[SYCL][E2E] Re-enable HIP tests disabled by flaky CI tracker #17464#22724
Open
zjin-lcf wants to merge 1 commit into
Open
[SYCL][E2E] Re-enable HIP tests disabled by flaky CI tracker #17464#22724zjin-lcf wants to merge 1 commit into
zjin-lcf wants to merge 1 commit into
Conversation
Removes UNSUPPORTED: hip from three tests that were disabled due to the flaky AMD/HIP CI failures tracked in intel#17464 (memory faults/hangs under parallel test execution, worked around with -j1). These tests were not themselves broken on HIP; they pass on an AMD Instinct MI210 (gfx90a). - WeakObject/weak_object_expired.cpp (intel#17415) - HostInteropTask/host-task-dependency3.cpp (intel#17235) - Adapters/retain_events.cpp (intel#17236)
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This re-enables three E2E tests that were marked
UNSUPPORTED: hipas part of the flaky AMD/HIP CI cluster tracked in #17464. Those failures were not defects in the tests themselves — they were the parallel-execution memory faults/hangs on the AMD runner (worked around with-j1). The tests run and pass on an AMD Instinct MI210 (gfx90a):WeakObject/weak_object_expired.cpp(e2e weak_object_expired GPU crash #17415)HostInteropTask/host-task-dependency3.cpp(SYCL :: HostInteropTask/host-task-dependency3.cpp flaky fails on AMD pre-commit #17235) — both%{run}invocations (default and10)Adapters/retain_events.cpp(SYCL :: Adapters/retain_events.cpp flaky timeout on AMD CI #17236)Verification
Built with a recent DPC++ (
-fsycl -fsycl-targets=amdgcn-amd-amdhsa,--offload-arch=gfx90a) and executed on an MI210 with ROCm 7.1.1; all three pass. Separately, the heavy-concurrency memory-fault signature from #17464 no longer reproduces on ROCm 7.1.1 (hundreds of concurrent runs, 0 faults/hangs), consistent with the underlying ROCm/driver bug having been resolved in newer ROCm.Note:
WorkGroupMemory/basic_usage.cpp(#17339) is intentionally left disabled — it currently triggers a separate device-compiler backend crash on gfx90a, unrelated to the #17464 concurrency issue.CI
Please run the HIP/AMD pre-commit to confirm these pass on the CI runners' ROCm version before merging.