[TRTLLM-9896][test] add vswa test cases coverage #10146
Conversation
Signed-off-by: Ivy Zhang <[email protected]>
Signed-off-by: Ivy Zhang <[email protected]>
Signed-off-by: Ivy Zhang <[email protected]>
crazydemo
force-pushed
the
add_vswa_case
branch
from
4b58f4a to
5b157c2
Compare
📝 WalkthroughWalkthroughThese changes extend the accuracy evaluation framework by introducing reference configurations for two new model variants (Gemma-3-1B-IT and GPT-OSS-120B-MXFP4) and expanding test coverage with VSWA block reuse and guided decoding test variants for Gemma and Eagle3 models across multiple GPU setups. Changes
Estimated code review effort🎯 2 (Simple) | ⏱️ ~15 minutes
Pre-merge checks and finishing touches❌ Failed checks (2 warnings)
✅ Passed checks (1 passed)
✨ Finishing touches
🧪 Generate unit tests (beta)
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. Comment |
![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 (3)
tests/integration/defs/accuracy/test_llm_api_pytorch.py (3)
1071-1115: Gemma3‑27B VSWA reuse tests look fine; confirm JsonModeEval reference for this modelThe FP8/NVFP4 VSWA reuse and guided‑decoding tests for
TestGemma3_27BInstructare consistent with the 1B variants (same VSWA pattern, just a larger window) and use reasonable KvCacheConfig settings.One thing to double‑check:
test_fp8_guided_decoding_vswa_reusecallstask = JsonModeEval(self.MODEL_NAME) # MODEL_NAME = "google/gemma-3-27b-it"but
json_mode_eval.yamlin this PR only adds a reference entry forgoogle/gemma-3-1b-it(plus GPT‑OSS and the existing Llama/DeepSeek models), not forgoogle/gemma-3-27b-it. Please confirm whetherJsonModeEvalis expected to run without a reference for 27B, or if you also intend to track accuracy for 27B and should add a corresponding entry.
1154-1183: Gemma3‑1B FP8 VSWA reuse + guided‑decoding tests are consistent; class‑level comment may be staleThe new
test_fp8_vswa_reuseandtest_fp8_guided_decoding_vswa_reuseforTestGemma3_1BInstruct:
- Reuse the same
max_attention_window=[512, 512, 512, 512, 512, 32768]VSWA configuration already exercised in the existing auto‑dtype VSWA tests.- Correctly point to the FP8 pre‑quantized checkpoint and assert via downstream tasks (GSM8K/MMLU or JsonModeEval) without changing core infra.
Given these additions, the class‑level comment:
# NOTE: Disable block reuse for SWA window model. kv_cache_config = KvCacheConfig(enable_block_reuse=True)is now misleading (and the attribute isn’t used by the new tests anyway). It would be clearer either to drop or update that comment/attribute so the VSWA reuse story for Gemma3‑1B is consistent in this class.
4492-4555: GPTOSS Eagle3 VSWA‑reuse and guided‑decoding 4‑GPU tests: configuration looks reasonable; a few knobs are worth double‑checkingThe new GPT‑OSS/Eagle3 tests do what the PR describes:
test_eagle3_vswa_reuse_4gpusexercises Eagle3 speculative decoding on 4 GPUs with VSWA enabled via:kv_cache_config = KvCacheConfig( free_gpu_memory_fraction=0.4, dtype="auto", enable_block_reuse=True, max_attention_window=[128, 32768], )
test_eagle3_guided_decoding_4gpuslayers xgrammar‑based guided decoding on top of Eagle3 (same 4‑GPU Eagle config, no VSWA‑specific flags), and correctly usesJsonModeEval("GPT-OSS/120B-MXFP4"), which now has a reference injson_mode_eval.yaml.A couple of details worth verifying while you’re here:
VSWA window shape for GPT‑OSS
Other VSWA tests (e.g., Gemma3) use a longer
max_attention_windowlist (multiple stage windows). Here you use a 2‑element list[128, 32768]. IfKvCacheConfig.max_attention_windowsemantics differ across models, that’s fine; just confirm this matches the intended GPT‑OSS attention schedule and the underlying model config.Overlap‑scheduler / SM‑specific behavior parity with
test_eagle3_4gpus
test_eagle3_4gpusabove carries explicit SM=90 skip logic and anoverlap_schedulerparameter to exercise/guard known accuracy issues for 2‑model Eagle3 + overlap scheduling.- The new VSWA/guided tests always run with
disable_overlap_schedulerleft at its default and no SM‑specific skips.If the same Hopper TP=4 Eagle3 limitations apply regardless of VSWA and guided decoding, you may want to mirror the
get_sm_version()==90skip and/or explicitly setdisable_overlap_schedulerbased onone_model(as is done for some llama/qwen Eagle3 tests), so these new tests don’t hit known-bad combinations unintentionally.If your local CG4/RTX6K runs already covered these scenarios and looked clean, you may decide to leave as‑is; otherwise, tightening those conditions would make the new tests more robust.
