
[2023-11-06 14:54:08,549] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 19, actual 21
[2023-11-06 14:54:08,549] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 18, actual 21
[2023-11-06 14:54:08,549] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 17, actual 21
[2023-11-06 14:54:08,549] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 16, actual 21
[2023-11-06 14:54:17,795] torch._dynamo.guards.__recompiles: [DEBUG] Recompiling function forward in /data/users/williamwen/torchvision/torchvision/models/resnet.py:284
[2023-11-06 14:54:17,795] torch._dynamo.guards.__recompiles: [DEBUG] triggered by the following guard failure(s):
[2023-11-06 14:54:17,795] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 21, actual 22
[2023-11-06 14:54:17,795] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 20, actual 22
[2023-11-06 14:54:17,795] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 19, actual 22
[2023-11-06 14:54:17,795] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 18, actual 22
[2023-11-06 14:54:17,795] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 17, actual 22
[2023-11-06 14:54:17,795] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 16, actual 22
[2023-11-06 14:54:27,430] torch._dynamo.guards.__recompiles: [DEBUG] Recompiling function forward in /data/users/williamwen/torchvision/torchvision/models/resnet.py:284
[2023-11-06 14:54:27,430] torch._dynamo.guards.__recompiles: [DEBUG] triggered by the following guard failure(s):
[2023-11-06 14:54:27,430] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 22, actual 23
[2023-11-06 14:54:27,430] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 21, actual 23
[2023-11-06 14:54:27,430] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 20, actual 23
[2023-11-06 14:54:27,430] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 19, actual 23
[2023-11-06 14:54:27,430] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 18, actual 23
[2023-11-06 14:54:27,430] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 17, actual 23
[2023-11-06 14:54:27,430] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 16, actual 23
[2023-11-06 14:54:36,744] torch._dynamo.guards.__recompiles: [DEBUG] Recompiling function forward in /data/users/williamwen/torchvision/torchvision/models/resnet.py:284
[2023-11-06 14:54:36,744] torch._dynamo.guards.__recompiles: [DEBUG] triggered by the following guard failure(s):
[2023-11-06 14:54:36,744] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 23, actual 24
[2023-11-06 14:54:36,744] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 22, actual 24
[2023-11-06 14:54:36,744] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 21, actual 24
[2023-11-06 14:54:36,744] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 20, actual 24
[2023-11-06 14:54:36,744] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 19, actual 24
[2023-11-06 14:54:36,744] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 18, actual 24
[2023-11-06 14:54:36,744] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 17, actual 24
[2023-11-06 14:54:36,744] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 16, actual 24
[2023-11-06 14:54:36,744] torch._dynamo.convert_frame: [WARNING] torch._dynamo hit config.cache_size_limit (8)
[2023-11-06 14:54:36,744] torch._dynamo.convert_frame: [WARNING] function: 'forward' (/data/users/williamwen/torchvision/torchvision/models/resnet.py:284)
[2023-11-06 14:54:36,744] torch._dynamo.convert_frame: [WARNING] last reason: tensor 'L['x']' size mismatch at index 0. expected 16, actual 24
[2023-11-06 14:54:36,744] torch._dynamo.convert_frame: [WARNING] To log all recompilation reasons, use TORCH_LOGS="recompiles".
[2023-11-06 14:54:36,744] torch._dynamo.convert_frame: [WARNING] To diagnose recompilation issues, see https://pytorch.org/docs/master/compile/troubleshooting.html.
