1) Using items stored in torch._tensor_classes to check item passed from python side;
2) Add SparsePrivateUse1 in backend_to_string, layout_from_backend and check_base_legacy_new;
3) Using more general API to get python module name in get_storage_obj and get_name functions.
Fixes #ISSUE_NUMBER
Pull Request resolved: https://github.com/pytorch/pytorch/pull/119263
Approved by: https://github.com/ezyang
I was just playing around with improving the typing of symbolic_shapes. The PR is not "complete" but I in particular wanted to get feedback on whether or not people liked making ValueRanges Generic; it seems that distinguishing if you have an Expr ValueRange or a SympyBoolean ValueRange is a lot of trouble for downstream. Using TypeGuard, we can perform refinements on the generic parameter inside methods, although we still have to cast back to ValueRange[T] due to https://github.com/python/mypy/issues/14425#issuecomment-1914852707
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/118529
Approved by: https://github.com/Skylion007
# Motivation
According to [[1/4] Intel GPU Runtime Upstreaming for Device](https://github.com/pytorch/pytorch/pull/116019), As mentioned in [[RFC] Intel GPU Runtime Upstreaming](https://github.com/pytorch/pytorch/issues/114842), this third PR covers the changes under `libtorch_python`.
# Design
This PR primarily offers device-related APIs in python frontend, including
- `torch.xpu.is_available`
- `torch.xpu.device_count`
- `torch.xpu.current_device`
- `torch.xpu.set_device`
- `torch.xpu.device`
- `torch.xpu.device_of`
- `torch.xpu.get_device_name`
- `torch.xpu.get_device_capability`
- `torch.xpu.get_device_properties`
- ====================
- `torch.xpu._DeviceGuard`
- `torch.xpu._is_compiled`
- `torch.xpu._get_device`
# Additional Context
We will implement the support of lazy initialization in the next PR.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/116850
Approved by: https://github.com/EikanWang, https://github.com/jgong5, https://github.com/gujinghui, https://github.com/malfet
I was just playing around with improving the typing of symbolic_shapes. The PR is not "complete" but I in particular wanted to get feedback on whether or not people liked making ValueRanges Generic; it seems that distinguishing if you have an Expr ValueRange or a SympyBoolean ValueRange is a lot of trouble for downstream. Using TypeGuard, we can perform refinements on the generic parameter inside methods, although we still have to cast back to ValueRange[T] due to https://github.com/python/mypy/issues/14425#issuecomment-1914852707
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/118529
Approved by: https://github.com/Skylion007
Fixes https://github.com/pytorch/pytorch/issues/118129
Suppressions automatically added with
```
import re
with open("error_file.txt", "r") as f:
errors = f.readlines()
error_lines = {}
for error in errors:
match = re.match(r"(.*):(\d+):\d+: error:.*\[(.*)\]", error)
if match:
file_path, line_number, error_type = match.groups()
if file_path not in error_lines:
error_lines[file_path] = {}
error_lines[file_path][int(line_number)] = error_type
for file_path, lines in error_lines.items():
with open(file_path, "r") as f:
code = f.readlines()
for line_number, error_type in sorted(lines.items(), key=lambda x: x[0], reverse=True):
code[line_number - 1] = code[line_number - 1].rstrip() + f" # type: ignore[{error_type}]\n"
with open(file_path, "w") as f:
f.writelines(code)
```
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Co-authored-by: Catherine Lee <csl@fb.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/118533
Approved by: https://github.com/Skylion007, https://github.com/zou3519
Fixes https://github.com/pytorch/pytorch/issues/118129
Suppressions automatically added with
```
import re
with open("error_file.txt", "r") as f:
errors = f.readlines()
error_lines = {}
for error in errors:
match = re.match(r"(.*):(\d+):\d+: error:.*\[(.*)\]", error)
if match:
file_path, line_number, error_type = match.groups()
if file_path not in error_lines:
error_lines[file_path] = {}
error_lines[file_path][int(line_number)] = error_type
for file_path, lines in error_lines.items():
with open(file_path, "r") as f:
code = f.readlines()
for line_number, error_type in sorted(lines.items(), key=lambda x: x[0], reverse=True):
code[line_number - 1] = code[line_number - 1].rstrip() + f" # type: ignore[{error_type}]\n"
with open(file_path, "w") as f:
f.writelines(code)
```
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/118533
Approved by: https://github.com/Skylion007, https://github.com/zou3519
Updates flake8 to v6.1.0 and fixes a few lints using sed and some ruff tooling.
- Replace `assert(0)` with `raise AssertionError()`
- Remove extraneous parenthesis i.e.
- `assert(a == b)` -> `assert a == b`
- `if(x > y or y < z):`->`if x > y or y < z:`
- And `return('...')` -> `return '...'`
Co-authored-by: Nikita Shulga <2453524+malfet@users.noreply.github.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/116591
Approved by: https://github.com/albanD, https://github.com/malfet
To codegen deferred runtime asserts, I need to be able to convert sympy expressions back into regular Python expressions that I can put in FX graphs. This PR adds some of the machinery to do this: it adds a new sympy analysis that runs operations on all FX traceable operations that can also be run with plain Python int/float/bool/etc. It's tested by symbolic tracing through the analysis, and then testing that this traced graph gives the same result as running the Python analysis directly.
