doc: update release notes #6438
Conversation
Signed-off-by: junq <[email protected]>
WalkthroughThe release notes for TensorRT-LLM v0.21.0 were updated to clarify that a known performance regression for full chunked attention support with the LLaMA4 model only affects Hopper GPUs. The update also notes that the root cause is identified and a fix is planned for a future release. Changes
Estimated code review effort🎯 1 (Trivial) | ⏱️ ~2 minutes Possibly related PRs
Suggested reviewers
Poem
Note ⚡️ Unit Test Generation is now available in beta!Learn more here, or try it out under "Finishing Touches" below. ✨ Finishing Touches🧪 Generate unit tests
🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
SupportNeed help? Create a ticket on our support page for assistance with any issues or questions. Note: Be mindful of the bot's finite context window. It's strongly recommended to break down tasks such as reading entire modules into smaller chunks. For a focused discussion, use review comments to chat about specific files and their changes, instead of using the PR comments. CodeRabbit Commands (Invoked using PR comments)
Other keywords and placeholders
Documentation and Community
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 0
🧹 Nitpick comments (1)
docs/source/release-notes.md (1)
76-76: Minor grammar & clarity touch-up recommendedThe sentence is hard to parse and has a few grammatical nits (missing space before parenthesis, subject–verb agreement, GPU mention). Consider tightening it for readability:
-In 0.21, full chunked attention support has been added to make sure LLaMA4 model can functionally run with > 8K seq length, while there is a known performance regression(only affect LLaMA4 model) on Hopper due to this functional enhancement. The root cause of the regression has been identified already and the fix will be part of the future release. +In 0.21, full chunked-attention support was added so that the LLaMA4 model can run with sequence lengths > 8 K. However, this functional enhancement introduces a known performance regression on Hopper GPUs that affects only the LLaMA4 model. The root cause has been identified, and a fix is planned for a future release.
📜 Review details
Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
docs/source/release-notes.md(1 hunks)
🧰 Additional context used
🧠 Learnings (2)
📓 Common learnings
Learnt from: moraxu
PR: NVIDIA/TensorRT-LLM#6303
File: tests/integration/test_lists/qa/examples_test_list.txt:494-494
Timestamp: 2025-07-28T17:06:08.621Z
Learning: In TensorRT-LLM testing, it's common to have both CLI flow tests (test_cli_flow.py) and PyTorch API tests (test_llm_api_pytorch.py) for the same model. These serve different purposes: CLI flow tests validate the traditional command-line workflow, while PyTorch API tests validate the newer LLM API backend. Both are legitimate and should coexist.
Learnt from: amitz-nv
PR: NVIDIA/TensorRT-LLM#5616
File: tensorrt_llm/executor/worker.py:375-384
Timestamp: 2025-07-17T09:01:27.402Z
Learning: In tensorrt_llm/executor/worker.py, the LoRA adapter cache optimization logic that checks `is_adapter_in_cpu_cache()` and conditionally passes None for weights/config has a known race condition issue that cannot be solved with simple error handling or verification checks. This is a known limitation that requires a more comprehensive solution.
Learnt from: yiqingy0
PR: NVIDIA/TensorRT-LLM#5198
File: jenkins/mergeWaiveList.py:0-0
Timestamp: 2025-07-22T08:33:49.109Z
Learning: In the TensorRT-LLM waive list merging system, removed lines are always located at the end of the merge waive lists, which is why the mergeWaiveList.py script uses reverse traversal - it's an optimization for this specific domain constraint.
Learnt from: yechank-nvidia
PR: NVIDIA/TensorRT-LLM#6254
File: tensorrt_llm/_torch/pyexecutor/model_engine.py:1201-1204
Timestamp: 2025-07-22T09:22:14.726Z
Learning: In TensorRT-LLM's multimodal processing pipeline, shared tensor recovery using `from_shared_tensor()` is only needed during the context phase. Generation requests reuse the already-recovered tensor data and only need to call `strip_for_generation()` to remove unnecessary multimodal data while preserving the recovered tensors. This avoids redundant tensor recovery operations during generation.
Learnt from: CR
PR: NVIDIA/TensorRT-LLM#0
File: CODING_GUIDELINES.md:0-0
Timestamp: 2025-07-28T15:39:50.377Z
Learning: Applies to **/*.{cpp,h,hpp,cc,cxx,cu,py} : All TensorRT-LLM Open Source Software code should contain an NVIDIA copyright header that includes the current year. This includes .cpp, .h, .cu, .py, and any other source files which are compiled or interpreted.
docs/source/release-notes.md (3)
Learnt from: moraxu
PR: #6303
File: tests/integration/test_lists/qa/examples_test_list.txt:494-494
Timestamp: 2025-07-28T17:06:08.621Z
Learning: In TensorRT-LLM testing, it's common to have both CLI flow tests (test_cli_flow.py) and PyTorch API tests (test_llm_api_pytorch.py) for the same model. These serve different purposes: CLI flow tests validate the traditional command-line workflow, while PyTorch API tests validate the newer LLM API backend. Both are legitimate and should coexist.
Learnt from: amitz-nv
PR: #5616
File: tensorrt_llm/executor/worker.py:375-384
Timestamp: 2025-07-17T09:01:27.402Z
Learning: In tensorrt_llm/executor/worker.py, the LoRA adapter cache optimization logic that checks is_adapter_in_cpu_cache() and conditionally passes None for weights/config has a known race condition issue that cannot be solved with simple error handling or verification checks. This is a known limitation that requires a more comprehensive solution.
Learnt from: yechank-nvidia
PR: #6254
File: tensorrt_llm/_torch/pyexecutor/model_engine.py:1201-1204
Timestamp: 2025-07-22T09:22:14.726Z
Learning: In TensorRT-LLM's multimodal processing pipeline, shared tensor recovery using from_shared_tensor() is only needed during the context phase. Generation requests reuse the already-recovered tensor data and only need to call strip_for_generation() to remove unnecessary multimodal data while preserving the recovered tensors. This avoids redundant tensor recovery operations during generation.
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)
- GitHub Check: Pre-commit Check
|
/bot skip --comment "doc changes" |
|
/bot skip --comment "doc changes" |
|
PR_Github #13328 [ skip ] triggered by Bot |
|
PR_Github #13328 [ skip ] completed with state |
Signed-off-by: junq <[email protected]>
Signed-off-by: junq <[email protected]>
Signed-off-by: junq <[email protected]>
Signed-off-by: junq <[email protected]>
Signed-off-by: junq <[email protected]> Signed-off-by: Lanyu Liao <[email protected]>
Signed-off-by: junq <[email protected]>
3 participants
Summary by CodeRabbit