[https://nvbugs/5444937][chore] Fixing KV events tests #7004
Conversation
Signed-off-by: Patrice Castonguay <[email protected]>
📝 WalkthroughWalkthroughRe-enabled and adjusted kv-cache events tests: removed explicit pytest skips and refactored event validation to avoid fixed event_id checks, accumulate stored blocks, detect removal events across batches, and special-case attention-dataplane paths. No public API/signature changes. Changes
Estimated code review effort🎯 4 (Complex) | ⏱️ ~45 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 #15629 [ run ] triggered by Bot |
Signed-off-by: Patrice Castonguay <[email protected]>
pcastonguay
changed the title
|
/bot run --disable-fail-fast |
![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
🔭 Outside diff range comments (1)
tests/unittest/llmapi/test_llm_kv_cache_events.py (1)
1-3: Add NVIDIA copyright header to Python test filePer repo guidelines, prepend the standard NVIDIA header (current year) to all source files. For this test, use:
+# Copyright 2025, NVIDIA CORPORATION & AFFILIATES. All rights reserved. + import asyncio import time• tests/unittest/llmapi/test_llm_kv_cache_events.py: insert header above the first import.
🧹 Nitpick comments (3)
tests/unittest/llmapi/test_llm_kv_cache_events.py (3)
187-187: Avoid hard-coding total_stored_blocks == 5; make the assertion resilientThe exact number of blocks can depend on tokens-per-block, model config, and runtime heuristics. Asserting equality to 5 is brittle and may flake across environments.
Consider relaxing to a minimal guarantee, or compute the expectation from configuration if available:
- assert total_stored_blocks == 5 # Should have 5 blocks in total + assert total_stored_blocks > 0, f"Expected at least 1 stored block, got {total_stored_blocks}"If there is a stable way to infer expected blocks (e.g., from llm.kv_cache_config.tokens_per_block and request length), prefer computing it and asserting equality.
195-209: Second batch checks: relax block count equality; keep removed-event checkGreat to verify a "removed" event between batches. However, asserting total_stored_blocks == 5 is likely to be environment-dependent.
Apply:
- assert total_stored_blocks == 5 # Should have 5 blocks in total + assert total_stored_blocks > 0, f"Expected some stored blocks after second batch, got {total_stored_blocks}" assert has_removed_eventOptionally, gate the removed-event assertion on block reuse being enabled if that configuration can vary:
- Only assert has_removed_event if block reuse is enabled in the active kv_cache_config.
217-231: Third batch checks: same resilience for block countsMirror the relaxation applied to the second batch to avoid brittleness.
Apply:
- assert total_stored_blocks == 5 # Should have 5 blocks in total + assert total_stored_blocks > 0, f"Expected some stored blocks after third batch, got {total_stored_blocks}" assert has_removed_event
📜 Review details
Configuration used: .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 (1)
tests/unittest/llmapi/test_llm_kv_cache_events.py(1 hunks)
🧰 Additional context used
📓 Path-based instructions (2)
**/*.py
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
**/*.py: Python code must target Python 3.8+
Python indentation: 4 spaces, no tabs
Maintain module namespace in imports (from package.subpackage import foo; then use foo.SomeClass())
Python file names use snake_case
Python class names use PascalCase
Python functions/methods and local variables use snake_case; variables starting with a number get k_ prefix (e.g., k_99th_percentile)
Global variables use G_ prefixed UPPER_SNAKE_CASE (e.g., G_MY_GLOBAL)
Constants use UPPER_SNAKE_CASE in Python
Avoid shadowing variables from outer scopes in Python
Initialize all externally visible members of a Python class in init
Prefer docstrings for interfaces used outside a file; comments for local code
Use Google-style docstrings for classes and functions (Sphinx-parsable)
Document attributes/variables inline with short docstrings
Avoid reflection when simple alternatives exist (e.g., prefer explicit parameters over dict(**locals()))
In try/except, catch the narrowest exceptions possible
For duck-typing with try/except, keep try body minimal and put logic in else
Files:
tests/unittest/llmapi/test_llm_kv_cache_events.py
**/*.{cpp,cxx,cc,cu,h,hpp,hxx,hh,cuh,py}
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
Prepend NVIDIA copyright header (current year) to all source files
Files:
tests/unittest/llmapi/test_llm_kv_cache_events.py
🔇 Additional comments (1)
tests/unittest/llmapi/test_llm_kv_cache_events.py (1)
164-175: Good move away from brittle event_id == 1; accumulating stored blocks is robustReplacing the hard-coded equality with event_id > 0 for "stored" events and summing block counts improves resilience across implementations.
pcastonguay
changed the title
|
PR_Github #15649 [ run ] triggered by Bot |
|
PR_Github #15629 [ run ] completed with state |
Signed-off-by: Patrice Castonguay <[email protected]>
pcastonguay
changed the title
|
/bot run --disable-fail-fast |
There was a problem hiding this comment.
Actionable comments posted: 0
♻️ Duplicate comments (1)
tests/unittest/llmapi/test_llm_kv_cache_events.py (1)
176-182: Guard is only partial; prefer get() and avoid pop(0) O(n) patternYou added a None guard, but still assert the key and use direct indexing. Also, popping from the head in a loop is O(n) and unnecessary here.