Also applies to: 4557-4599
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (4)
tests/integration/defs/accuracy/references/json_mode_eval.yaml(1 hunks)tests/integration/defs/accuracy/test_llm_api_pytorch.py(3 hunks)tests/integration/test_lists/qa/llm_function_core.txt(2 hunks)tests/integration/test_lists/qa/llm_function_rtx6k.txt(1 hunks)
🧰 Additional context used
📓 Path-based instructions (2)
**/*.py
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
**/*.py: Code developed for TensorRT-LLM should conform to Python 3.8+
Indent Python code with 4 spaces. Do not use tabs
Always maintain the namespace when importing in Python, even if only one class or function from a module is used
Python files should use snake_case naming:some_file.py
Python classes should use PascalCase naming:class SomeClass
Python functions and methods should use snake_case naming:def my_awesome_function():
Python local variables should use snake_case naming:my_variable = ...
Python variable names that start with a number should be prefixed with 'k':k_99th_percentile = ...
Python global variables should use upper snake_case with prefix 'G':G_MY_GLOBAL = ...
Python constants should use upper snake_case naming:MY_CONSTANT = ...
Avoid shadowing variables declared in an outer scope in Python
Initialize all externally visible members of a Python class in the constructor
For Python interfaces that may be used outside a file, prefer docstrings over comments
Python comments should be reserved for code within a function, or interfaces that are local to a file
Use Google style docstrings in Python for classes and functions, which can be parsed by Sphinx
Python attributes and variables can be documented inline with type and description
Avoid using reflection in Python when functionality can be easily achieved without reflection
When using try-except blocks in Python, limit the except to the smallest set of errors possible
When using try-except blocks in Python to handle multiple possible variable types (duck-typing), keep the body of the try as small as possible, using the else block for logic
Files:
tests/integration/defs/accuracy/test_llm_api_pytorch.py
**/*.{cpp,h,cu,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
All TensorRT-LLM Open Source Software code should contain an NVIDIA copyright header that includes the year of its latest meaningful modification
Files:
tests/integration/defs/accuracy/test_llm_api_pytorch.py
🧠 Learnings (7)
📓 Common learnings
Learnt from: pengbowang-nv
Repo: NVIDIA/TensorRT-LLM PR: 7192
File: tests/integration/test_lists/test-db/l0_dgx_b200.yml:56-72
Timestamp: 2025-08-26T09:49:04.956Z
Learning: In TensorRT-LLM test configuration files, the test scheduling system handles wildcard matching with special rules that prevent duplicate test execution even when the same tests appear in multiple yaml files with overlapping GPU wildcards (e.g., "*b200*" and "*gb200*").
📚 Learning: 2025-09-09T09:40:45.658Z
Learnt from: fredricz-20070104
Repo: NVIDIA/TensorRT-LLM PR: 7645
File: tests/integration/test_lists/qa/llm_function_core.txt:648-648
Timestamp: 2025-09-09T09:40:45.658Z
Learning: In TensorRT-LLM test lists, it's common and intentional for the same test to appear in multiple test list files when they serve different purposes (e.g., llm_function_core.txt for comprehensive core functionality testing and llm_function_core_sanity.txt for quick sanity checks). This duplication allows tests to be run in different testing contexts.