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/113978
Approved by: https://github.com/aakhundov, https://github.com/lezcano
This PR also contains a basket of fixes that were turned up by now testing more arguments with SymInt. I fixed as many of the easy ones as I could easily get earlier in this stack and a bunch here, but there are some more annoying ones I xfailed.
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/113452
Approved by: https://github.com/Chillee
ghstack dependencies: #113877, #113911
`@classproperty` decorator adds another wrapper, so warning with default stacklevel (2) would always point to the wrapper implementation rather than at callee.
For example, before this change following code
```python
import torch
print(torch.FloatStorage.dtype)
```
will produce inactionable warning:
```
/Users/nshulga/git/pytorch/pytorch/torch/_utils.py:836: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()
return self.fget.__get__(instance, owner)()
```
But after the change warning turns into:
```
/Users/nshulga/test/bar.py:2: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()
print(torch.FloatStorage.dtype)
```
Discovered while reading https://github.com/pytorch/pytorch/issues/109108
Pull Request resolved: https://github.com/pytorch/pytorch/pull/113601
Approved by: https://github.com/kit1980
`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
Summary:
CUDA Graph does not work well with CUPTI teardown.
1) crashes on 1st lazy CUPTI re-init after teardown (CUDA 11)
2) crashes on 2nd non-lazy CUPTI re-init after teardown (CUDA 12)
Workaround: turn off CUPTI teardown when using CUDA Graphs completely.
Test Plan: CI
Differential Revision: D50811284
Pulled By: aaronenyeshi
Pull Request resolved: https://github.com/pytorch/pytorch/pull/112507
Approved by: https://github.com/davidberard98
This PR:
- Moves TrueDiv, LShift, RShift, IsNonOverlappingAndDenseIndicator to `_sympy.functions.py`
- Moves SymNode to `fx.experimental.sym_node`.
- This file does not have any SymPy dependencies at import time
- It installs the magic methods in Sym{Bool,Int,Float}.
- N.b. With this split, we may be able to move Sym{Bool,Int,Float} to this file, and remove quite a few of the hacks around these classes
- Imports `sym_node` in `torch/__init__.py` rather than the whole `symbolic_shapes.py`.
This breaks the import-time dependency between torch and SymPy
Pull Request resolved: https://github.com/pytorch/pytorch/pull/112037
Approved by: https://github.com/peterbell10
ghstack dependencies: #112035, #112036
This PR supports sym_ite. This is useful for converting SymBool to SymInt in e.g. #109916. Internally, it uses sympy.Piecewise. We cannot use sympy.ITE because it expects the arguments and output all to be boolean type but we want return SymInt type when converting a SymBool to SymInt. So we use sympy.Piecewise to denote the symbolic relationship.
Note that this pr uses the range analysis for sympy.Piecewise implemented in https://github.com/pytorch/pytorch/blob/main/torch/utils/_sympy/value_ranges.py.
Test Plan:
See added test.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/111440
Approved by: https://github.com/ezyang
This pr expose torch._higher_order_ops.cond as torch.cond.
1. Need to add #noqa: F811 to the _check calls in torch/__init__.py to address some confusing linter error "Redefinition of unused 'cond'" but only one cond is imported and for these lines that have this error, they don't define the cond but just use it as an argument.
2. Also add cond to the list that allows it to be traced through so as dynamo could trigger the CondHigherOrder logic instead of creating a TorchVariable.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/110293
Approved by: https://github.com/zou3519
This pr expose torch._higher_order_ops.cond as torch.cond.
1. Need to add #noqa: F811 to the _check calls in torch/__init__.py to address some confusing linter error "Redefinition of unused 'cond'" but only one cond is imported and for these lines that have this error, they don't define the cond but just use it as an argument.
2. Also add cond to the list that allows it to be traced through so as dynamo could trigger the CondHigherOrder logic instead of creating a TorchVariable.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/110293
Approved by: https://github.com/zou3519
Changelog:
- torch.library.impl_abstract optionally accepts a torch.library.Library
object. If passed in, then the lifetime of the registration is tied to
the Library object.
- we've also changed torch.library.impl_abstract to work on all
operators, including overloads.
- we refactored the `torch._custom_ops.*` and `torch._custom_op.*`
impl_abstract APIs and put them under torch._library. This is the
final resting place for them. I will follow-up with deleting
all the `torch._custom_ops.*` stuff later.
- There is a new "SimpleOperatorRegistry" where we actually collect the
abstract_impl. We will expand this to also hold the other
torch._custom_ops.* APIs when we move those to torch.library
NB: Previously we had designed
`impl_abstract` assuming a very high-level Python-only custom op API.
We've revisited that since; now, impl_abstract works for all custom ops,
no matter python or C++, no matter the schema. The new refactored design
reflects this better.
Test Plan:
- existing and new tests
Pull Request resolved: https://github.com/pytorch/pytorch/pull/109912
Approved by: https://github.com/ezyang
Bugfix:
- previously, SymBool does not implement `__eq__`, Python falls back to default `__eq__ `and `__hash__`
- in this PR, we make SymBool implement `__eq__`
- symbolic SymBool now raises an error when hashed just like SymInt/SymFloat
New feature:
- previously, SymInt and SymFloat are unhashable (even if you are singleton or constant)
- in this PR, SymInt and SymBool are hashable if singleton/constant
Stay the same:
- SymNode are hashable due to default Python behavior
Pull Request resolved: https://github.com/pytorch/pytorch/pull/109170
Approved by: https://github.com/ezyang
ghstack dependencies: #109169
In this PR:
- When Constant SymNode are detected in unary/binary ops demote them to plain int/bool before proceeding. Sometimes this means doing a unary op with a Constant SymNode would result in a plain bool.
