[https://nvbugs/5845769][fix] B300(sm103) support on VLMs

[yechank-nvidia]

Summary by CodeRabbit

• Chores
  • Improved GPU kernel compatibility for additional hardware configurations.

[coderabbitai]

📝 Walkthrough

Walkthrough

Adds a conditional SM value remapping in getXMMAKernelsV2 where kSM_103 is remapped to kSM_100 before kernel retrieval, mirroring an existing kSM_121 to kSM_120 mapping pattern. The change adjusts SM selection without altering function signatures or control flow logic.

Changes

Cohort / File(s)	Summary
SM Kernel Remapping cpp/tensorrt_llm/kernels/contextFusedMultiHeadAttention/fused_multihead_attention_v2.cpp	Added conditional to remap kSM_103 to kSM_100 in getXMMAKernelsV2 function, following existing SM architecture remapping pattern.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

🚥 Pre-merge checks | ✅ 1 | ❌ 2

❌ Failed checks (2 warnings)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.
Description check	⚠️ Warning	The PR description is empty except for the template comments; it lacks required sections including a description of the issue/solution, test coverage information, and PR checklist verification.	Add a meaningful description explaining the B300 support changes, list relevant test coverage, and complete the PR checklist items to meet repository standards.

✅ Passed checks (1 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title follows the required template format with [https://nvbugs/5845769][fix] prefix and clearly describes the change: adding B300 (sm103) support for VLMs.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

• 📝 Generate docstrings

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

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

[yechank-nvidia]

/bot run

[tensorrt-cicd]

PR_Github #34758 [ run ] triggered by Bot. Commit: 30a8c0c

[tensorrt-cicd]

PR_Github #34788 [ run ] triggered by Bot. Commit: 30a8c0c

[tensorrt-cicd]

PR_Github #34788 [ run ] completed with state SUCCESS. Commit: 30a8c0c
/LLM/release-1.2/L0_MergeRequest_PR pipeline #318 completed with status: 'FAILURE'

⚠️ 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

[yechank-nvidia]

/bot run

[tensorrt-cicd]

PR_Github #34864 [ run ] triggered by Bot. Commit: 30a8c0c

[tensorrt-cicd]

PR_Github #34864 [ run ] completed with state SUCCESS. Commit: 30a8c0c
/LLM/release-1.2/L0_MergeRequest_PR pipeline #327 completed with status: 'FAILURE'

⚠️ 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

[yechank-nvidia]

/bot run

[tensorrt-cicd]

PR_Github #34898 [ run ] triggered by Bot. Commit: 30a8c0c

[tensorrt-cicd]

PR_Github #34898 [ run ] completed with state SUCCESS. Commit: 30a8c0c
/LLM/release-1.2/L0_MergeRequest_PR pipeline #330 completed with status: 'SUCCESS'