Applied to files:
tests/integration/test_lists/qa/llm_function_rtx6k.txttests/integration/test_lists/qa/llm_function_core.txttests/integration/defs/accuracy/test_llm_api_pytorch.py
📚 Learning: 2025-09-17T02:48:52.732Z
Learnt from: tongyuantongyu
Repo: NVIDIA/TensorRT-LLM PR: 7781
File: tests/integration/test_lists/waives.txt:313-313
Timestamp: 2025-09-17T02:48:52.732Z
Learning: In TensorRT-LLM, `tests/integration/test_lists/waives.txt` is specifically for waiving/skipping tests, while other test list files like those in `test-db/` and `qa/` directories are for different test execution contexts (pre-merge, post-merge, QA tests). The same test appearing in both waives.txt and execution list files is intentional - the test is part of test suites but will be skipped due to the waiver.
Applied to files:
tests/integration/test_lists/qa/llm_function_rtx6k.txttests/integration/test_lists/qa/llm_function_core.txt
📚 Learning: 2025-07-28T17:06:08.621Z
Learnt from: moraxu
Repo: NVIDIA/TensorRT-LLM 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.
Applied to files:
tests/integration/test_lists/qa/llm_function_rtx6k.txttests/integration/test_lists/qa/llm_function_core.txttests/integration/defs/accuracy/test_llm_api_pytorch.py
📚 Learning: 2025-11-27T09:23:18.742Z
Learnt from: fredricz-20070104
Repo: NVIDIA/TensorRT-LLM PR: 9511
File: tests/integration/defs/examples/serve/test_serve.py:136-186
Timestamp: 2025-11-27T09:23:18.742Z
Learning: In TensorRT-LLM testing, when adding test cases based on RCCA commands, the command format should be copied exactly as it appears in the RCCA case, even if it differs from existing tests. For example, some RCCA commands for trtllm-serve may omit the "serve" subcommand while others include it.
Applied to files:
tests/integration/test_lists/qa/llm_function_rtx6k.txttests/integration/test_lists/qa/llm_function_core.txt
📚 Learning: 2025-08-06T13:58:07.506Z
Learnt from: galagam
Repo: NVIDIA/TensorRT-LLM PR: 6487
File: tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py:1-12
Timestamp: 2025-08-06T13:58:07.506Z
Learning: In TensorRT-LLM, test files (files under tests/ directories) do not require NVIDIA copyright headers, unlike production source code files. Test files typically start directly with imports, docstrings, or code.
Applied to files:
tests/integration/test_lists/qa/llm_function_rtx6k.txt
📚 Learning: 2025-08-26T09:49:04.956Z
Learnt from: pengbowang-nv
Repo: NVIDIA/TensorRT-LLM PR: 7192
File: tests/integration/test_lists/test-db/l0_dgx_b200.yml:56-72
Timestamp: 2025-08-26T09:49:04.956Z
Learning: In TensorRT-LLM test configuration files, the test scheduling system handles wildcard matching with special rules that prevent duplicate test execution even when the same tests appear in multiple yaml files with overlapping GPU wildcards (e.g., "*b200*" and "*gb200*").
Applied to files:
tests/integration/test_lists/qa/llm_function_rtx6k.txttests/integration/test_lists/qa/llm_function_core.txttests/integration/defs/accuracy/test_llm_api_pytorch.py
⏰ 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
🔇 Additional comments (3)
tests/integration/test_lists/qa/llm_function_core.txt (1)
428-429: VSWA and Eagle3 test registrations look consistentThe added entries line up with the new test names and parametrization ids in
TestGemma3_1BInstructandTestGPTOSSand follow the existing list format. Nothing to change here.Also applies to: 594-597
tests/integration/defs/accuracy/references/json_mode_eval.yaml (1)
11-21: New JsonModeEval references match added testsThe new entries for
google/gemma-3-1b-itandGPT-OSS/120B-MXFP4are well‑formed and aligned with the model names used in the new JsonModeEval tests. No issues from a structure or naming perspective.tests/integration/test_lists/qa/llm_function_rtx6k.txt (1)
34-49: RTX6K Eagle3 coverage entries are correctly wiredThe newly added
TestGPTOSS::test_eagle3_*_4gpus[...]entries (including VSWA reuse and guided decoding variants) match the test function names and parametrization ids and follow the established pattern for this file. Looks good.