- Introduce an is_symbolic method, only available from Python. We need this because isinstance(x, SymInt) is no longer sufficient to check whether a given int/SymInt is symbolic or not. See later PR in the stack to see how this is used.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/109169
Approved by: https://github.com/ezyang
Reland - the previous PR was reverted by internal with this error:
```
File "/data/sandcastle/boxes/eden-trunk-hg-fbcode-fbsource/buck-out/v2/gen/fbcode/363cd7e240f5d021/caffe2/torch/fb/trainer/data_modules/tests/__test_dataloader__/test_dataloader#link-tree/torch/__init__.py", line 29, in <module>
from ._utils_internal import _functionalize_sync as _sync
ImportError: cannot import name '_functionalize_sync' from 'torch._utils_internal'
```
I couldn't figure out why internal was unhappy with the import. One potential reason is that I see a build rule for *another* `_utils_internal.py` in the fb folder here ([link](https://www.internalfb.com/code/fbsource/[30ed85cd88409af98b7490be137aaa5dfd7afd01]/fbcode/caffe2/TARGETS?lines=444))
Rather than burn more time investigating, I confirmed internally that the error goes away if I move the util from `torch/_utils_internal.py` to `torch/_utils.py`.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/109518
Approved by: https://github.com/albanD
Added two new utils to help with turning python functionalization on in AOTAutograd (next PR):
(1) updated `torch._sync()`. Previously, this API could only handle `torch.Tensor` instances that had a `FunctionalTensorWrapper` TensorImpl. It now needs to handle python `FunctionalTensor`'s. In theory I can probably break BC and change this API (since it's private?), but I decided not to do it in this PR stack do minimize the chance of reverts. Instead of updating that API directly (which is in C++), I just added a python shim that first tries to unwrap the python `FunctionalTensor` if there is one, then calls the existing C++ logic
(2) `mirror_autograd_meta` is now a standalone API that tries to mirror the `requires_grad` and `is_leaf` autograd metadata from one tensor to another. Previously this was hardcoded into `torch._to_functional_tensor()`. But I now need to use it in a more standalone way: later in AOTAutograd when we unwrap and re-wrap a tensor subclasses, we need to manually mirror the autograd metadata from the original to the updated version of the subclass.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/107917
Approved by: https://github.com/ezyang
ghstack dependencies: #106404
Or any other distro that have different purelib and platlib paths Regression was introduced, when small wheel base dependency was migrated from CUDA-11 to CUDA-12
Not sure why, but minor version of the package is no longer shipped with following CUDA-12:
- nvidia_cuda_nvrtc_cu12-12.1.105
- nvidia-cuda-cupti-cu12-12.1.105
- nvidia-cuda-cupti-cu12-12.1.105
But those were present in CUDA-11 release, i.e:
``` shell
bash-5.2# curl -OL 922c5996aa/nvidia_cuda_nvrtc_cu11-11.7.99-2-py3-none-manylinux1_x86_64.whl; unzip -t nvidia_cuda_nvrtc_cu11-11.7.99-2-py3-none-manylinux1_x86_64.whl |grep \.so
testing: nvidia/cuda_nvrtc/lib/libnvrtc-builtins.so.11.7 OK
testing: nvidia/cuda_nvrtc/lib/libnvrtc.so.11.2 OK
bash-5.2# curl -OL c64c03f49d/nvidia_cuda_nvrtc_cu12-12.1.105-py3-none-manylinux1_x86_64.whl; unzip -t nvidia_cuda_nvrtc_cu12-12.1.105-py3-none-manylinux1_x86_64.whl|grep \.so
testing: nvidia/cuda_nvrtc/lib/libnvrtc-builtins.so.12.1 OK
testing: nvidia/cuda_nvrtc/lib/libnvrtc.so.12 OK
```
Fixes https://github.com/pytorch/pytorch/issues/109221
Pull Request resolved: https://github.com/pytorch/pytorch/pull/109244
Approved by: https://github.com/huydhn
This updates ruff to 0.285 which is faster, better, and have fixes a bunch of false negatives with regards to fstrings.
I also enabled RUF017 which looks for accidental quadratic list summation. Luckily, seems like there are no instances of it in our codebase, so enabling it so that it stays like that. :)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/107519
Approved by: https://github.com/ezyang
This updates ruff to 0.285 which is faster, better, and have fixes a bunch of false negatives with regards to fstrings.
I also enabled RUF017 which looks for accidental quadratic list summation. Luckily, seems like there are no instances of it in our codebase, so enabling it so that it stays like that. :)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/107519
Approved by: https://github.com/ezyang
The documentation for `torch.set_float32_matmul_precision()` mentions a datatype called "bfloat16_3x". This doesn't appear to be a very standard term, and I had a hard time figuring out what exactly it meant. I now assume it refers to [[Henry2019]](http://arxiv.org/abs/1904.06376), which describes an algorithm by which a float32 multiplication is approximated via three bfloat16 multiplications. This PR updates the documentation to include this reference and to briefly describe how this algorithm works.
