Add SQTT marker support to rocSHMEM to improve device-level performance tracing

[mberenjk]

Motivation

Adding sqtt_marker to device functions and critical sections of the code to mprove device-level performance tracing

Technical Details

Add SQTT marker support to rocSHMEM to enable device-side performance tracing. Similar to ROCTX, this provides annotations for performance analysis, but unlike ROCTX, it works on the device side, allowing correlation of rocSHMEM operations directly with GPU execution. This will improve trace visibility and help identify kernel-level bottlenecks more effectively.

JIRA ID

AIROCSHMEM-431

Test Plan

A complete test will be added in the CI with rocprofiler-sdk

Test Result

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/rocshmem/src/backend_type.hpp, projects/rocshmem/src/gda/bnxt/queue_pair_bnxt.cpp, projects/rocshmem/src/gda/context_gda_device.cpp, projects/rocshmem/src/gda/context_gda_tmpl_device.hpp, projects/rocshmem/src/gda/ionic/queue_pair_ionic.cpp (+3 more); 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.

[omor1]

Please hold until #7217 is merged to prevent conflicts.

[ApoKalipse-V]

Is the switch case itself a performance concern?
If not, I'd recommend to have just enter post_wqe_rma here
(it can be if gda_provider_ can be slow to load)

[ApoKalipse-V]

Would it be useful if we provided a way for enter/exit to contain extra information? E.g.

sqtt_marker_enter("mymarker")

if (a)
sqtt_marker_exit("mymarker", "A")
else
sqtt_marker_exit("mymarker", "B")

[ApoKalipse-V]

I think you need a flag here to switch these off when needed, like you do in #ifdef ENABLE_ROCTX

Imagine we changed the path, or sqtt_trace.hpp wasn't available for some reason

[ApoKalipse-V]

We plan on having this available globally in rocm soon. I dont think it hurts for now, though.

[ApoKalipse-V]

Doesn't hurt, but it's off by default.