Issue Station
Sign In
Tools hereToolbox

pytorch - ✅(Solved) Fix DISABLED test_mutated_benchmark (__main__.ComboKernelBenchmarkTestsPerSubkernelBlocks) [2 pull requests, 1 comments, 2 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#180012•Fetched 2026-04-11 06:10:13
View on GitHub
Comments
1
Participants
2
Timeline
98
Reactions
0
Author
praguptapragupta
Participants
praguptapraguptapytorch-bot[bot]pytorch-bot[bot]
Timeline (top)
mentioned ×38subscribed ×38unsubscribed ×10labeled ×9

Root Cause

This test was disabled because it is failing on main branch (recent examples).

PR fix notes

PR #180277: [ROCm][inductor][UT] Preserve combo kernel HIP compile options

Description (problem / solution / changelog)

Fixes #180011 Fixes #180012 Fixes #180013 Fixes #180014 Fixes #180015 Fixes #180016 Fixes #180021 Fixes #179952 Fixes #180022 Fixes #180025 Fixes #180027 Fixes #180028 Fixes #180029 Fixes #180549

This appears to have regressed in #177715, where combo-kernel autotuning started materializing alternate per-subkernel configs. This change keeps Triton HIP compile options such as waves_per_eu, matrix_instr_nonkdim, and kpack kernel-wide when combo kernels rewrite per-subkernel configs, avoiding invalid names like waves_per_eu_0. It also routes both baseline combo configs and sequential combo autotune trials through the same kwarg rewrite helper and adds a regression test covering combo-kernel kwarg rewriting for AMD special config args.

Made with Cursor

cc @jeffdaily @sunway513 @jithunnair-amd @pruthvistony @ROCmSupport @jataylo @hongxiayang @pragupta @jerrymannil @xinyazhang @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @kadeng @muchulee8 @amjames @chauhang @aakhundov @coconutruben @mlazos

Changed files

  • test/inductor/test_combo_kernels.py (modified, +26/-0)
  • torch/_inductor/runtime/triton_heuristics.py (modified, +60/-26)

PR #180319: [ROCm] Handle profiler trace format differences in combo kernel tests

Description (problem / solution / changelog)

Fixes #180017. Fixes #180018. Fixes #180019.

Three combo kernel tests access event["args"]["grid"] from profiler chrome traces. The grid key structure differs between CUPTI and ROCTracer trace formats, so this throws KeyError on ROCm.

Use .get() and only assert on grid values when present. All other validation (numerical correctness, kernel counts, codegen) still runs. No change to CUDA path where grid is always present there so assertions execute as before.

cc @jeffdaily @sunway513 @jithunnair-amd @pruthvistony @ROCmSupport @jataylo @hongxiayang @naromero77amd @pragupta @jerrymannil @xinyazhang @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @kadeng @muchulee8 @amjames @chauhang @aakhundov @coconutruben

Changed files

  • test/inductor/test_combo_kernels.py (modified, +25/-14)
RAW_BUFFERClick to expand / collapse

Platforms: rocm

This test was disabled because it is failing on main branch (recent examples).

cc @jeffdaily @sunway513 @jithunnair-amd @pruthvistony @ROCmSupport @jataylo @hongxiayang @naromero77amd @jerrymannil @xinyazhang @mruberry @chauhang @penguinwu @voznesenskym @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @kadeng @muchulee8 @amjames @aakhundov @coconutruben

extent analysis

TL;DR

The test failure in test_combo_kernels.py on the main branch may be resolved by investigating and addressing the issues causing the failures in ComboKernelBenchmarkTestsPerSubkernelBlocks.

Guidance

  • Investigate the recent examples of test failures on torch-ci.com to identify patterns or common causes of the failures.
  • Review the test_combo_kernels.py file, specifically the ComboKernelBenchmarkTestsPerSubkernelBlocks class, to understand the test logic and potential failure points.
  • Consider re-enabling the test and running it locally or on a CI environment to gather more information about the failure.

Notes

The provided information lacks specific technical details about the failure, making it challenging to provide a more targeted solution.

Recommendation

Apply workaround: Investigate and address the underlying causes of the test failures, as upgrading to a fixed version is not mentioned in the issue.

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…
#agent setup#task chaining#parallel task#integration issue#index setup

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 DISABLED test_mutated_benchmark (__main__.ComboKernelBenchmarkTestsPerSubkernelBlocks) [2 pull requests, 1 comments, 2 participants]