[None][test] add kv cache size in bench metric and fix failed cases #7211
Conversation
Signed-off-by: ruodil <[email protected]>
📝 WalkthroughWalkthroughAdds a new KV cache size performance metric to perf tests: enum extension, log parsing regexes for normal and bench runs, threshold configuration, and inclusion in bench metrics. Updates QA test lists: removes/adjusts several heavy tests, adds new 4-GPU variants, renames/normalizes some test names, and reorders entries. Changes
Sequence Diagram(s)sequenceDiagram
autonumber
participant T as Perf Test Runner
participant S as SUT (Model/Bench)
participant L as Log Stream
participant P as Perf Parser
participant E as Evaluator
T->>S: Start perf run (normal or bench)
S-->>L: Emit logs (throughput, latency, "Allocated X GiB for max tokens in paged KV cache", etc.)
T->>P: Provide log lines
P->>P: Match regexes for metrics incl. KV_CACHE_SIZE
P-->>T: Parsed metric values
T->>E: Compare metrics vs thresholds (incl. KV_CACHE_SIZE: rel=-0.1, abs=50)
E-->>T: Pass/Fail per metric
T-->>T: Aggregate and report results
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~25 minutes Possibly related PRs
Suggested reviewers
Tip 🔌 Remote MCP (Model Context Protocol) integration is now available!Pro plan users can now connect to remote MCP servers from the Integrations page. Connect with popular remote MCPs such as Notion and Linear to add more context to your reviews and chats. ✨ Finishing Touches
🧪 Generate unit tests
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. 🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
SupportNeed help? Create a ticket on our support page for assistance with any issues or questions. CodeRabbit Commands (Invoked using PR/Issue comments)Type Other keywords and placeholders
Status, Documentation and Community
|
|
/bot run |
|
PR_Github #16431 [ run ] triggered by Bot |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (1)
tests/integration/defs/perf/test_perf.py (1)
269-271: Broaden KV cache size regex to match both GiB and MiBTo prevent false negatives when parsing bench or integration logs that may emit “MiB” instead of “GiB” for the KV cache metric, update both regex mappings in tests/integration/defs/perf/test_perf.py.
Files to update:
- BENCH_PERF_METRIC_LOG_QUERIES mapping (around line 269)
- PERF_METRIC_LOG_QUERIES mapping (around line 255)
Apply this unified diff:
--- a/tests/integration/defs/perf/test_perf.py +++ b/tests/integration/defs/perf/test_perf.py @@ BENCH_PERF_METRIC_LOG_QUERIES @@ - PerfMetricType.KV_CACHE_SIZE: - re.compile(r".*Allocated ([\d\.]+) GiB for max tokens in paged KV cache.*"), + PerfMetricType.KV_CACHE_SIZE: + re.compile(r".*Allocated ([\d\.]+) (?:GiB|MiB) for max tokens in paged KV cache.*"), @@ PERF_METRIC_LOG_QUERIES @@ - PerfMetricType.KV_CACHE_SIZE: re.compile(r".*Allocated ([\d\.]+) GiB for max tokens in paged KV cache.*"), + PerfMetricType.KV_CACHE_SIZE: re.compile(r".*Allocated ([\d\.]+) (?:GiB|MiB) for max tokens in paged KV cache.*"),This ensures both GiB- and MiB-based outputs are correctly captured.
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
💡 Knowledge Base configuration:
- MCP integration is disabled by default for public repositories
- Jira integration is disabled by default for public repositories
- Linear integration is disabled by default for public repositories
You can enable these sources in your CodeRabbit configuration.
📒 Files selected for processing (3)
tests/integration/defs/perf/test_perf.py(2 hunks)tests/integration/test_lists/qa/llm_perf_cluster.yml(2 hunks)tests/integration/test_lists/qa/llm_perf_full.yml(3 hunks)
🧰 Additional context used
🧠 Learnings (1)
📚 Learning: 2025-07-28T17:06:08.621Z
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.
Applied to files:
tests/integration/test_lists/qa/llm_perf_full.ymltests/integration/test_lists/qa/llm_perf_cluster.yml
🧬 Code graph analysis (1)
tests/integration/defs/perf/test_perf.py (1)
tests/integration/defs/perf/utils.py (1)
PerfMetricType(84-101)
⏰ 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 (10)
tests/integration/test_lists/qa/llm_perf_full.yml (3)
77-78: TIMEOUT annotations for heavy 20k-token Mistral-Small tests look reasonableAdding TIMEOUT(120) to the 20k maxnt cases is appropriate to keep CI predictable on 1-GPU boxes.
182-183: Correct: remove quant for bench-pytorch variants
PerfTestConfig.validate()forbidsquant:whenbackend == "pytorch". Droppingquant:fp8avoids an assertion at runtime. Good fix.
265-266: Rename tollama_v3.3_70b_instructaligns with MODEL_PATH_DICTThis matches the keys present in
MODEL_PATH_DICTand avoids lookup failures during test generation.tests/integration/test_lists/qa/llm_perf_cluster.yml (7)
48-48: Reduce reqs from 500 to 300 for 24B/20k/4096 case: good stabilizationThis trims runtime for an extremely heavy configuration while preserving coverage.
67-67: Mirror req reduction for the 2-GPU variant: goodKeeps the 2-GPU run practical for CI.
78-78: Add TIMEOUT(120) to deepseek_r1_nvfp4 4-GPU heavy testReasonable safeguard for cluster queues.
80-80: Add TIMEOUT(120) to deepseek_r1_nvfp4 4-GPU maxbs/maxnt caseSame here; appropriate for CI stability.
86-86: TIMEOUT(120) on 405B FP4 4-GPU 20k tokensMakes sense given the workload magnitude.
90-91: Adjust 70B FP4 4-GPU reqs to 1000Reasonable reduction for runtime/manageability without losing perf signal.
94-96: Add 4-GPU Llama v4 Scout FP4 variants (including 20k TIMEOUT): good coverageThese round out the 4-GPU FP4 matrix while bounding runtime.
|
PR_Github #16431 [ run ] completed with state |
…7211) Signed-off-by: ruodil <[email protected]>
…VIDIA#7211) Signed-off-by: ruodil <[email protected]> Signed-off-by: Wangshanshan <[email protected]>
…VIDIA#7211) Signed-off-by: ruodil <[email protected]> Signed-off-by: Wangshanshan <[email protected]>
…VIDIA#7211) Signed-off-by: ruodil <[email protected]> Signed-off-by: Wangshanshan <[email protected]>
…VIDIA#7211) Signed-off-by: ruodil <[email protected]> Signed-off-by: Wangshanshan <[email protected]>
…VIDIA#7211) Signed-off-by: ruodil <[email protected]> Signed-off-by: Wangshanshan <[email protected]>
3 participants
Summary by CodeRabbit
New Features
Tests
Description
Test Coverage
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.
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.