Skip aten.alias calls in proxy_tensor

[Chillee]

Stack from ghstack (oldest at bottom):

• Add memory_budget prototype to checkpoint #123335
• -> Skip aten.alias calls in proxy_tensor #126018

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/126018

• 📄 Preview Python docs built from this PR
• 📄 Preview C++ docs built from this PR
• ❓ Need help or want to give feedback on the CI? Visit the bot commands wiki or our office hours

Note: Links to docs will display an error until the docs builds have been completed.

❌ 39 New Failures

As of commit 8c97118 with merge base 0dda338 ():

NEW FAILURES - The following jobs have failed:

• inductor / cuda12.1-py3.10-gcc9-sm80 / test (inductor_torchbench_smoketest_perf, 1, 1, linux.gcp.a100) (gh)
  Process completed with exit code 1.
• pull / linux-focal-cuda12.1-py3.10-gcc9 / test (default, 1, 5, linux.4xlarge.nvidia.gpu) (gh)
  RuntimeError: doctests 1/1 failed
• pull / linux-focal-cuda12.1-py3.10-gcc9 / test (default, 2, 5, linux.4xlarge.nvidia.gpu) (gh)
  test_proxy_tensor.py::TestFakeProxyTensor::test_alias
• pull / linux-focal-cuda12.1-py3.10-gcc9 / test (default, 3, 5, linux.4xlarge.nvidia.gpu) (gh)
  functorch/test_aotdispatch.py::TestAOTAutograd::test_input_mutation_set__nop
• pull / linux-focal-cuda12.1-py3.10-gcc9 / test (default, 4, 5, linux.4xlarge.nvidia.gpu) (gh)
  test_functionalization.py::TestFunctionalization::test_instance_norm
• pull / linux-focal-cuda12.1-py3.10-gcc9 / test (default, 5, 5, linux.4xlarge.nvidia.gpu) (gh)
  inductor/test_fused_attention.py::SDPAPatternRewriterCpuTests::test_sdpa_rewriter_11_cpu
• pull / linux-focal-cuda12.1-py3.10-gcc9-sm86 / test (default, 1, 5, linux.g5.4xlarge.nvidia.gpu) (gh)
  RuntimeError: doctests 1/1 failed
• pull / linux-focal-cuda12.1-py3.10-gcc9-sm86 / test (default, 2, 5, linux.g5.4xlarge.nvidia.gpu) (gh)
  test_proxy_tensor.py::TestFakeProxyTensor::test_alias
• pull / linux-focal-cuda12.1-py3.10-gcc9-sm86 / test (default, 3, 5, linux.g5.4xlarge.nvidia.gpu) (gh)
  functorch/test_aotdispatch.py::TestAOTAutograd::test_input_mutation_set__nop
• pull / linux-focal-py3.11-clang10 / test (crossref, 1, 2, linux.2xlarge) (gh)
  functorch/test_eager_transforms.py::TestLinearizeCPU::test_linearize_basic_cpu_float32
• pull / linux-focal-py3.11-clang10 / test (crossref, 2, 2, linux.2xlarge) (gh)
  test_proxy_tensor.py::TestFakeProxyTensor::test_alias
• pull / linux-focal-py3.11-clang10 / test (default, 1, 3, linux.2xlarge) (gh)
  RuntimeError: doctests 1/1 failed
• pull / linux-focal-py3.11-clang10 / test (default, 2, 3, linux.2xlarge) (gh)
  functorch/test_aotdispatch.py::TestAOTAutograd::test_input_mutation_set__nop
• pull / linux-focal-py3.11-clang10 / test (default, 3, 3, linux.2xlarge) (gh)
  test_proxy_tensor.py::TestFakeProxyTensor::test_alias
• pull / linux-focal-py3.11-clang10 / test (dynamo, 1, 3, linux.2xlarge) (gh)
  test_nn.py::TestNN::test_transformerdecoder
• pull / linux-focal-py3.11-clang10 / test (dynamo, 2, 3, linux.2xlarge) (gh)
  functorch/test_eager_transforms.py::TestLinearizeCPU::test_linearize_basic_cpu_float32
• pull / linux-focal-py3.11-clang10 / test (dynamo, 3, 3, linux.2xlarge) (gh)
  functorch/test_aotdispatch.py::TestAOTAutograd::test_input_mutation_set__nop
• pull / linux-focal-py3.12-clang10 / test (default, 1, 3, linux.2xlarge) (gh)
  RuntimeError: doctests 1/1 failed
• pull / linux-focal-py3.12-clang10 / test (default, 2, 3, linux.2xlarge) (gh)