[2023-11-06 14:54:45,922] torch._dynamo.guards.__recompiles: [DEBUG] Recompiling function _forward_impl in /data/users/williamwen/torchvision/torchvision/models/resnet.py:266
[2023-11-06 14:54:45,922] torch._dynamo.guards.__recompiles: [DEBUG] triggered by the following guard failure(s):
[2023-11-06 14:54:45,922] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 24, actual 25
[2023-11-06 14:54:54,691] torch._dynamo.guards.__recompiles: [DEBUG] Recompiling function _forward_impl in /data/users/williamwen/torchvision/torchvision/models/resnet.py:266
[2023-11-06 14:54:54,691] torch._dynamo.guards.__recompiles: [DEBUG] triggered by the following guard failure(s):
[2023-11-06 14:54:54,691] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 25, actual 26
[2023-11-06 14:54:54,691] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 24, actual 26
[2023-11-06 14:55:03,591] torch._dynamo.guards.__recompiles: [DEBUG] Recompiling function _forward_impl in /data/users/williamwen/torchvision/torchvision/models/resnet.py:266
[2023-11-06 14:55:03,591] torch._dynamo.guards.__recompiles: [DEBUG] triggered by the following guard failure(s):
[2023-11-06 14:55:03,591] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 26, actual 27
[2023-11-06 14:55:03,591] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 25, actual 27
[2023-11-06 14:55:03,591] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 24, actual 27
[2023-11-06 14:55:12,384] torch._dynamo.guards.__recompiles: [DEBUG] Recompiling function _forward_impl in /data/users/williamwen/torchvision/torchvision/models/resnet.py:266
[2023-11-06 14:55:12,384] torch._dynamo.guards.__recompiles: [DEBUG] triggered by the following guard failure(s):
[2023-11-06 14:55:12,384] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 27, actual 28
[2023-11-06 14:55:12,384] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 26, actual 28
[2023-11-06 14:55:12,384] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 25, actual 28
[2023-11-06 14:55:12,384] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 24, actual 28
[2023-11-06 14:55:21,442] torch._dynamo.guards.__recompiles: [DEBUG] Recompiling function _forward_impl in /data/users/williamwen/torchvision/torchvision/models/resnet.py:266
[2023-11-06 14:55:21,442] torch._dynamo.guards.__recompiles: [DEBUG] triggered by the following guard failure(s):
[2023-11-06 14:55:21,442] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 28, actual 29
[2023-11-06 14:55:21,442] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 27, actual 29
[2023-11-06 14:55:21,442] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 26, actual 29
[2023-11-06 14:55:21,442] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 25, actual 29
[2023-11-06 14:55:21,442] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 24, actual 29
[2023-11-06 14:55:30,315] torch._dynamo.guards.__recompiles: [DEBUG] Recompiling function _forward_impl in /data/users/williamwen/torchvision/torchvision/models/resnet.py:266
[2023-11-06 14:55:30,315] torch._dynamo.guards.__recompiles: [DEBUG] triggered by the following guard failure(s):
[2023-11-06 14:55:30,315] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 29, actual 30
[2023-11-06 14:55:30,315] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 28, actual 30
[2023-11-06 14:55:30,315] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 27, actual 30
[2023-11-06 14:55:30,315] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 26, actual 30
[2023-11-06 14:55:30,315] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 25, actual 30
[2023-11-06 14:55:30,315] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 24, actual 30
[2023-11-06 14:55:39,839] torch._dynamo.guards.__recompiles: [DEBUG] Recompiling function _forward_impl in /data/users/williamwen/torchvision/torchvision/models/resnet.py:266
[2023-11-06 14:55:39,839] torch._dynamo.guards.__recompiles: [DEBUG] triggered by the following guard failure(s):
[2023-11-06 14:55:39,839] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 30, actual 31
[2023-11-06 14:55:39,839] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 29, actual 31
[2023-11-06 14:55:39,839] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 28, actual 31
[2023-11-06 14:55:39,839] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 27, actual 31
[2023-11-06 14:55:39,839] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 26, actual 31
[2023-11-06 14:55:39,839] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 25, actual 31
[2023-11-06 14:55:39,839] torch._dynamo.guards.__recompiles: [DEBUG] - tensor 'L['x']' size mismatch at index 0. expected 24, actual 31
```
Pull Request resolved: https://github.com/pytorch/pytorch/pull/110325
Approved by: https://github.com/ezyang, https://github.com/jon-chuang
This PR:
- adds the pt2 compliant tag. This tag specifies that the operator works
with the PT2 compilation APIs. A custom op author should test their
ops with opcheck if they choose to add this tag.
- adds a config for Dynamo to allow only pt2 compliant ops into the
graph and graph break on all other OpOverload/OpOverloadPacket.
Bikeshedding help wanted on the name of the tag. It should be easily
grep-able so we can set up rules for it.
Test Plan:
- new tests
Reviewers:
Subscribers:
Tasks:
Tags:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/111933
Approved by: https://github.com/ezyang
ghstack dependencies: #111912, #111915, #111948
`is_allowed` is a tricky bit of functionality - it sits early up in builder and is used to drive the creation of TorchVariable (more notes here, meta only https://fb.workplace.com/groups/pytorch.dev/permalink/1393563781222098/)
If we are tracing distributed in full, we want to route certain calls in distributed to NOT PASS is_allowed (this does not, confusingly, mean that they are not allowed, lol, but rather that we dont want them to become TorchVariable), others, we are fine with preserving.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/110894
Approved by: https://github.com/ezyang
Summary: The runtime assertions inserted in the `torch._export.export` by the `_AddRuntimeAssertionsForInlineConstraintsPass` lead to errors in AOT Inductor like #109884. In `torch._export.aot_compile` export and AOT compilation are run consecutively which would lead to the above issue if any assertions are inserted.