Note that I just learned everything that I wrote here, so I'd appreciate if someone more expert in this topic could check to make sure that I didn't get anything significantly wrong.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/107169
Approved by: https://github.com/colesbury
This pattern shows up in torchrec KeyedJaggedTensor. Most
of the change in this PR is mechanical: whenever we failed
an unbacked symint test due to just error checking, replace the
conditional with something that calls expect_true (e.g.,
torch._check or TORCH_SYM_CHECK).
Some of the changes are a bit more nuanced, I've commented on the PR
accordingly.
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/106788
Approved by: https://github.com/lezcano
ghstack dependencies: #106720
issues resolved: https://github.com/pytorch/pytorch/issues/101832
**context**: get torch.compile config for further usage. E.g, the training platform wants to get if model is compiled with cudagraph enabled and trigger further action
**how it is implemented**
* the core logic is backend.get_compiler_config() in torch/_dynamo/eval_frame.py
* for backend='inductor' / _TorchCompileInductorWrapper, we have inductor-specific implementation in get_compiler_config in torch/_inductor/compile_fx.py and torch/__init__.py
**how to use it**: Below is an example.
```
model = DummyModule()
optimized_module = torch.compile(
model, options={"triton.cudagraphs": True}
)
compiler_config = optimized_module.get_compiler_config()
if compiler_config["triton.cudagraphs"]:
pass
```
Pull Request resolved: https://github.com/pytorch/pytorch/pull/105026
Approved by: https://github.com/yanboliang, https://github.com/jansel
This PR re-lands
- [Typing] Fix PEP 484 Violation (#105022)
- Update mypy to 1.4.1 (#91983)
That were reverted due to the conflict with internal source repo.
Mostly fixes for PEP-484 violation (i.e. when default arg is set to None, but type is not annotated as optional)
Plus few real fixes:
- Add missing `_get_upgraders_entry_map` to `torch/_C/__init__.pyi`
- Add missing return statement to `torch._export. deserialize_graph`
- Fix error message in `torch.ao.ns.fx.weight_utils.get_lstm_mod_weights`
- Add assert it `torch/optim/optimizer.py` that Optional list is not None
TODO (in followup PR):
- Fix erroneous `isinstance` check in `torch/ao/quantization/_pt2e/qat_utils.py`
Unrelated, to bypass CI failures due to the gcc9 dependency update in Ubuntu-18.04:
- Add hack to squash older libstdc++ from conda environment in favor one from OS to `.ci/docker/install_conda.sh`
- Update bazel cuda builds to focal, as with libstdc++-6.0.32 bazel builds loose the ability to catch exceptions (probably because they link with cupti statically, but I could not found where it is done)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/105227
Approved by: https://github.com/atalman, https://github.com/albanD, https://github.com/Skylion007
This PR re-lands
- [Typing] Fix PEP 484 Violation (#105022)
- Update mypy to 1.4.1 (#91983)
That were reverted due to the conflict with internal source repo.
Mostly fixes for PEP-484 violation (i.e. when default arg is set to None, but type is not annotated as optional)
Plus few real fixes:
- Add missing `_get_upgraders_entry_map` to `torch/_C/__init__.pyi`
- Add missing return statement to `torch._export. deserialize_graph`
- Fix error message in `torch.ao.ns.fx.weight_utils.get_lstm_mod_weights`
- Add assert it `torch/optim/optimizer.py` that Optional list is not None
TODO (in followup PR):
- Fix erroneous `isinstance` check in `torch/ao/quantization/_pt2e/qat_utils.py`
Pull Request resolved: https://github.com/pytorch/pytorch/pull/105227
Approved by: https://github.com/atalman, https://github.com/albanD, https://github.com/Skylion007
The general idea is to do a separate CUDA graph for each size. Because of cuda graph trees, these graphs will all share the same memory pool, so your memory usage will only be the worst case memory usage of the biggest dynamic size you want. This requires an extra dispatch in the cudagraphified callable. You must pay for a CUDA graph recording for every dynamic size you encounter, but this is MUCH cheaper than running the entire PT2 compile stack, so I expect you to still see benefits.
This was surprisingly easy to do.
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/105064
Approved by: https://github.com/voznesenskym
This reduces total number of imported modules by default from 1419 to 1322 according to
```
time python -c "import sys;before=len(sys.modules);import torch;after=len(sys.modules);print(f'torch-{torch.__version__} imported {after-before} modules')"
```
and slightly reduces import time, while having no effect on UX (i.e. `torch.onnx.` submodule is kept intact)
Suppress lint errors that appear after mypy accidentally starts listing more files, for more details see: https://github.com/pytorch/pytorch/issues/104940
Pull Request resolved: https://github.com/pytorch/pytorch/pull/104843
Approved by: https://github.com/jansel, https://github.com/albanD
New elements added to a tensor by `torch.Tensor.resize_` are set to NaN/MAX_INT when deterministic mode is turned on.
When `torch.Tensor.resize_` is called on a quantized tensor and deterministic mode is turned on, a nondeterministic error is raised.
