[https://nvbugs/6384622][fix] increase free_mem_frac for failed perf test

[reasonsolo]

Summary by CodeRabbit

• Bug Fixes
  • Updated a performance sanity configuration to reserve a bit more GPU memory for context caching, which may improve stability for the affected test scenario.
  • Removed an outdated test waiver entry, allowing the test to be evaluated normally again.

Description

Test Coverage

PR Checklist

Please review the following before submitting your PR:

• PR description clearly explains what and why. If using CodeRabbit's summary, please make sure it makes sense.

• PR Follows TRT-LLM CODING GUIDELINES to the best of your knowledge.

• Test cases are provided for new code paths (see test instructions)

• If PR introduces API changes, an appropriate PR label is added - either api-compatible or api-breaking. For api-breaking, include BREAKING in the PR title.

• Any new dependencies have been scanned for license and vulnerabilities

• CODEOWNERS updated if ownership changes

• Documentation updated as needed

• Update tava architecture diagram if there is a significant design change in PR.

• The reviewers assigned automatically/manually are appropriate for the PR.

• Please check this after reviewing the above items as appropriate for this PR.

GitHub Bot Help

To see a list of available CI bot commands, please comment /bot help.

[coderabbitai]

📝 Walkthrough

Walkthrough

This PR removes a SKIP waiver entry for a disaggregated NIXL performance sanity test case and increases the free_gpu_memory_fraction value in its associated context worker configuration from 0.3 to 0.4.

Changes

Test waiver and config update

Layer / File(s)	Summary
Unwaive test and adjust KV cache config tests/integration/test_lists/waives.txt, tests/scripts/perf-sanity/disaggregated/gb200_deepseek-r1-fp4_128k8k_con128_ctx1_pp8_gen1_dep16_eplb0_mtp1_ccb-NIXL.yaml	Removes the SKIP/nvbugs waiver for the disagg_upload NIXL test case and increases worker_config.ctx.kv_cache_config.free_gpu_memory_fraction from 0.3 to 0.4.

Estimated code review effort: 1 (Trivial) | ~3 minutes

Possibly related PRs

• NVIDIA/TensorRT-LLM#15825: Both PRs modify waives.txt for the same disagg_upload NIXL perf sanity test case.
• NVIDIA/TensorRT-LLM#15800: Both PRs remove SKIP waiver entries for disaggregated tests from waives.txt.
• NVIDIA/TensorRT-LLM#15811: Both PRs change SKIP waiver entries for disaggregated tests in waives.txt.

Suggested reviewers: mzweilz, chzblych, xinhe-nv

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Description check	⚠️ Warning	The PR description is just the template and checklist, with no actual issue summary, solution, or test coverage filled in.	Add a real Description and Test Coverage section that explains the change, why it was needed, and what tests validate it.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title matches the main change: increasing free GPU memory fraction for a perf-test fix.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

_{Comment @coderabbitai help to get the list of available commands.}

[reasonsolo]

/bot run --disable-fail-fast

[reasonsolo]

/bot run --disable-fail-fast

[tensorrt-cicd]

PR_Github #57286 [ run ] triggered by Bot. Commit: 9564cd2 Link to invocation

[tensorrt-cicd]

PR_Github #57286 [ run ] completed with state FAILURE. Commit: 9564cd2
/LLM/main/L0_MergeRequest_PR pipeline #46047 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

• Please check the failed tests and fix your PR
• If you cannot view the failures, ask the CI triggerer to share details
• Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

[reasonsolo]

/bot run --disable-fail-fast

[tensorrt-cicd]

PR_Github #57644 [ run ] triggered by Bot. Commit: 9564cd2 Link to invocation

[tensorrt-cicd]

PR_Github #57644 [ run ] completed with state SUCCESS. Commit: 9564cd2
/LLM/main/L0_MergeRequest_PR pipeline #46367 completed with status: 'SUCCESS'

CI Report

Link to invocation