[BMG] [upstream_ut][Failed to collect] UT collection failures #6165

(myvenv) root@0b9b86ce49de:/workspace/pytorch/test/inductor# python -m pytest -v test_aoti_torchbind_constants.py
========================================================== test session starts ==========================================================
platform linux -- Python 3.12.16, pytest-7.3.2, pluggy-1.7.0 -- /workspace/myvenv/bin/python
cachedir: .pytest_cache
hypothesis profile 'dev' -> database=None, max_examples=10, suppress_health_check=[HealthCheck.too_slow]
benchmark: 5.3.1 (defaults: timer=time.perf_counter disable_gc=False min_rounds=5 min_time=0.000005 max_time=1.0 calibration_precision=10 warmup=False warmup_iterations=100000)
rootdir: /workspace/pytorch
configfile: pytest.ini
plugins: platformdirs-4.14.5, xdist-3.3.1, cpp-2.3.0, hypothesis-6.56.4, xdoctest-1.3.0, subtests-0.13.1, anyio-4.16.0, flakefinder-1.1.0, rerunfailures-14.0, hydra-core-1.3.8, timeout-2.4.0, benchmark-5.3.1
collected 0 items / 1 error
Running 0 items in this shard

================================================================ ERRORS =================================================================
____________________________________ ERROR collecting test/inductor/test_aoti_torchbind_constants.py ____________________________________
Traceback (most recent call last):
  File "/workspace/myvenv/lib/python3.12/site-packages/torch/testing/_internal/common_device_type.py", line 576, in _init_and_get_primary_device
    return cls.get_primary_device()
           ^^^^^^^^^^^^^^^^^^^^^^^^
  File "/workspace/myvenv/lib/python3.12/site-packages/torch/testing/_internal/common_device_type.py", line 949, in get_primary_device
    return cls.primary_device
           ^^^^^^^^^^^^^^^^^^
AttributeError: type object 'TestTorchbindAOTIDeviceXPU' has no attribute 'primary_device'. Did you mean: 'get_primary_device'?

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/workspace/pytorch/test/inductor/test_aoti_torchbind_constants.py", line 122, in <module>
    instantiate_device_type_tests(TestTorchbindAOTIDevice, globals(), allow_xpu=True)
  File "/workspace/myvenv/lib/python3.12/site-packages/torch/testing/_internal/common_device_type.py", line 1324, in instantiate_device_type_tests
    device_type_test_class.instantiate_test(
  File "/workspace/myvenv/lib/python3.12/site-packages/torch/testing/_internal/common_device_type.py", line 767, in instantiate_test
    instantiate_test_helper(
  File "/workspace/myvenv/lib/python3.12/site-packages/torch/testing/_internal/common_device_type.py", line 654, in instantiate_test_helper
    device_arg: str = cls._init_and_get_primary_device()
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/workspace/myvenv/lib/python3.12/site-packages/torch/testing/_internal/common_device_type.py", line 581, in _init_and_get_primary_device
    cls.setUpClass()
  File "/workspace/myvenv/lib/python3.12/site-packages/torch/testing/_internal/common_device_type.py", line 1302, in _setUpClass
    generic_test_class.setUpClass.__func__(cls)
  File "/workspace/pytorch/test/inductor/test_aoti_torchbind_constants.py", line 27, in setUpClass
    init_torchbind_implementations()
  File "/workspace/myvenv/lib/python3.12/site-packages/torch/testing/_internal/torchbind_impls.py", line 19, in init_torchbind_implementations
    load_torchbind_test_lib()
  File "/workspace/myvenv/lib/python3.12/site-packages/torch/testing/_internal/torchbind_impls.py", line 188, in load_torchbind_test_lib
    lib_file_path = find_library_location("libtorchbind_test.so")
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/workspace/myvenv/lib/python3.12/site-packages/torch/testing/_internal/common_utils.py", line 6079, in find_library_location
    raise FileNotFoundError(f"Could not find library '{lib_name}' in installed or build paths")
FileNotFoundError: Could not find library 'libtorchbind_test.so' in installed or build paths
======================================================== short test summary info ========================================================
ERROR test_aoti_torchbind_constants.py - FileNotFoundError: Could not find library 'libtorchbind_test.so' in installed or build paths
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Interrupted: 1 error during collection !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
=========================================================== 1 error in 1.95s ============================================================

(myvenv) root@0b9b86ce49de:/workspace/pytorch/test/inductor# python -m pytest -v test_comm_analysis.py
========================================================== test session starts ==========================================================
platform linux -- Python 3.12.16, pytest-7.3.2, pluggy-1.7.0 -- /workspace/myvenv/bin/python
cachedir: .pytest_cache
hypothesis profile 'dev' -> database=None, max_examples=10, suppress_health_check=[HealthCheck.too_slow]
benchmark: 5.3.1 (defaults: timer=time.perf_counter disable_gc=False min_rounds=5 min_time=0.000005 max_time=1.0 calibration_precision=10 warmup=False warmup_iterations=100000)
rootdir: /workspace/pytorch
configfile: pytest.ini
plugins: platformdirs-4.14.5, xdist-3.3.1, cpp-2.3.0, hypothesis-6.56.4, xdoctest-1.3.0, subtests-0.13.1, anyio-4.16.0, flakefinder-1.1.0, rerunfailures-14.0, hydra-core-1.3.8, timeout-2.4.0, benchmark-5.3.1
collected 0 items / 1 error
Running 0 items in this shard