Part of #82004
Pull Request resolved: https://github.com/pytorch/pytorch/pull/104300
Approved by: https://github.com/albanD
Summary:
This diff is reverting D46920584
D46920584: Make `torch.empty*` deterministic by filling with NaN or max int value (#101849) by generatedunixname499836121 has been identified to be causing the following test or build failures:
Tests affected:
- [torchrec/distributed/composable/tests:test_fsdp - torchrec.distributed.composable.tests.test_fsdp.FullyShardTest: test_composable_checkpoint](https://www.internalfb.com/intern/test/281475062923125/)
Here's the Multisect link:
https://www.internalfb.com/multisect/2341386
Here are the tasks that are relevant to this breakage:
We're generating a revert to back out the changes in this diff, please note the backout may land if someone accepts it.
If you believe this diff has been generated in error you may Commandeer and Abandon it.
Test Plan: NA
Reviewed By: huydhn, osalpekar
Differential Revision: D46997394
Pull Request resolved: https://github.com/pytorch/pytorch/pull/104302
Approved by: https://github.com/osalpekar
Use [PEP-562](https://peps.python.org/pep-0562) to import `_dynamo` and `_inductor` only when needed.
- Remove redundant imports from tests
- Add `test_lazy_imports_are_lazy` to make sure they will not get imported by accident
<!--
copilot:poem
-->
### <samp>🤖 Generated by Copilot at bae8e90</samp>
> _Sing, O Muse, of the daring deeds of PyTorch, the swift and fiery_
> _framework of deep learning, that with skill and cunning wrought_
> _many wonders of dynamic compilation, using the hidden powers_
> _of `_dynamo` and `_inductor`, the secret modules of LLVM and MLIR._
Pull Request resolved: https://github.com/pytorch/pytorch/pull/104368
Approved by: https://github.com/msaroufim, https://github.com/albanD
# torch.compiler public API
## Goal
The goal of this document is to describe the public facing API for torchdynamo and torchinductor.
Today both dynamo and torchinductor are in `torch/_dynamo` and `torch/_inductor` namespace with the only public function
`torch.compile()` which is directly placed in `torch/__init__.py`
This poses a few problems for users trying to take dependencies on PyTorch 2.0
1. Unclear BC guarantees
2. No builtin discovery mechanism outside of reading the source code
3. No hard requirements for docstrings or type annotations
Most importantly it mixes two personas the PyTorch 2.0 developer vs the PyTorch 2.0 customer so this is an attempt to address this. We draw a lot of inspiration from the `functorch` migration to the `func` namespace.
## Alternate names
We did discuss some other alternative names
1. `torch.compile` -> problem is this would break BC on the existing `torch.compile` function
2. `torch.dynamo` -> `dynamo` is so far not something we've deliberately hidden from users but problem is now figuring out what it's `_dynamo` vs `dynamo` might be confusing
3. `torch.compiler` -> 1 would be better but to keep BC this is a good compromise
# The general approach
## Proposal 1
In https://github.com/pytorch/pytorch/blob/main/torch/_dynamo/__init__.py
We have function called `reset()`, this function is essential if users are trying to `torch.compile()` a model under different settings
```python
# in _dynamo/
def reset():
do_reset_stuff()
```
Instead we propose
```python
# in compiler/
def reset():
do_reset_stuff() # As in copy paste the logic from _dynamo.reset
# in _dynamo/
import warnings
import inspect
def reset():
function_name = inspect.currentframe().f_code.co_name
warnings.warn(f"{function_name} is deprecated, use compiler.{function_name} instead", DeprecationWarning)
return compiler.reset()
```
## Proposal 2
```python
# in compiler/
def reset():
“””
Docstrings here
“””
_dynamo.reset()
# in _dynamo/
No changes
```
Consensus so far seems to be proposal 2 since fewer warnings will be less jarring and it’ll make it quite easy to merge the public API
## Docstrings
The above was an example of a function that has no inputs or outputs but there are other functions which could use an improvement in their docstrings, for example allow_in_graph actually works over lists of functions but that’s not mentioned anywhere in the example only if you read the source code.
def allow_in_graph(fn):
"""
Customize which functions TorchDynamo will include in the generated
graph. Similar to `torch.fx.wrap()`.
Parameters:
fn (callable or list/tuple): The function(s) to be allowed in the graph.
Returns:
callable or list/tuple: The input function(s) included in the graph.
Examples:
Customize inclusion of a single function:
::
torch._dynamo.allow_in_graph(my_custom_function)
Customize inclusion of multiple functions:
::
torch._dynamo.allow_in_graph([my_custom_function1, my_custom_function2])
@torch._dynamo.optimize(...)
def fn(a):
x = torch.add(x, 1)
x = my_custom_function(x)
x = torch.add(x, 1)
return x
fn(...)
Notes:
The `allow_in_graph` function allows customization of which functions TorchDynamo
includes in the generated graph. It can be used to include specific functions that
are not automatically captured by TorchDynamo.
If `fn` is a list or tuple, `allow_in_graph` will be called recursively on each
element in the sequence.
Once a function is allowed in the graph using `allow_in_graph`, it will be captured
in the graph generated by TorchDynamo. This customization enables more fine-grained
control over the functions included in the graph.
Note that `allow_in_graph` expects the input `fn` to be a callable.
"""
if isinstance(fn, (list, tuple)):
return [allow_in_graph(x) for x in fn]
assert callable(fn), "allow_in_graph expects a callable"
allowed_functions._allowed_function_ids.add(id(fn))
allowed_functions._disallowed_function_ids.remove(id(fn))
return fn
So to make the API public, we’d have to write similar docstrings for all public functions we’d like to create.
