pytorch - ✅(Solved) Fix [CUDA] `test_dlpack_tensor_on_different_device` fails with `torch.device` instead of `torch.cuda.device` [1 pull requests, 1 participants]

pytorch2026-04-21 21:01:30

ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

GitHub issue URL

Helpful · Quick feedback

GitHub stats

pytorch/pytorch#181024•Fetched 2026-04-22 07:43:01

View on GitHub

Comments

Participants

Timeline

Reactions

Author

gderossi

Participants

gderossi

Timeline (top)

mentioned ×7subscribed ×7labeled ×6cross-referenced ×1

Error Message

====================================================================== FAIL: test_dlpack_tensor_on_different_device_cuda (main.TestTorchDlPackCUDA.test_dlpack_tensor_on_different_device_cuda)

Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/torch/testing/_internal/common_utils.py", line 3448, in wrapper method(*args, **kwargs) File "/usr/local/lib/python3.12/dist-packages/torch/testing/_internal/common_utils.py", line 3448, in wrapper method(*args, **kwargs) File "/usr/local/lib/python3.12/dist-packages/torch/testing/_internal/common_device_type.py", line 439, in instantiated_test result = test(self, **param_kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/torch/testing/_internal/common_device_type.py", line 1255, in dep_fn return fn(slf, *args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/torch/testing/_internal/common_device_type.py", line 1468, in only_fn return fn(slf, *args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/torch/testing/_internal/common_device_type.py", line 1498, in multi_fn return fn(slf, devices, *args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/opt/pytorch/pytorch/test/test_dlpack.py", line 374, in test_dlpack_tensor_on_different_device with self.assertRaisesRegex( AssertionError: BufferError not raised

To execute this test, run the following from the base repo dir: python test/test_dlpack.py TestTorchDlPackCUDA.test_dlpack_tensor_on_different_device_cuda

This message can be suppressed by setting PYTORCH_PRINT_REPRO_ON_FAILURE=0

Ran 1 test in 0.435s

FAILED (failures=1)

Fix Action

Fixed

Fixed by PR: [CUDA] Use torch.cuda.device in dlpack test (https://github.com/pytorch/pytorch/pull/181025)

PR fix notes

PR #181025: [CUDA] Use torch.cuda.device in dlpack test

Repository: pytorch/pytorch
Author: gderossi
State: open | merged: False
Link: https://github.com/pytorch/pytorch/pull/181025

Description (problem / solution / changelog)

Fixes #181024 by using torch.cuda.device on CUDA and retaining torch.device for XPU. @shangerxin can you or somebody else please confirm that this change doesn't break this test on XPU?

cc @eqy

Changed files

test/test_dlpack.py (modified, +6/-3)

Code Example

- with torch.cuda.device(dev1):
+ with torch.device(dev1):

---

======================================================================
FAIL: test_dlpack_tensor_on_different_device_cuda (__main__.TestTorchDlPackCUDA.test_dlpack_tensor_on_different_device_cuda)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/usr/local/lib/python3.12/dist-packages/torch/testing/_internal/common_utils.py", line 3448, in wrapper
    method(*args, **kwargs)
  File "/usr/local/lib/python3.12/dist-packages/torch/testing/_internal/common_utils.py", line 3448, in wrapper
    method(*args, **kwargs)
  File "/usr/local/lib/python3.12/dist-packages/torch/testing/_internal/common_device_type.py", line 439, in instantiated_test
    result = test(self, **param_kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/testing/_internal/common_device_type.py", line 1255, in dep_fn
    return fn(slf, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/testing/_internal/common_device_type.py", line 1468, in only_fn
    return fn(slf, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/testing/_internal/common_device_type.py", line 1498, in multi_fn
    return fn(slf, devices, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/pytorch/pytorch/test/test_dlpack.py", line 374, in test_dlpack_tensor_on_different_device
    with self.assertRaisesRegex(
AssertionError: BufferError not raised

To execute this test, run the following from the base repo dir:
    python test/test_dlpack.py TestTorchDlPackCUDA.test_dlpack_tensor_on_different_device_cuda

This message can be suppressed by setting PYTORCH_PRINT_REPRO_ON_FAILURE=0

----------------------------------------------------------------------
Ran 1 test in 0.435s

FAILED (failures=1)

RAW_BUFFERClick to expand / collapse

🐛 Describe the bug

#173760 made the following change to test_dlpack_tensor_on_different_device:

- with torch.cuda.device(dev1):
+ with torch.device(dev1):

On CUDA, torch.device does not update the current CUDA device in the same way that torch.cuda.device does, causing the test to fail.