Signed-off-by: Ivy Zhang <[email protected]>
1 participant
The new cases are validated on CG4 and RTX6K.
Summary by CodeRabbit
✏️ Tip: You can customize this high-level summary in your review settings.
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)
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
/bot [-h] ['run', 'kill', 'skip', 'reuse-pipeline'] ...Provide a user friendly way for developers to interact with a Jenkins server.
Run
/bot [-h|--help]to print this help message.See details below for each supported subcommand.
Details
run [--reuse-test (optional)pipeline-id --disable-fail-fast --skip-test --stage-list "A10-PyTorch-1, xxx" --gpu-type "A30, H100_PCIe" --test-backend "pytorch, cpp" --add-multi-gpu-test --only-multi-gpu-test --disable-multi-gpu-test --post-merge --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" --detailed-log --debug(experimental)]Launch build/test pipelines. All previously running jobs will be killed.
--reuse-test (optional)pipeline-id(OPTIONAL) : Allow the new pipeline to reuse build artifacts and skip successful test stages from a specified pipeline or the last pipeline if no pipeline-id is indicated. If the Git commit ID has changed, this option will be always ignored. The DEFAULT behavior of the bot is to reuse build artifacts and successful test results from the last pipeline.--disable-reuse-test(OPTIONAL) : Explicitly prevent the pipeline from reusing build artifacts and skipping successful test stages from a previous pipeline. Ensure that all builds and tests are run regardless of previous successes.--disable-fail-fast(OPTIONAL) : Disable fail fast on build/tests/infra failures.--skip-test(OPTIONAL) : Skip all test stages, but still run build stages, package stages and sanity check stages. Note: Does NOT update GitHub check status.--stage-list "A10-PyTorch-1, xxx"(OPTIONAL) : Only run the specified test stages. Examples: "A10-PyTorch-1, xxx". Note: Does NOT update GitHub check status.--gpu-type "A30, H100_PCIe"(OPTIONAL) : Only run the test stages on the specified GPU types. Examples: "A30, H100_PCIe". Note: Does NOT update GitHub check status.--test-backend "pytorch, cpp"(OPTIONAL) : Skip test stages which don't match the specified backends. Only support [pytorch, cpp, tensorrt, triton]. Examples: "pytorch, cpp" (does not run test stages with tensorrt or triton backend). Note: Does NOT update GitHub pipeline status.--only-multi-gpu-test(OPTIONAL) : Only run the multi-GPU tests. Note: Does NOT update GitHub check status.--disable-multi-gpu-test(OPTIONAL) : Disable the multi-GPU tests. Note: Does NOT update GitHub check status.--add-multi-gpu-test(OPTIONAL) : Force run the multi-GPU tests in addition to running L0 pre-merge pipeline.--post-merge(OPTIONAL) : Run the L0 post-merge pipeline instead of the ordinary L0 pre-merge pipeline.--extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx"(OPTIONAL) : Run the ordinary L0 pre-merge pipeline and specified test stages. Examples: --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx".--detailed-log(OPTIONAL) : Enable flushing out all logs to the Jenkins console. This will significantly increase the log volume and may slow down the job.--debug(OPTIONAL) : Experimental feature. Enable access to the CI container for debugging purpose. Note: Specify exactly one stage in thestage-listparameter to access the appropriate container environment. Note: Does NOT update GitHub check status.For guidance on mapping tests to stage names, see
docs/source/reference/ci-overview.mdand the
scripts/test_to_stage_mapping.pyhelper.kill
killKill all running builds associated with pull request.
skip
skip --comment COMMENTSkip testing for latest commit on pull request.
--comment "Reason for skipping build/test"is required. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.reuse-pipeline
reuse-pipelineReuse a previous pipeline to validate current commit. This action will also kill all currently running builds associated with the pull request. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.