The benefit of this approach is that
1. No BC risks, internal and external users relying on our tooling can slowly wean off the private functions.
2. We will also have to write correct docstrings which will automatically make our documentation easier to maintain and render correctly on pytorch.org
3. We already have some BC guarantees already, we don’t kill OptimizedModule, we rejected the PR to change the config system
The con of this approach is that
Will be stuck with some potentially suboptimal functions/classes that you can’t kill
## Testing strategy
If the approach is to mostly make a public function call an already tested private function then all we need to do is ensure that the function signatures don't change
## Which functions should be in the public API
Our heuristic for deciding whether something should be public or not is are users already relying on it for lack of other options or have we recommended some non public functions for users to debug their PT 2.0 programs.
Heuristic for not putting something in public is that it’s an experimental subsystem with the goal of turning it on by default, it’s very core dev centric, meta centric, a bunch of different configs that should be batched into a single user facing one, or something that needs to be renamed because the name is confusing
#### Top level
`torch.compile()` -> already is a public API it does require some minor improvements like having configs be passed in to any backend and not just inductor (EDIT: This was already done https://github.com/pytorch/pytorch/pull/99645l) and renaming `mode=reduce-overhead` to `mode=cudagraph`
To make sure that PT 2.0 is supported with a given pytorch version users can create a new public function and this would replace the need for `try/except` blocks around `import torch._dynamo` that has been populating user code.
```python
def pt2_enabled():
if hasattr(torch, 'compile'):
return True
else:
return False
```
For all of the below they will be translated to `torch.compiler.function_name()`
#### From _dynamo
As a starting point we looked at https://github.com/pytorch/pytorch/blob/main/torch/_dynamo/__init__.py and we suggest redefining these functions in `pytorch/torch/compiler/__init__.py`
It might also make sense to split them over multiple files and import them in `__init__.py` but because the number of functions is small it'd probably be fine to add them all into a single compiler/__init__.py until this list becomes larger
1. `reset()`
2. `allow_in_graph()`
10. `list_backends()`
12. `compile()`: torch.compile() would be mostly a shell function passing arguments to torch.compiler.compile()
13. `assume_constant_result()`: TODO: Double check how this is useful
15. `torch._dynamo.disable()`
Some notable omissions
11. `explain()`: We need to clean up the output for this function, make it a data class and pretty printable
1. `forbid_in_graph()`: Considered adding this but should instead consolidate on `disallow_in_graph`
2. `optimize_assert()`: Already covered by `torch.compile(fullgraph=True)`
3. `check_if_dynamo_supported()`: this would be supplanted by pt2_enabled()
4. `compilation_metrics`, `graph_breaks_reasons` ..: would all be accessed via `torch.compiler.explain()`
5. `replay` does not seem useful to end customers
6. . `graph_break()`: Mostly useful for debugging or unit tests
9. `register_backend()`: End users will just pass a string backend to torch.compile, only devs will create new backends
10. `export()` : Eventually this needs to public but for now it’s not ready so just highlighting that it will be in the public API eventually
11. `disallow_in_graph()`: Usage is limited
12. `mark_static()`: we can keep this private until dynamic=True is recommended in stable
13. `mark_dynamic()`: we can keep this private until dynamic=True is recommended in trunk
14. 8. `OptimizedModule`: This is the only class that we'd expose but is crucial since users are running code like `if isinstance(mod, OptimizedModule): torch.save(mod._orig_mod)` EDIT: because we fixed pickling we no longer need to
expose this
15. `is_compiling()`: Still not clear how this useful to end users
There are also config variables which we need to expose https://github.com/pytorch/pytorch/blob/main/torch/_dynamo/config.py
Some of our configs are useful dev flags, others are to gate experimental functionality and others are essential debugging tools and we seperate out the essential debugging and logging tools to a public facing config.
TODO: I still need to think of a good way of porting the config in a BC way here are some ideas
1. Just make all passes available and controllable via `torch.compile(options={})` but only show docstrings for the ones users should care about.
The current problem with our config system is we have 3 ways of setting them once via `options={}`, environment variables and variables in `config.py`, it'd be worth settling on one source of truth and have that be the public API.
The configs we should make public are
1. `log_file_name`
2. `verbose`
3. `cache_size_limit`
4. `repro_level` and `repro_after`: Although we can rename these to minifier and give human readable names to the levels
Everything else should stay private in particular
1. `print_graph_breaks`, `print_specializations`: should be supplanted by `explain()` for public users
2. dynamic shape configs : Users should only have to worry about `torch.compile(dynamic=True/False)`
3. The distributed flags, hook or guard configs: If we tell a user to use FSDP and DDP then the flag should be enabled by default or be in a private namespace
4. The fbcode flags: Obviously no need to be user facing
5. Skip/Allow lists: Not something normal users should play around with
#### From _inductor
Very little of inductor should be exposed in a public facing API, our core audience as in people writing models mostly just need information on what certain passes mean and how to control them a high level and they can do this with `torch.compile(options={})` so the goal here should be more to make available passes clearer and ideally consolidate them into `torch.compile()` docstrings or modes.
