oneapi-src / oneapi-src/unified-runtime

Heap APIs aborting on enqueue tests on L0 adapter

Open
#1,094 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

bug ci/cd conformance level-zero
Dominant language
C++
Stars
57
Forks
120
Avg merge
1d 14h
Merged PRs (30d)
1

Description

While testing adapter third bump PR, L0 enqueue tests was failing sometimes with a heap APIs problems like here, it have aborted with free(): corrupted unsorted chunks, on a rerun it sometimes also abort with malloc(): unsorted double linked list corrupted like here. On a third rerun, it sometimes also fail normally like here. The normal failure is expected from this test, but the heap calls aborting shouldn't be expected and wasn't consistent. This was also only happening on CI with me.

Note: This problem wasn't happening before the third adapter bump as the tests was seg-faulting earlier in the tests pipeline in urEnqueueDeviceGetGlobalVariableReadTest.InvalidEventWaitInvalidEvent but the adapter third bump will fix this test so it allowed the tests pipeline to reach and test following tests which one of them have caused this.

I believe this is coming from one of urEnqueueEventsWaitTest tests but not entirely sure of that.

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the CI runs linked in the issue and the urEnqueueEventsWaitTest tests, comparing their intermittent heap-abort failures with the expected normal failure. Also review urEnqueueDeviceGetGlobalVariableReadTest.InvalidEventWaitInvalidEvent, which previously stopped the pipeline. Done means identifying the source of the corrupted heap behavior and confirming the enqueue test sequence no longer aborts intermittently.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp
Domain
testing
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
30/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.