inductor.test_cpu_select_algorithm tests are failing because of OSError: Bad file descriptor and AssertionError: Scalars are not equal! on main branch
Tests Failing:
test_aoti_bmm_unique_identifiers_cpu_float32
Env details:
PyTorch version: 2.10.0
Branch: main
OS: RHEL 9.6
CPU: Intel
python version: 3.12
commit id : 6de6685797cabc6256df76803f3a5f772d5275a7 (tag: trunk/6de6685797cabc6256df76803f3a5f772d5275a7, origin/main, origin/HEAD)
Steps to repro:
Log in to H200.
Login to quay.io: podman login quay.io
Pull base image: podman pull quay.io/aipcc/pytorch:rhel_cuda_build_without_pins
Run the image and specify the GPU to be used: podman run -it <IMAGE_NAME>
Run the PyTorch UT: TEST_CONFIG=cpu python3 test/run_test.py -i inductor.test_cpu_select_algorithm
Expected result: UTs should run fine.
Actual result:
OSError: [Errno 9] Bad file descriptor - The test fails consistently with a bad file descriptor error occurring during pytest teardown, indicating a file descriptor issue in the subprocess execution environment.
AssertionError: Scalars are not equal! - Expected 1 but got 0. The test expects cpp_templated_kernel_counter to be 1, but it is 0, indicating that the templated kernel was not generated or counted as expected.
Logs are attached below