)

Fix pytorch#140462 .

Horace found that when we implicitly fallback to eager, some eager kernels may not work correctly if Inductor provide non-contiguous inputs (due to padding etc.). The original issue is found for the backward op of weight_norm. The fix in this PR is a general one: we force inputs to all implicit fallback kernels to be contiguous.

I have to refactor the code a bit to make it work. Previously we apply layout constraint in `GraphLowering.run_node`. We looks for implicit fallback in `call_function`. The problem here is, when we setup the implicit fallback in `call_function` with a layout constraint, we don't have a chance to apply the constraints.. The refactor moves the code that applies layout constraints to `call_function`.

Pull Request resolved: pytorch#140996
Approved by: https://github.com/jansel

…0084)

Summary: The way we've been de/serializing sympy.Exprs is not roundtrippable in all cases (serialize by calling `str(expr)`, and deserialize by calling `sympy.sympify(expr_str)`). This has led to expressions being mathematically equivalent but structurally different, causing issues in ValueRanges. Example issue: pytorch#136797

This starts to deprecate the use of `expr_str` and stores expressions in AST format instead. For BC purposes, `expr_str` deserialization is still supported, but we will always serialize to `expr_ast`. We'll kill this once the serialization upgrader design is finalized and implemented.

Test Plan: test_export

Differential Revision: D65638757

Pull Request resolved: pytorch#140084
Approved by: https://github.com/angelayi

…41003)

Fixes `python test/inductor/test_fused_attention.py SDPAPatternRewriterCpuTests.test_pattern_fails_with_unsupported_mask_cpu` when `specialize_float=False`. You might wonder how it's related, it's because there is a "negative" test that expects us not to match. Previously it would fail on isinstance(param, Tensor), but now that we tensorify the float, it did match and caused a failure. This check ensures the mask has the same shape to ensure this negative test case actually fails.

Pull Request resolved: pytorch#141003
Approved by: https://github.com/ezyang
ghstack dependencies: pytorch#140983

Implement RNG Generator by falling back to CPUGeneratorImpl.

Pull Request resolved: pytorch#138449
Approved by: https://github.com/ezyang

…1022)

Detailed description:

The codes below will raise an error
```Python
import torch
from torch.fx.experimental.proxy_tensor import make_fx

def func(a):
    b = a + 1
    c = b.view(-1)
    c.add_(1)
    return b

input = torch.randn(2)
out = make_fx(func)(input)
```

The error info are like below:
```Python
...
  File "/root/Git.d/pytorch/pytorch/torch/_dynamo/codegen.py", line 34, in <module>
    from .variables.torch_function import TensorWithTFOverrideVariable
  File "/root/Git.d/pytorch/pytorch/torch/_dynamo/variables/torch_function.py", line 185, in <module>
    populate_builtin_to_tensor_fn_map()
  File "/root/Git.d/pytorch/pytorch/torch/_dynamo/variables/torch_function.py", line 146, in populate_builtin_to_tensor_fn_map
    inp0 = torch.ones(1)
  File "/root/Git.d/pytorch/pytorch/torch/fx/experimental/proxy_tensor.py", line 1240, in __torch_function__
    return func(*args, **kwargs)
  File "/root/Git.d/pytorch/pytorch/torch/utils/_stats.py", line 21, in wrapper
    return fn(*args, **kwargs)
  File "/root/Git.d/pytorch/pytorch/torch/fx/experimental/proxy_tensor.py", line 1342, in __torch_dispatch__
    return proxy_call(self, func, self.pre_dispatch, args, kwargs)
  File "/root/Git.d/pytorch/pytorch/torch/fx/experimental/proxy_tensor.py", line 907, in proxy_call
    name=proxy_mode.tracer.graph._target_to_str(func.overloadpacket.__name__),
AttributeError: 'PythonKeyTracer' object has no attribute 'graph'
...
```

Solutions:
Import torch._dynamo before dispatch_trace is called to avoid the context set before dispatch_trace from affecting the torch._dynamo import.