There are some exceptions though from https://github.com/pytorch/pytorch/blob/main/torch/_inductor/__init__.py
1. `list_mode_options()`
2. `list_options()`: this needs an additional pass to hide internal or debug options
For both of these we’d rename them to compiler.inductor_list_mode_options and compiler.inductor_list_options() since they would be in the same init file as the one for dynamo
Notable omissions
1. `_inductor.compile()`: Because of users are coming in with their own fx graph, they are likely developers
2. `_inductor.aot_compile()`:Again this is about capturing and modifying fx graphs so users APIs don't need to be public
However the configs are a slightly different story, because we can choose to either
1. Make all configs public
2. Make some configs public and keep most of the private ones. If public config is set it should override the private version
3. Make all configs controllable via `torch.compile(options={})` but make list_options() hide more things
For now 3 seems like the most reasonable choice with some high level configs we’ll keep like TORCH_COMPILE_DEBUG
Regardless here's what should probably be public or advertised more
1. `disable_progress` and verbose_progress: Combine and enable by default
2. `fallback_random`: We could make the case this shouldn't be public if a top level deterministic mode enables this
3. `profile_bandwidth`: Or could make the case that this should be in TORCH_COMPILE_DEBUG
Notable omissions
1. Any config that would generally improve performance for most that we should probably enable by default but might be disabled in the short term because of stability: example `epilogue_fusion`, `pattern_matcher`, `reordering`
2. Autotuning flags: Should just sit behind `torch.compile(mode="max-autotune")` like `max_autotune`, `max_autotune_gemm`
3. `coordinate_descent_tuning`: This one I'm a but mixed about, maybe it just also fall into `mode="max-autotune"`
4. `trace`: `TORCH_COMPILE_DEBUG` is the best flag for all of this
5. `triton.cudagraphs`: Default should be `torch.compile(mode="reduce-overhead")` - I'd go further and rename the `mode=cudagraph` and we can keep reduce-overhead for BC reasons
6. `triton_unique_kernel_names`: Mostly useful for devs debugging
7. `dce`: which doesnt really do anything
8. `shape_padding`: Elias is working on enabling this by default in which case we also remove it
## Mechanics
This PR would include the public functions with their docstrings
Another PR will take a stab at the configs
And for work where the APIs are still being cleaned up whether its minifier or escape hatches, export or dynamic shapes, aot_inductor etc.. we’ll keep them private until a public commitment can be made
Pull Request resolved: https://github.com/pytorch/pytorch/pull/102182
Approved by: https://github.com/jansel, https://github.com/albanD
Previously, cudagraphs and dynamic_shapes were incompatible and enabling
dynamic shapes would forcibly disable cudagraphs. This new strategy
I think is better. The idea is essentially that cudagraphs is an
"optimization" that happens to guard on every input. When cudagraphs
is on, we force everything static, and this automatically does the right
thing because we will force a recompile if sizes change.
This obsoletes https://github.com/pytorch/pytorch/pull/101813
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/103290
Approved by: https://github.com/voznesenskym, https://github.com/eellison
Previously, cudagraphs and dynamic_shapes were incompatible and enabling
dynamic shapes would forcibly disable cudagraphs. This new strategy
I think is better. The idea is essentially that cudagraphs is an
"optimization" that happens to guard on every input. When cudagraphs
is on, we force everything static, and this automatically does the right
thing because we will force a recompile if sizes change.
This obsoletes https://github.com/pytorch/pytorch/pull/101813
Signed-off-by: Edward Z. Yang <ezyang@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/103290
Approved by: https://github.com/voznesenskym
<!--
copilot:summary
-->
### <samp>🤖 Generated by Copilot at 08f7a6a</samp>
This pull request adds support for triton kernels in `torch` and `torch/cuda`, and refactors and tests the existing triton kernel for BSR matrix multiplication. It also adds a test case to ensure that importing `torch` does not implicitly import `triton`.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/98403
Approved by: https://github.com/malfet, https://github.com/cpuhrsch
Summary: cuda graph doesn't work with cuda 11's cupti lazy reinit. So we'll turn it off if any modules turn on cudagraph
Test Plan: test with cuda graph on
Reviewed By: aaronenyeshi
Differential Revision: D45967197
Pull Request resolved: https://github.com/pytorch/pytorch/pull/101848
Approved by: https://github.com/aaronenyeshi
Also not sure if this should be a public function or not. Leaving it private for now but let me know if you prefer for it to be public.
FYI @nikitaved this will logically conflict with your triton kernel PR.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/101420
Approved by: https://github.com/malfet
Description:
- As suggested by Nikita, created `torch.backends.cpu` submodule and exposed `get_cpu_capability`.
- In torchvision Resize method we want to know current cpu capability in order to pick appropriate codepath depending on cpu capablities
Newly coded vectorized resize of uint8 images on AVX2 supported CPUs is now faster than older way (uint8->float->resize->uint8). However, on non-avx hardware (e.g. Mac M1) certain configs are slower using native uint8.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/100164
Approved by: https://github.com/albanD, https://github.com/malfet
Caches output tensors for the common case when the output Tensor storage is unaliased for all graph outputs in all paths. For these persisted tensors we adjust the liveness tracking by also checking that the output tensor does not have an additional python reference.
I limit cached output tensors to be unaliased. If a descendent node discovers it has an alias of a prior output, then the aliased output will no longer be persisted in the ancestor.
The large majority of tensors are unaliased, and preserving aliased output tensors would add significant additional complexity with marginal gains. For instance, when do checkpointing and re-recordings, we need to remove the persisted tensors otherwise it would prevent memory from being reclaimed. If a single persisted tensor was present in multiple paths then that would create an inter-path dependence which adds complexity. Additionally, each further caching of the output would affect the reference count of the other caches, and that reference count would also need to be adjusted depending on if a node was checkpointed.