In this PR, we're adding a new parameter / flag to `torch._export.aot_compile`, `remove_runtime_assertions`, to remove the assertions inserted during export before AOT compilation. The flag is set to `False` for BC.
Additionally, we remove the flag `add_runtime_assertions_for_inline_constraints` recently added to `torch._dynamo.config`, as it can lead to undesirable `torch._export` behavior and is 's no longer required for the AOT Inductor testing purposes.
Test Plan: CI
Reviewers:
Subscribers:
Tasks:
Tags:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/110710
Approved by: https://github.com/zhxchen17, https://github.com/chenyang78
Summary: with the grid computed in terms of unbacked `SymInt`s, it can happen that the grid is zero size. This causes CUDA error on `cuLaunchKernel` in the AOT Inductor codegen.
In this PR, when the grid contains unbacked `SymInt`s, a check is added around the `launchKernel` in the AOT Inductor's C++ wrapper codegen to make sure that the grid is not zero-size.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/110312
Approved by: https://github.com/chenyang78
This flag is requested by @Chillee who is seeing recompilations with simple gpt experiments. We are observing recompilations because `_parameters` ordered dict keeps changing from run to run, and its unclear why that is happening.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/110039
Approved by: https://github.com/Chillee
ghstack dependencies: #110023
Pull Request resolved: https://github.com/pytorch/pytorch/pull/105489
NOTE: this PR is tagged "not user facing", because it's not ready to be announced externally yet.
This PR implements torch.compile + selective activation checkpoint (SAC) integration, by using `TagActivationCheckpoint` (same backend as torch.compile + full activation checkpoint integration).
TorchDispatchMode based implementation cannot support including inplace ops in the checkpointed region at the moment (the reason for this needs investigation), and there is also no way to ban them (because TorchDispatchMode now only sees "after-functionalization" ops, so can't detect if an op is in-place). Hence we hide torch.compile + SAC behind a flag (`torch._dynamo.config._experimental_support_context_fn_in_torch_utils_checkpoint`) and will only use it internally for cases that are known to not have in-place ops. This state won't last too long, because in-place op will at least be able to be detected after Brian's mode reordering and related functionalization changes.
So next steps after this PR:
1. Wait for Brian's mode reordering and related functionalization changes to land, and then try to enable the "inplace ops" unit test for torch.compile + selective activation checkpoint (if it doesn't work, investigate why).
2. Unify selective- and full-checkpoint under TorchDispatchMode based implementation.
Differential Revision: D47497145
Pull Request resolved: https://github.com/pytorch/pytorch/pull/105489
Approved by: https://github.com/anijain2305
Summary:
The basic concept behind this diff is to modify Dynamo's tracing behavior when it encounters a KeyedJaggedTensor that is synced (aka has `_length_per_key` and `_offset_per_key` populated). These fields are lists of integers; ordinarily, Dynamo will optimistically try to specialize on integers, however, for KJTs, we know that these integers will definitely vary from run-to-run. Furthermore, ordinarily, we would also specialize these integers if they are 0/1, but we will frequently expect features in KJTs to be 0/1.
The fix is to detect KJTs and treat these integers as *unbacked integers*. This is NOT a universally sound optimization: when treating these integers as unbacked, we never report them as equal to zero or one. In return, we always generate graphs that generalize no matter the length of values on features. This is enough to trace through APS sparse arch, torchrec_dlrm and some small split-cat examples.
The special integer behavior is triggered by a dynamically scoped `force_unspec_int_unbacked_size_like` variable on TracingContext, which we trigger when we wrap a KJT. There probably are other ways to do this, but this was simple and worked.