================================================================ ERRORS =================================================================
_________________________________________ ERROR collecting test/inductor/test_comm_analysis.py __________________________________________
Traceback (most recent call last):
  File "/workspace/pytorch/test/inductor/test_comm_analysis.py", line 13, in <module>
    sys.exit(0)
SystemExit: 0
------------------------------------------------------------ Captured stderr ------------------------------------------------------------
c10d NCCL not available, skipping tests
======================================================== short test summary info ========================================================
ERROR test_comm_analysis.py - SystemExit: 0
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Interrupted: 1 error during collection !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
=========================================================== 1 error in 0.03s ============================================================

(myvenv) root@0b9b86ce49de:/workspace/pytorch/test/mobile# python -m pytest -v test_lite_script_module.py
========================================================== test session starts ==========================================================
platform linux -- Python 3.12.16, pytest-7.3.2, pluggy-1.7.0 -- /workspace/myvenv/bin/python
cachedir: .pytest_cache
hypothesis profile 'dev' -> database=None, max_examples=10, suppress_health_check=[HealthCheck.too_slow]
benchmark: 5.3.1 (defaults: timer=time.perf_counter disable_gc=False min_rounds=5 min_time=0.000005 max_time=1.0 calibration_precision=10 warmup=False warmup_iterations=100000)
rootdir: /workspace/pytorch
configfile: pytest.ini
plugins: platformdirs-4.14.5, xdist-3.3.1, cpp-2.3.0, hypothesis-6.56.4, xdoctest-1.3.0, subtests-0.13.1, anyio-4.16.0, flakefinder-1.1.0, rerunfailures-14.0, hydra-core-1.3.8, timeout-2.4.0, benchmark-5.3.1
collected 0 items / 1 error
Running 0 items in this shard

================================================================ ERRORS =================================================================
________________________________________ ERROR collecting test/mobile/test_lite_script_module.py ________________________________________
ImportError while importing test module '/workspace/pytorch/test/mobile/test_lite_script_module.py'.
Hint: make sure your test modules/packages have valid Python names.
Traceback:
/root/.local/share/uv/python/cpython-3.12.16-linux-x86_64-gnu/lib/python3.12/importlib/__init__.py:90: in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
test_lite_script_module.py:5: in <module>
    from tempfile import TemporaryFileName
E   ImportError: cannot import name 'TemporaryFileName' from 'tempfile' (/root/.local/share/uv/python/cpython-3.12.16-linux-x86_64-gnu/lib/python3.12/tempfile.py). Did you mean: 'TemporaryFile'?
======================================================== short test summary info ========================================================
ERROR test_lite_script_module.py
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Interrupted: 1 error during collection !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
=========================================================== 1 error in 0.04s ============================================================

(myvenv) root@0b9b86ce49de:/workspace/pytorch/test/quantization/core/experimental# python -m pytest -v test_fake_quantize.py
========================================================== test session starts ==========================================================
platform linux -- Python 3.12.16, pytest-7.3.2, pluggy-1.7.0 -- /workspace/myvenv/bin/python
cachedir: .pytest_cache
hypothesis profile 'dev' -> database=None, max_examples=10, suppress_health_check=[HealthCheck.too_slow]
benchmark: 5.3.1 (defaults: timer=time.perf_counter disable_gc=False min_rounds=5 min_time=0.000005 max_time=1.0 calibration_precision=10 warmup=False warmup_iterations=100000)
rootdir: /workspace/pytorch
configfile: pytest.ini
plugins: platformdirs-4.14.5, xdist-3.3.1, cpp-2.3.0, hypothesis-6.56.4, xdoctest-1.3.0, subtests-0.13.1, anyio-4.16.0, flakefinder-1.1.0, rerunfailures-14.0, hydra-core-1.3.8, timeout-2.4.0, benchmark-5.3.1
collected 0 items / 1 error
Running 0 items in this shard

================================================================ ERRORS =================================================================
______________________________ ERROR collecting test/quantization/core/experimental/test_fake_quantize.py _______________________________
ImportError while importing test module '/workspace/pytorch/test/quantization/core/experimental/test_fake_quantize.py'.
Hint: make sure your test modules/packages have valid Python names.
Traceback:
/root/.local/share/uv/python/cpython-3.12.16-linux-x86_64-gnu/lib/python3.12/importlib/__init__.py:90: in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
test_fake_quantize.py:6: in <module>
    from torch.ao.quantization.experimental.fake_quantize import APoTFakeQuantize
E   ModuleNotFoundError: No module named 'torch.ao.quantization.experimental'
======================================================== short test summary info ========================================================
ERROR test_fake_quantize.py
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Interrupted: 1 error during collection !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
=========================================================== 1 error in 0.03s ============================================================