Apply:
- while events: - event = events.pop(0) - if not event: - continue - assert "attention_dp_rank" in event - if event["attention_dp_rank"] == attention_dp_rank: + for event in events: + if not event: + continue + if event.get("attention_dp_rank") != attention_dp_rank: + continue
🧹 Nitpick comments (3)
tests/unittest/llmapi/test_llm_kv_cache_events.py (3)
169-175: Minor: iterate instead of popping for clarity and to avoid O(n^2) behaviorSame nit as above for the non-attention branch.
- while events: - event = events.pop(0) - if event: - assert event["data"]["type"] == "stored" - assert event["event_id"] > 0 - total_stored_blocks += len(event["data"]["blocks"]) + for event in events: + if not event: + continue + assert event["data"]["type"] == "stored" + assert event["event_id"] > 0 + total_stored_blocks += len(event["data"]["blocks"])
189-189: Avoid hard-coded “5” block count; derive from first batch to reduce flakinessHardcoding 5 ties the test to internal block sizing. Derive the expected count from the first batch, then assert subsequent batches match.
Apply:
- assert total_stored_blocks == 5 # Should have 5 blocks in total + expected_total_stored_blocks = total_stored_blocks- assert total_stored_blocks == 5 # Should have 5 blocks in total + assert total_stored_blocks == expected_total_stored_blocks- assert total_stored_blocks == 5 # Should have 5 blocks in total + assert total_stored_blocks == expected_total_stored_blocksAdditionally, initialize the variable at the top of check_events:
# inside check_events(...) near the top expected_total_stored_blocks = NoneAlso applies to: 210-210, 232-232
203-211: Add sanity check: stored events should have event_id > 0You relaxed this check earlier; keep it consistent in later batches too.
Apply:
elif event["data"]["type"] == "stored": + assert event["event_id"] > 0 total_stored_blocks += len(event["data"]["blocks"])Apply the same in the third batch loop:
elif event["data"]["type"] == "stored": + assert event["event_id"] > 0 total_stored_blocks += len(event["data"]["blocks"])Also applies to: 226-233
📜 Review details
Configuration used: .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 (1)
tests/unittest/llmapi/test_llm_kv_cache_events.py(1 hunks)
🧰 Additional context used
📓 Path-based instructions (2)
**/*.py
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
**/*.py: Python code must target Python 3.8+
Python indentation: 4 spaces, no tabs
Maintain module namespace in imports (from package.subpackage import foo; then use foo.SomeClass())
Python file names use snake_case
Python class names use PascalCase
Python functions/methods and local variables use snake_case; variables starting with a number get k_ prefix (e.g., k_99th_percentile)
Global variables use G_ prefixed UPPER_SNAKE_CASE (e.g., G_MY_GLOBAL)
Constants use UPPER_SNAKE_CASE in Python
Avoid shadowing variables from outer scopes in Python
Initialize all externally visible members of a Python class in init
Prefer docstrings for interfaces used outside a file; comments for local code
Use Google-style docstrings for classes and functions (Sphinx-parsable)
Document attributes/variables inline with short docstrings
Avoid reflection when simple alternatives exist (e.g., prefer explicit parameters over dict(**locals()))
In try/except, catch the narrowest exceptions possible
For duck-typing with try/except, keep try body minimal and put logic in else
Files:
tests/unittest/llmapi/test_llm_kv_cache_events.py
**/*.{cpp,cxx,cc,cu,h,hpp,hxx,hh,cuh,py}
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
Prepend NVIDIA copyright header (current year) to all source files
Files:
tests/unittest/llmapi/test_llm_kv_cache_events.py
🧬 Code Graph Analysis (1)
tests/unittest/llmapi/test_llm_kv_cache_events.py (3)
tensorrt_llm/llmapi/llm.py (1)
get_kv_cache_events(471-490)tensorrt_llm/serve/openai_server.py (1)
get_kv_cache_events(242-250)tensorrt_llm/llmapi/utils.py (2)
get(354-367)get(420-437)
⏰ 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 (2)
tests/unittest/llmapi/test_llm_kv_cache_events.py (2)
164-175: Good move: relax event_id check and aggregate stored blocksSwitching to event_id > 0 and summing stored blocks makes the test less brittle than pinning to a specific id. Looks correct for non-attention DP.
173-174: Fix remaining brittleevent_idassertion in executor bindings testOne instance of
assert events[0].event_id == 1still exists. Update it to a non‐strict check:
- File:
tests/unittest/bindings/test_executor_bindings.py, line 2023- Change:
- assert events[0].event_id == 1
- assert events[0].event_id > 0
> Likely an incorrect or invalid review comment. </details> </blockquote></details> </details> <!-- This is an auto-generated comment by CodeRabbit for review status -->
|
PR_Github #15651 [ run ] triggered by Bot |
|
PR_Github #15649 [ run ] completed with state |
|
PR_Github #15651 [ run ] completed with state |
|
/bot run |
|
PR_Github #15787 [ run ] triggered by Bot |
|
PR_Github #15787 [ run ] completed with state |
3 participants
Summary by CodeRabbit
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.