Test log:

======================================================================
FAIL: test_dlpack_tensor_on_different_device_cuda (__main__.TestTorchDlPackCUDA.test_dlpack_tensor_on_different_device_cuda)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/usr/local/lib/python3.12/dist-packages/torch/testing/_internal/common_utils.py", line 3448, in wrapper
    method(*args, **kwargs)
  File "/usr/local/lib/python3.12/dist-packages/torch/testing/_internal/common_utils.py", line 3448, in wrapper
    method(*args, **kwargs)
  File "/usr/local/lib/python3.12/dist-packages/torch/testing/_internal/common_device_type.py", line 439, in instantiated_test
    result = test(self, **param_kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/testing/_internal/common_device_type.py", line 1255, in dep_fn
    return fn(slf, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/testing/_internal/common_device_type.py", line 1468, in only_fn
    return fn(slf, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/testing/_internal/common_device_type.py", line 1498, in multi_fn
    return fn(slf, devices, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/pytorch/pytorch/test/test_dlpack.py", line 374, in test_dlpack_tensor_on_different_device
    with self.assertRaisesRegex(
AssertionError: BufferError not raised

To execute this test, run the following from the base repo dir:
    python test/test_dlpack.py TestTorchDlPackCUDA.test_dlpack_tensor_on_different_device_cuda

This message can be suppressed by setting PYTORCH_PRINT_REPRO_ON_FAILURE=0

----------------------------------------------------------------------
Ran 1 test in 0.435s

FAILED (failures=1)

Versions

GB200 on current main

cc @ptrblck @msaroufim @eqy @jerryzh168 @tinglvv @nWEIdia @mruberry

extent analysis

TL;DR

Revert the change in test_dlpack_tensor_on_different_device to use torch.cuda.device instead of torch.device to update the current CUDA device.

Guidance

The issue is caused by the difference in behavior between torch.device and torch.cuda.device when updating the current CUDA device.
To fix the test failure, revert the change made in test_dlpack_tensor_on_different_device to use torch.cuda.device instead of torch.device.
Verify the fix by re-running the test test_dlpack_tensor_on_different_device_cuda after making the change.
Consider adding a comment or documentation to explain the reason for using torch.cuda.device in this specific test case.

Example

# Revert the change to use torch.cuda.device
with torch.cuda.device(dev1):
    # test code here

Notes

The fix assumes that the test is intended to run on a CUDA device and that torch.cuda.device is the correct way to update the current CUDA device in this context.

Recommendation

Apply workaround: Revert the change to use torch.cuda.device to ensure the test runs correctly on CUDA devices.

Vote matrix · Quick signals

Works

Did the solution work? Tap to confirm.

Easy Fix

Was it a quick fix?

Time Saver

Did it save you time?

Blocking

Was it severely blocking?

Common Issue

Are others likely hitting this too?

Flaky / Intermittent

Is it intermittent?

Verified / Reproducible

Can you reproduce it reliably?

#file not found #serialization error #model compatibility #GPU setup #container setup

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Data

Security

Network

Code

UI/UX

Text

System

Multimedia

Protocol

API

Engineering

pytorch - ✅(Solved) Fix [CUDA] `test_dlpack_tensor_on_different_device` fails with `torch.device` instead of `torch.cuda.device` [1 pull requests, 1 participants]

Recommended Tools

GitHub issue graph ai analysis

Error Message

====================================================================== FAIL: test_dlpack_tensor_on_different_device_cuda (main.TestTorchDlPackCUDA.test_dlpack_tensor_on_different_device_cuda)

Fix Action

Fixed

PR fix notes

PR #181025: [CUDA] Use torch.cuda.device in dlpack test

Description (problem / solution / changelog)

Changed files

Code Example

🐛 Describe the bug

Versions

extent analysis

TL;DR

Guidance

Example

Notes

Recommendation

Still need to ship something?

TRENDING

pytorch - ✅(Solved) Fix [CUDA] `test_dlpack_tensor_on_different_device` fails with `torch.device` instead of `torch.cuda.device` [1 pull requests, 1 participants]

Recommended Tools

GitHub issue graph ai analysis

Error Message

====================================================================== FAIL: test_dlpack_tensor_on_different_device_cuda (main.TestTorchDlPackCUDA.test_dlpack_tensor_on_different_device_cuda)

Fix Action

Fixed

PR fix notes

PR #181025: [CUDA] Use torch.cuda.device in dlpack test

Description (problem / solution / changelog)

Changed files

Code Example

🐛 Describe the bug

Versions

extent analysis

TL;DR

Guidance

Example

Notes

Recommendation

Still need to ship something?

RELATED_DISCOVERY

TRENDING