Test Plan:
```
buck2 test mode/dev-nosan //pytorch/benchmark/fb/test_gpu:run_test_gpu
```
from aakhundov
1. first build feed_lower_benchmark:
```
buck2 build --show-output mode/opt -c python.package_style=inplace -c fbcode.enable_gpu_sections=true -c fbcode.platform=platform010 -c fbcode.split-dwarf=true hpc/new/models/feed/benchmark:feed_lower_benchmark
```
2. then run the lowering of the model with it:
```
TORCHINDUCTOR_MAX_AUTOTUNE=1 TORCHINDUCTOR_UNIQUE_KERNEL_NAMES=1 TORCH_LOGS="output_code,graph_code" TORCH_COMPILE_DEBUG=1 ../buck-out/v2/gen/fbcode/79c6b019ee0f9469/hpc/new/models/feed/benchmark/__feed_lower_benchmark__/feed_lower_benchmark.par --load=manifold://ig_inference_model/tree/user/facebook/fblearner/predictor/960999465/60/gpu_lowering/input.predictor --skip-trt --skip-ait --sync-mode=0 --enable-aot-inductor --lower-presets="ig_stories" --gpu-trace
```
cf https://docs.google.com/document/d/1yD30xYrdmM8r2HTdmXnZTg0-MHVexfVrAa0294m1AUE/edit?pli=1#heading=h.qiv3fp7e6zg0
From torchrec: https://www.internalfb.com/intern/wiki/Torchrec/Development/Testing_production_models/
From ge0405
baseline (without your diff): f477293168
your diff: f477292363
```
buck2 test //caffe2/test/dynamo:test_dynamo_torchrec
buck2 run 'fbcode//mode/opt' fbcode//pytorch/benchmark/fb/test_gpu:run_test_gpu -- 'pytorch.benchmark.fb.test_gpu.test_gpu.TestBenchmarkFbGpu.test_train_blue_reels_vdd_v3_inductor_speedup'
```
Differential Revision: D49236757
Pull Request resolved: https://github.com/pytorch/pytorch/pull/109216
Approved by: https://github.com/voznesenskym
**Motivation:**
We try to make torch.cond use torch.compile automatically so that we could error out when there is side-effects in the branches and correctly handle the closures.
Before this PR, we have a warning if we don't turn on a config raise_on_backend_change (turning it on gives us an error) for the following code:
```python
def foo()
# Inside torch.cond, we'd like to do something like
torch.compile(foo, backend="eager", fullgraph=True)(...)
...
# Users may then call torch.compile somewhere else.
# Dynamo will use the cached code of foo for "eager" backend
# but we expect dynamo to recompile with "inductor" backend.
torch.compile(foo, backend="inductor")(...)
```
This PR adds a BACKEND_MATCH guard. Effectively, it implements a per-backend cache. In the above example, the cached code for "eager" won't work for "inductor" due to guard check failures and the second torch.compile will do a re-compilation. In the future, it might be useful to have something like a configuration guard that guards against dynamo configuration changes across different compiles (e.g. compile a function with fullgraph=False then compile it again with fullgraph=True).
**Implementation:**
1. We add a guarded_backend_cache and check the most_recent_backend against the backend associated with cached code. We also remove the raise_on_backend_change flag.
Note: More lines are printed for debug log due to newly added context manager and guard adds .
**Test Plan:**
Removed original tests that raise on different backend and add a new test to test whether the BACKEND_MATCH guard can guard against backend change.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/107337
Approved by: https://github.com/jansel
Fix several issues with `torch._numpy.random` functions on eager
1. actually return scalars when `size is None`
2. fix dispatch with USE_NUMPY_STREAM
3. make tnp.random functions composable: make numpy functions receive numpy arguments, not `tnp.ndarray`s
4. fix random.shuffle for e.g. lists
The main need for this gymnastics is due to `np.random` functions returning an ndarray or python scalar depending on the `size` argument. We decided a while ago to replicate this behavior in `tnp.random` and not elsewhere where we always return 0D arrays instead.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/108944
Approved by: https://github.com/lezcano
This PR introduces record and replay functionality for `ShapeEnv` instances. In short,
throughout the execution of a program, we record events (e.g. function calls that modify
its state) so that, in the future, we are able to reproduce any intermediary state of the
instance.
In summary, this PR introduces the following changes (they mostly belong to
_symbolic_shapes.py_ unless otherwise stated):
- Create `ShapeEnvEvent` class for recording function calls + arguments
- Create `record_shapeenv_event` decorator and decorate every function that changes the
state of a `ShapeEnv`: it creates an appropriate event and add it to the available
ShapeEnv instance (sometimes it has to extract from `SymTypes`).
