[rocprofiler-sdk] Fix queue interposition completion waiter ordering

[venkat1361]

Motivation

• transpose-sampling test from rocprofiler-systems can hang because queue interposition may start the async completion-signal waiter before the final real doorbell is published, so the waiter can block waiting for GPU completion while the GPU has not yet been told about the packet.

Technical Details

• Defer completion-waiter enqueue from write_interceptor() to process_doorbell_impl(), after publish_submitted_packets() publishes the real write index and rings the doorbell.

JIRA ID

Test Plan

Test Result

• https://github.com/ROCm/rocm-systems/actions/runs/28333482927/job/83974368084(includes commits from PR which is causing test failures).

Submission Checklist

• Look over the contributing guidelines at https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests.

[therock-pr-bot]

❌ PR Check — Action Required

Check	Status	Details
🌿 Branch Name	✅ Pass	—
📝 PR Title/Description	❌ Fail	Error: Title does not follow Conventional Commits style. Expected: start with a valid type (feat, fix, docs, …). Desired format: type(optional-scope): short description ─── Error: PR description must reference a JIRA ID, ISSUE ID, or a GitHub closing keyword. Expected: include a JIRA ID / ISSUE ID line (separator : or -, or omitted; value may be a JIRA key, a number with/without #, or a link), OR a closing keyword + issue reference. Accepted examples: • JIRA ID : TESTAUTO-6039 • JIRA ID - #330 • JIRA ID #330 • ISSUE ID : TESTUTO-3334 • ISSUE ID #3334 • ISSUE ID - TESTAUTO-3433 • ISSUE ID : https://github.com/<org_name>/<repo_name>/issues/1234 • Closes #10 • Fixes octo-org/octo-repo#100 • Resolves: #123 • #123 • https://github.com/<org_name>/<repo_name>/issues/123 Current: no valid JIRA/ISSUE/closing-keyword reference found
⛔ Forbidden Files	✅ Pass	—
🧪 Unit Test	❌ Fail	Error: Source/code files changed without an accompanying unit test. Expected: add at least one test file named like test_<name>.py / test_<name>.cpp (or <name>_test.*). Current: code file(s) changed: projects/rocprofiler-sdk/source/lib/rocprofiler-sdk/hsa/queue_interposition.cpp, projects/rocprofiler-sdk/source/lib/rocprofiler-sdk/hsa/queue_interposition.hpp, projects/rocprofiler-sdk/source/lib/rocprofiler-sdk/tests/queue_interposition.cpp; no test file found
🔎 pre-commit	⏳ Pending	⏳ Still running…
🚫 Draft PR	🔜 To Be Enabled	—
🚩 Feature Flag	🔜 To Be Enabled	—
📊 Code Coverage	🔜 To Be Enabled	—

⚠️ 2 policy check(s) failed. Please address the issues above before this PR can be Reviewed.

🚫 Please fix the failed policies

• ❌ PR Title/Description
• ❌ Unit Test

The Not ready to Review label was added to this PR. Once all policies pass, the label is removed automatically.

📖 Need help? See the Policy FAQ for details on every check and how to fix failures.

[therock-pr-bot]

🚫 Please fix the failed policies before requesting reviews.

The following policy checks failed:

• ❌ PR Title/Description
• ❌ Unit Test

The Not ready to Review label has been added to this PR.
Once all policies pass, the label will be removed automatically.

[Copilot]

Pull request overview

This PR addresses a potential hang in rocprofiler-sdk’s inline HSA queue interposition path by ensuring completion-signal waiters are not enqueued until after the real write index has been published and the final doorbell has been rung.

Changes:

• Defers async completion-waiter enqueue from write_interceptor() to process_doorbell_impl() (after publish_submitted_packets()).
• Refactors async-signal-handler thread-count selection into get_async_signal_handler_thread_count().
• Adds unit tests covering env-driven thread-count selection and clamping behavior.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated no comments.

File	Description
projects/rocprofiler-sdk/source/lib/rocprofiler-sdk/tests/queue_interposition.cpp	Adds unit tests for async-signal-handler thread-count selection via env vars.
projects/rocprofiler-sdk/source/lib/rocprofiler-sdk/hsa/queue_interposition.hpp	Exposes get_async_signal_handler_thread_count() (and includes <cstddef> for size_t).
projects/rocprofiler-sdk/source/lib/rocprofiler-sdk/hsa/queue_interposition.cpp	Implements deferred waiter scheduling and centralizes thread-count env logic.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[minseobshin11]

LGTM

[bwelton]

While i don't have opposition to this change.

queue interposition may start the async completion-signal waiter before the final real doorbell is published. so the waiter can block waiting for GPU completion while the GPU has not yet been told about the packet.

This shouldn't matter. The async just wont trigger until the completion signal is set. Its not clear why this change fixes anything other than changing the ordering and holding the lock for potentially less time.

[rahulc-gh]

LGTM - wait for CI to Pass