Issue Station
Sign In
Tools hereToolbox

pytorch - ✅(Solved) Fix Add a document for reusing the testcases in PyTorch. [1 pull requests, 1 participants]

Official PRs (…)
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.

Helpful · Quick feedback

Loading…
GitHub stats
pytorch/pytorch#181474•Fetched 2026-04-26 05:05:44
View on GitHub
Comments
0
Participants
1
Timeline
25
Reactions
0
Author
fffrogfffrog
Participants
fffrogfffrog
Timeline (top)
mentioned ×8subscribed ×8labeled ×6project_v2_item_status_changed ×2

PR fix notes

PR #180820: [Test] Support skipping test classes and methods in device-type tests

Description (problem / solution / changelog)

Fixes #177253

Summary

Introduces a mechanism to selectively skip test classes and test methods during device-specific test instantiation, built on top of instantiate_device_type_tests and DeviceTypeTestBase.

Backends can now declaratively specify which tests to skip without modifying upstream test files.

Changes

  • Adds skipped_testcases to DeviceTypeTestBase.
  • Device-specific test class can define this field to control which test classes or methods are skipped.

Changed files

  • test/cpp_extensions/open_registration_extension/torch_openreg/tests/test_testing.py (modified, +39/-0)
  • torch/testing/_internal/common_device_type.py (modified, +27/-1)
RAW_BUFFERClick to expand / collapse

A new dedicated page need to be added to the Accelerator Integration Documentation Serials to assist developers of new accelerators in reusing PyTorch's built-in test cases. This guide primarily focuses on the following aspects:

This page will be updated continuously as the PyTorch testing refactor progresses.

cc @svekars @sekyondaMeta @AlannaBurke @mruberry

extent analysis

TL;DR

Create a new dedicated page in the Accelerator Integration Documentation Serials to provide guidance on reusing PyTorch's built-in test cases for developers of new accelerators.

Guidance

  • Review the referenced GitHub issues and pull request to understand the current state of PyTorch testing refactor and its relevance to the new page.
  • Identify the key aspects of PyTorch's built-in test cases that need to be documented for reuse by new accelerator developers.
  • Determine the format and content of the new page, considering the continuous updates as the PyTorch testing refactor progresses.
  • Collaborate with the mentioned individuals (@svekars, @sekyondaMeta, @AlannaBurke, @mruberry) to ensure the new page meets the requirements and is accurate.

Notes

The creation of the new page is dependent on the ongoing PyTorch testing refactor, and its content may need to be updated accordingly.

Recommendation

Apply workaround by creating the new page with the currently available information and updating it as the PyTorch testing refactor progresses, to provide timely guidance to developers of new accelerators.

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?
Loading…
#callback error#memory management#API rate limit#retriever error#indexing error

Still need to ship something?

×6

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

Back to top recommendations

TRENDING

pytorch - ✅(Solved) Fix Add a document for reusing the testcases in PyTorch. [1 pull requests, 1 participants]