  functorch/test_aotdispatch.py::TestAOTAutograd::test_input_mutation_set__nop
• pull / linux-focal-py3.12-clang10 / test (default, 3, 3, linux.2xlarge) (gh)
  test_proxy_tensor.py::TestFakeProxyTensor::test_alias
• pull / linux-focal-py3.12-clang10 / test (dynamo, 1, 3, linux.2xlarge) (gh)
  test_nn.py::TestNN::test_transformerdecoder
• pull / linux-focal-py3.12-clang10 / test (dynamo, 2, 3, linux.2xlarge) (gh)
  functorch/test_eager_transforms.py::TestLinearizeCPU::test_linearize_basic_cpu_float32
• pull / linux-focal-py3.12-clang10 / test (dynamo, 3, 3, linux.2xlarge) (gh)
  functorch/test_aotdispatch.py::TestAOTAutograd::test_input_mutation_set__nop
• pull / linux-focal-py3.8-clang10 / test (crossref, 1, 2, linux.2xlarge) (gh)
  functorch/test_aotdispatch.py::TestAOTAutograd::test_input_mutation_set__nop
• pull / linux-focal-py3.8-clang10 / test (crossref, 2, 2, linux.2xlarge) (gh)
  test_proxy_tensor.py::TestFakeProxyTensor::test_alias
• pull / linux-focal-py3.8-clang10 / test (default, 1, 3, linux.2xlarge) (gh)
  RuntimeError: doctests 1/1 failed
• pull / linux-focal-py3.8-clang10 / test (default, 2, 3, linux.2xlarge) (gh)
  functorch/test_aotdispatch.py::TestAOTAutograd::test_input_mutation_set__nop
• pull / linux-focal-py3.8-clang10 / test (default, 3, 3, linux.2xlarge) (gh)
  test_proxy_tensor.py::TestFakeProxyTensor::test_alias
• pull / linux-focal-py3.8-clang10 / test (dynamo, 1, 3, linux.2xlarge) (gh)
  test_nn.py::TestNN::test_transformerdecoder
• pull / linux-focal-py3.8-clang10 / test (dynamo, 3, 3, linux.2xlarge) (gh)
  functorch/test_aotdispatch.py::TestAOTAutograd::test_input_mutation_set__nop
• pull / linux-focal-py3.8-clang10-onnx / test (default, 1, 2, linux.2xlarge) (gh)
  onnx/test_fx_to_onnx_with_onnxruntime.py::TestFxToOnnxWithOnnxRuntime_op_level_debug_True_dynamic_shapes_True_model_type_TorchModelType.TORCH_NN_MODULE::test_expand_as_fill_separate_tensor
• pull / linux-focal-py3.8-clang10-onnx / test (default, 2, 2, linux.2xlarge) (gh)
  onnx/test_fx_to_onnx_with_onnxruntime.py::TestFxToOnnxWithOnnxRuntime_op_level_debug_True_dynamic_shapes_True_model_type_TorchModelType.TORCH_NN_MODULE::test_expand_as_fill_tensor
• pull / linux-jammy-py3.10-clang15-asan / test (default, 2, 6, linux.4xlarge) (gh)
  inductor/test_fused_attention.py::SDPAPatternRewriterCpuTests::test_sdpa_rewriter_11_cpu
• pull / linux-jammy-py3.10-clang15-asan / test (default, 3, 6, linux.4xlarge) (gh)
  test_proxy_tensor.py::TestFakeProxyTensor::test_alias
• pull / linux-jammy-py3.10-clang15-asan / test (default, 5, 6, linux.4xlarge) (gh)
  dynamo/test_dynamic_shapes.py::DynamicShapesFuncTorchHigherOrderOpTests::test_linearize_jvp_fn_dynamic_shapes
• pull / linux-jammy-py3.10-clang15-asan / test (default, 6, 6, linux.4xlarge) (gh)
  dynamo/test_dynamic_shapes.py::DynamicShapesFuncTorchHigherOrderOpTests::test_linearize_disable_capture_dynamic_shapes
• pull / linux-jammy-py3.8-gcc11 / test (default, 1, 3, linux.2xlarge) (gh)
  RuntimeError: doctests 1/1 failed
• pull / linux-jammy-py3.8-gcc11 / test (default, 2, 3, linux.2xlarge) (gh)
  test_proxy_tensor.py::TestFakeProxyTensor::test_alias
• pull / linux-jammy-py3.8-gcc11 / test (default, 3, 3, linux.2xlarge) (gh)
  functorch/test_aotdispatch.py::TestAOTAutograd::test_input_mutation_set__nop

This comment was automatically generated by Dr. CI and updates every 15 minutes.

[ezyang]

Naw man, you can't do this, inside proxy tensor it's not guaranteed you are post functionalization