- Create `SymNode.with_shape_env` convenient function for replacing `ShapeEnv` references
- Wraps `ShapeEnv` initialization method: so that we also save the exact way a `ShapeEnv`
was constructed, i.e. arguments
- Introduces a way to compare two `ShapeEnv` instances, defining a concept of state for
that class. In short, the state of `ShapeEnv` is every variable that may change the
execution flow
- Create `check_shape_env_recorded_events` dynamo configuration for enabling the check for
equality the state of `ShapeEnv` with another one that was constructed by replaying all
the recorded events. This check takes place inside `produce_guards`
- Create `replay_shape_env_events` function for replaying given events. It assumes the
first event is `ShapeEnv` initialization function
Pull Request resolved: https://github.com/pytorch/pytorch/pull/107989
Approved by: https://github.com/ezyang
**Motivation:**
We try to make torch.cond use torch.compile automatically so that we could error out when there is side-effects in the branches and correctly handle the closures.
Before this PR, we have a warning if we don't turn on a config raise_on_backend_change (turning it on gives us an error) for the following code:
```python
def foo()
# Inside torch.cond, we'd like to do something like
torch.compile(foo, backend="eager", fullgraph=True)(...)
...
# Users may then call torch.compile somewhere else.
# Dynamo will use the cached code of foo for "eager" backend
# but we expect dynamo to recompile with "inductor" backend.
torch.compile(foo, backend="inductor")(...)
```
This PR adds a BACKEND_MATCH guard. Effectively, it implements a per-backend cache. In the above example, the cached code for "eager" won't work for "inductor" due to guard check failures and the second torch.compile will do a re-compilation. In the future, it might be useful to have something like a configuration guard that guards against dynamo configuration changes across different compiles (e.g. compile a function with fullgraph=False then compile it again with fullgraph=True).
**Implementation:**
1. We add a guarded_backend_cache and check the most_recent_backend against the backend associated with cached code. We also remove the raise_on_backend_change flag.
2. Then newly added context manager and guard adds more lines for debug log so we change the uppper limit from 50 to 55.
**Test Plan:**
Removed original tests that raise on different backend and add a new test to test whether the BACKEND_MATCH guard can guard against backend change.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/107337
Approved by: https://github.com/jansel
**This PR is a 99% copy paste of Sam Gross** (@colesbury) work at https://github.com/pytorch/pytorch/pull/100642. Copied from there
--------
The NN_MODULE guard now subsumes guards on Module attributes. The check_fn will fail if the module attributes are changed (such as Module.training), parameters, submodules, and buffers are added or removed, and if fields are changed on the type itself.
This gives up specificity in the guard check -- if any field is changed the check_fn fails -- for faster overall checks.
-----
Pull Request resolved: https://github.com/pytorch/pytorch/pull/108528
Approved by: https://github.com/ezyang
This PR wraps `InstructionTranslator` run with a try-catch block so as to run the
translation validation (TV) if it ends up raising an error.
In this context, we run TV so as to catch simplification errors. These may turn
`ShapeEnv.divisible` and `ShapeEnv.replacements` incorrect.
For example: #101173 describes a SymPy simplification bug that doesn't reach TV, since
it's run only in the end of the tracing.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/106645
Approved by: https://github.com/ezyang
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 a new configuration that enables shapes of torch.nn.Parameter to be treated as dynamic in order to avoid extensive recompilation when Paramters are used instead of Tensor.
This features addresses part of issue #105279
Pull Request resolved: https://github.com/pytorch/pytorch/pull/105855
Approved by: https://github.com/ezyang
D47969512 was the original diff to revert this, but the diff train doesn't work well, so I have to split it into two part: this OSS PR and another separate diff to revert the fbcode change.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/106562
Approved by: https://github.com/angelayi
Summary:
We are working toward full model compilation, where when compilation error happens, we just fall back to eager mode rather than error out.
But at the same time, we should fix these issues if they are bugs. We will:
* 1/ log warnings in OSS;
* 2/ log warnings and write them into Scuba in fbcode;
to prevent us from ignoring these issues.
Test Plan: Manual test
Differential Revision: D47506314
Pull Request resolved: https://github.com/pytorch/pytorch/pull/105307
Approved by: https://github.com/jansel
Fixes: #105143
In summary, the changes are:
- Check if Z3 is installed when the module is loaded
- Naming consistently as "translation validation" (not "validator")
- Skipping tests if Z3 is not installed
Pull Request resolved: https://github.com/pytorch/pytorch/pull/105168
Approved by: https://github.com/ezyang
As of now, translation validation runs to its completion. However, Z3 is time
consuming. PR #104464, for example, disables translation validation for a few benchmarks.
Instead, this PR introduces a timeout for translation validation. In that case, Z3 will
return `unknown`, since it wasn't able to prove or disprove the assertions. Then, we log
it as a warning, but don't stop execution.
