-
Notifications
You must be signed in to change notification settings - Fork 27.7k
test_torch.py/test_inverse_cuda causes illegal memory access on some platforms #45403
Copy link
Copy link
Closed
Labels
high prioritymodule: dependency bugProblem is not caused by us, but caused by an upstream library we useProblem is not caused by us, but caused by an upstream library we usemodule: linear algebraIssues related to specialized linear algebra operations in PyTorch; includes matrix multiply matmulIssues related to specialized linear algebra operations in PyTorch; includes matrix multiply matmulmodule: regressionIt used to work, and now it doesn'tIt used to work, and now it doesn'tmodule: testsIssues related to tests (not the torch.testing module)Issues related to tests (not the torch.testing module)triage review
Milestone
Description
Metadata
Metadata
Assignees
Labels
high prioritymodule: dependency bugProblem is not caused by us, but caused by an upstream library we useProblem is not caused by us, but caused by an upstream library we usemodule: linear algebraIssues related to specialized linear algebra operations in PyTorch; includes matrix multiply matmulIssues related to specialized linear algebra operations in PyTorch; includes matrix multiply matmulmodule: regressionIt used to work, and now it doesn'tIt used to work, and now it doesn'tmodule: testsIssues related to tests (not the torch.testing module)Issues related to tests (not the torch.testing module)triage review
Run the test -> illegal memory access on CUDA (on some platforms).
This is indicative of the operation illegally accessing memory.
Likely (almost definitely) caused by #42403.
Environment
PyTorch version: 1.7.0a0+db4690c
Is debug build: True
CUDA used to build PyTorch: 10.1
ROCM used to build PyTorch: N/A
OS: Ubuntu 18.04.3 LTS (x86_64)
GCC version: (Ubuntu 7.4.0-1ubuntu1~18.04.1) 7.4.0
Clang version: 8.0.0 (tags/RELEASE_800/final)
CMake version: version 3.14.0
Python version: 3.8 (64-bit runtime)
Is CUDA available: True
CUDA runtime version: 10.1.105
GPU models and configuration:
GPU 0: Quadro GP100
GPU 1: Quadro GP100
Nvidia driver version: 440.82
cuDNN version: Could not collect
HIP runtime version: N/A
MIOpen runtime version: N/A
Versions of relevant libraries:
[pip3] numpy==1.19.1
[pip3] torch==1.7.0a0
[pip3] torchvision==0.7.0
[conda] blas 1.0 mkl
[conda] magma-cuda101 2.5.2 1 pytorch
[conda] mkl 2020.1 217
[conda] mkl-include 2020.1 217
[conda] mkl-service 2.3.0 py38he904b0f_0
[conda] mkl_fft 1.2.0 py38h23d657b_0
[conda] mkl_random 1.1.1 py38h0573a6f_0
[conda] numpy 1.19.1 py38hbc911f0_0
[conda] numpy-base 1.19.1 py38hfa32c7d_0
[conda] torch 1.7.0a0 dev_0
[conda] torchvision 0.7.0 pypi_0 pypi
cc @ezyang @gchanan @zou3519 @mruberry @VitalyFedyunin @vincentqb @vishwakftw @jianyuh @nikitaved @pearu
cc @xwang233