Pull Request resolved: pytorch#141022
Approved by: https://github.com/ezyang

Detailed description:

As the title stated.
Pull Request resolved: pytorch#141023
Approved by: https://github.com/Skylion007, https://github.com/ezyang

Tracking issue: pytorch#138399

This PR changes the `pow` C++ implementation, making its C++ meta kernel consistent with
its Python ref implementation. The following example shows the inconsistency between the
two:

```python
def run(device):
    S = (5,)
    a = torch.rand(S, device=device, dtype=torch.float32)
    b = 2
    out = torch.empty(S, device=device, dtype=torch.float64)
    return torch.pow(a, b, out=out)

>>> run("cpu")
Traceback (most recent call last):
  File "test.py", line 34, in run
    return torch.pow(a, b, out=out)
RuntimeError: Found dtype Double but expected Float

>>> run("meta")
tensor(..., device='meta', size=(5,), dtype=torch.float64)
```

**~Update:~**

~Note that this happens only for `pow.Tensor_Scalar` overloads. Therefore, this PR needed
further 2 modifications:~

- ~Split the `pow` ref implementation, making `pow.Tensor_Scalar` error on mismatching
output dtypes~
- ~Create a dispatch for `pow` when `_refs.pow()` is called~

**Update:**

Changing the `TensorIteratorConfig` for `pow.Tensor_Scalar` was easier and,
after the discussion below, more correct. The solution was to change the
`TensorIteratorBase::build_output_borrowing_argument_owning_unary_op` function,
setting:

- `cast_common_dtype_to_outputs`; and
- `enforce_safe_casting_to_output`.

Pull Request resolved: pytorch#140287
Approved by: https://github.com/ezyang

Pull Request resolved: pytorch#141092
Approved by: https://github.com/Skylion007, https://github.com/kulinseth
ghstack dependencies: pytorch#141089, pytorch#141090

Instead of calling `REGISTER_FUSED_ADAM_OP` macro with 7 parameters 16 times, 4 type parameter macros for each op and then one op to define the quartet of ops: Adam, AdamW and their grad functions
Pull Request resolved: pytorch#141103
Approved by: https://github.com/kulinseth
ghstack dependencies: pytorch#141089, pytorch#141090, pytorch#141092

* Automatically applies ruff rule 401. Turns loops into equivalent list comprehensions which are faster and do not leak the scope of the loop variables.
* list comprehensions not only often have better typing, but are 50+% faster than for loops on overhead. They also preserve length information etc and are better for the interpreter to optimize.
* Manually went back and made mypy happy after the change.
* Also fixed style lints in files covered by flake8 but not by pyfmt

Pull Request resolved: pytorch#140980
Approved by: https://github.com/justinchuby, https://github.com/malfet

Pull Request resolved: pytorch#140250
Approved by: https://github.com/jansel
ghstack dependencies: pytorch#140223

Easy fix to missing reserve calls in NT Matmul CUDA kernel to improve perf.
Pull Request resolved: pytorch#141130
Approved by: https://github.com/malfet

This uploads the MPS benchmark results to benchmark database.  The data can then be queried, for example:

```
select benchmark, model, metric from oss_ci_benchmark_v3 where head_sha = '99a133116fee15aa1467165f2b209b37da53f189' and metric.name in ['eager_peak_mem', 'dynamo_peak_mem', 'speedup'] and model.name = 'BERT_pytorch'
```

I'm documenting the JSON format at https://github.com/pytorch/pytorch/wiki/How-to-integrate-with-PyTorch-OSS-benchmark-database

### Testing

Locally,

```
PYTHONPATH=/Users/huydo/Storage/mine/benchmark python benchmarks/dynamo/torchbench.py --performance --only resnet152 --backend eager --training --devices mps --output test/test-reports/torchbench_training.csv
```

Workflow dispatch https://github.com/pytorch/pytorch/actions/runs/11927990520

Pull Request resolved: pytorch#141087
Approved by: https://github.com/malfet

I'm trying to make this benchmark results available on OSS benchmark database, so that people can query it from outside.  The first step is to also record the results in the JSON format compatible with the database schema defined in pytorch/test-infra#5839.

Existing CSV files remain unchanged.

### Testing

The JSON results are uploaded as artifacts to S3 https://github.com/pytorch/pytorch/actions/runs/11809725848/job/32901411180#step:26:13, for example https://gha-artifacts.s3.amazonaws.com/pytorch/pytorch/11809725848/1/artifact/test-jsons-test-pr_time_benchmarks-1-1-linux.g4dn.metal.nvidia.gpu_32901411180.zip

Pull Request resolved: pytorch#140493
Approved by: https://github.com/laithsakka

…#141145)

A small fix I just realize after pytorch#141087.

Pull Request resolved: pytorch#141145
Approved by: https://github.com/malfet

Fixes pytorch#107302

This is a clone and fix for pytorch#139199.

This PR is a small step for the overall NumPy 2 support.
It adds a new CI job for testing with NumPy 2 with one test file only.
More tests to be fixed and added later in follow-up pull requests.

Pull Request resolved: pytorch#140586
Approved by: https://github.com/malfet

Co-authored-by: Nikita Shulga <[email protected]>