Here's a summary of the changes:
- Added an environment variable for turning translation validation on and off
- Added an environment variable for setting the translation validation timeout
- Possibly reverts the changes in #104464
- ~~Move from "QF_NRA" to "QF_NIRA" logic~~
- ~~It makes more sense, given the nature of the problems~~
- "QF_NRA" seems to solve more instances of _dynamo/test_dynamic_shapes.py_
Pull Request resolved: https://github.com/pytorch/pytorch/pull/104654
Approved by: https://github.com/ezyang
Some notes:
* I now manually turn off `_generate` jobs from running with cudagraphs, as it is unrealistic to expect to cudagraph autoregressive generation up to max sequence length, this would imply compiling the entire unrolled sequence generation. Concretely, cm3leon_generate was timing out post this change, likely due to the compile time slowdown of dynamic shapes ON TOP OF accidentally unrolling all the loops
* A few torch._dynamo.reset tactically inserted to force recompiles on tests that expected it
* expectedFailureAutomaticDynamic flip into patching automatic_dynamic_shapes=False
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/103623
Approved by: https://github.com/voznesenskym
This PR introduces a translation validator for dynamo guards. In summary, it verifies
whether the guards issued as Python code are sound, w.r.t the initial guards.
The main changes in this PR are:
- Create an FX graph for dynamic shapes
- Translate "the original" guards from the FX graph to Z3
- Check if the guards produced by `produce_guards` are sound w.r.t. the ones from the FX graph
gh-stack version of the PR #101146.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/102563
Approved by: https://github.com/ezyang
Summary:
Replace _dynamo.config with an object instead of module
Current usage patterns of setting and reading fields on config will work
unchanged.
Only changes needed going forward:
1. import torch._dynamo.config will not work. However, just doing
import torch._dynamo is sufficient to access dynamo config
as torch._dynamo.config.
2. Files inside of _dynamo folder need to access config via
from torch._dynamo.config_util import config instead of
from torch._dynamo import config. Because _dynamo/__init__.py
imports some of the files so it would be circular import.
Test Plan:
Reviewers:
Subscribers:
Tasks:
Tags:
Fixes #ISSUE_NUMBER
Pull Request resolved: https://github.com/pytorch/pytorch/pull/96455
Approved by: https://github.com/jansel
This PR adds support for tracing autograd.Function with grad.
A few important bullet points outlining our approach:
1) Our goal is to verify soundness in order to add a call_function to the autograd.Function's `apply` to the graph.
2) We achieve (1) by either verifying soundness or rejecting soundness, by ensuring that both forward and backward of the autograd.Function are sound.
3) For the forward, if we verify soundness, we install its guards into the graph.
4) For the backward, if we verify soundness, we throw it out. However, backwards soundness verification is more onerous, and has a config driven set of banned attrs and methods for tensors.
1-4 above are achieved by turning the forward and backward into UserDefinedFunctionVariables, and inlining through them, relying on dynamo's soundness detection. If we graph break in these, we raise and treat them as unsound. As noted above, backwards is stricter yet.
For the tracing, the safety comes from dynamo's HigherOrderOperator system. That system ensures that not only do we trace soundly, but that no new variables are lifted into inputs during the tracing, and that the forward and backwards are entirely self contained.
Whenever we reject a function as unsound, we restore back, as usual.
Due to some limitations in the lifting logic, we have an escape hatch we implemented for tensors that are known in forward, but cross into backwards through save_tensors (save) /saved_tensors (load). We escape hatch here to avoid having the known saved tensors coming from forward end up being accidentally treated as lifted variables (and rejected). This is sound, but a little hacky feeling.
Additionally, due to some limitations in fx node removal, combined with how we produce subgraphs for the traces installed from HigherOrderOperators, we had to improve our node removal logic. In the event of a restore, we remove the old nodes from the graph, as usual in dynamo. However, because the references to these nodes may exist in subgraphs, we traverse any nodes users and remove them first if and only if they are in another graph. This is always sound, because removal should only be downstream of restoration at this point.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/99483
Approved by: https://github.com/zou3519
The changes:
* Add config knob `same_two_models_use_fp64` for toggling whether or not to use fp64
* Add a test showing that RMSE is superior to atol/rtol
* Add `--strict-accuracy` options, which allows for testing against integral/boolean accuracy. Regular accuracy by default now ONLY. There's a test which exercises this, it's a little delicate but I had trouble thinking of a good test otherwise.
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/100447
Approved by: https://github.com/voznesenskym