Still need to do a complete a run but for the models I tried makes the performance extremely close between trees and non trees impl.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/98944
Approved by: https://github.com/jansel, https://github.com/ngimel
using the existing deterministic implementation via `index_put` which has a deterministic implementation based on sorting indices.
With the `accumulate` arg in `index_put`, this can work for both scatter and scatter_reduce with sum/mean reduction mode.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/98060
Approved by: https://github.com/mikaylagawarecki
Summary:
Adds NNC-like logging that is configured through an env var `TORCH_COMPILE_LOGS`
Examples:
`TORCH_LOGS="dynamo,guards" python script.py` - prints dynamo logs at level INFO with guards of all functions that are compiled
`TORCH_LOGS="+dynamo,guards,graph" python script.py` - prints dynamo logs at level DEBUG with guards and graphs (in tabular) format of all graphs that are compiled
[More examples with full output](https://gist.github.com/mlazos/b17f474457308ce15e88c91721ac1cce)
Implementation:
The implementation parses the log settings from the environment, finds any components (aot, dynamo, inductor) or other loggable objects (guards, graph, etc.) and generates a log_state object. This object contains all of the enabled artifacts, and a qualified log name -> level mapping. _init_logs then adds handlers to the highest level logs (the registered logs), and sets any artifact loggers to level DEBUG if the artifact is enabled.
Note: set_logs is an alternative for manipulating the log_state, but if the environment contains TORCH_LOGS, the environment settings will be prioritized.
Adding a new log:
To add a new log, a dev should add their log name to torch._logging._registrations (there are examples there already).
Adding a new artifact:
To add a new artifact, a dev should add their artifact name to torch._logging._registrations as well.
Additionally, wherever the artifact is logged, `torch._logging.getArtifactLogger(__name__, <artifact_name>)` should be used instead of the standard logging implementation.
[design doc](https://docs.google.com/document/d/1ZRfTWKa8eaPq1AxaiHrq4ASTPouzzlPiuquSBEJYwS8/edit#)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/94858
Approved by: https://github.com/ezyang
Finding out what the inductor configs mean has been a confusing point for the community so creating some top level functions that just print them out to console if people don't wanna muck around the source code
Pull Request resolved: https://github.com/pytorch/pytorch/pull/95824
Approved by: https://github.com/jansel
Following the same logic of preloading cudnn and cublas from the pypi folder in multi-arch disributions, where Pure-lib vs Plat-lib matters, this PR adds the logic for the rest of the cuda pypi libraries that were integrated.
I have tested this PR by running the code block locally and installing/uninstalling nvidia pypi libraries:
```
import sys
import os
def _preload_cuda_deps():
"""Preloads cudnn/cublas deps if they could not be found otherwise."""
# Should only be called on Linux if default path resolution have failed
cuda_libs = {
'cublas': 'libcublas.so.11',
'cudnn': 'libcudnn.so.8',
'cuda_nvrtc': 'libnvrtc.so.11.2',
'cuda_runtime': 'libcudart.so.11.0',
'cuda_cupti': 'libcupti.so.11.7',
'cufft': 'libcufft.so.10',
'curand': 'libcurand.so.10',
'cusolver': 'libcusolver.so.11',
'cusparse': 'libcusparse.so.11',
'nccl': 'libnccl.so.2',
'nvtx': 'libnvToolsExt.so.1',
}
cuda_libs_paths = {lib_folder: None for lib_folder in cuda_libs.keys()}
for path in sys.path:
nvidia_path = os.path.join(path, 'nvidia')
if not os.path.exists(nvidia_path):
continue
for lib_folder, lib_name in cuda_libs.items():
candidate_path = os.path.join(nvidia_path, lib_folder, 'lib', lib_name)
if os.path.exists(candidate_path) and not cuda_libs_paths[lib_folder]:
cuda_libs_paths[lib_folder] = candidate_path
if all(cuda_libs_paths.values()):
break
if not all(cuda_libs_paths.values()):
none_libs = [lib for lib in cuda_libs_paths if not cuda_libs_paths[lib]]
raise ValueError(f"{', '.join(none_libs)} not found in the system path {sys.path}")
_preload_cuda_deps()
```
I don't have access to a multi-arch environment, so if somebody could verify a wheel with this patch on a multi-arch distribution, that would be great!
Pull Request resolved: https://github.com/pytorch/pytorch/pull/94355
Approved by: https://github.com/atalman
Per @ezyang's advice, added magic sym_int method. This works for 1.0 * s0 optimization, but can't evaluate `a>0` for some args, and still misses some optimization that model rewrite achieves, so swin still fails
(rewrite replaces `B = int(windows.shape[0] / (H * W / window_size / window_size))` with `B = (windows.shape[0] // int(H * W / window_size / window_size))` and model passes)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/94365
Approved by: https://github.com/ezyang
Per @ezyang's advice, added magic sym_int method. This works for 1.0 * s0 optimization, but can't evaluate `a>0` for some args, and still misses some optimization that model rewrite achieves, so swin still fails
(rewrite replaces `B = int(windows.shape[0] / (H * W / window_size / window_size))` with `B = (windows.shape[0] // int(H * W / window_size / window_size))` and model passes)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/94365
Approved by: https://github.com/ezyang
