[None][test] Test trtllm-bench AD vs, PT BEs on H100 single gpu #6487
Conversation
📝 WalkthroughWalkthroughThe changes update test configurations for A30, B200, and H100 GPU systems by excluding the Changes
Sequence Diagram(s)sequenceDiagram
participant TestRunner as Pytest
participant Benchmark as run_benchmark()
participant Subprocess as trtllm_bench subprocess
participant Parser as parse_kv_cache_metrics()
participant Validator as validate_kv_cache_metrics_dynamic()
participant Comparator as compare_backends_performance()
TestRunner->>Benchmark: Invoke run_benchmark() for backend A
Benchmark->>Subprocess: Run trtllm_bench with backend A
Subprocess-->>Benchmark: Return stdout/stderr logs
Benchmark->>Parser: Parse KV cache metrics from logs
Parser-->>Benchmark: Return metrics
Benchmark->>Validator: Validate metrics against expectations
Validator-->>Benchmark: Validation result
TestRunner->>Benchmark: Invoke run_benchmark() for backend B
Benchmark->>Subprocess: Run trtllm_bench with backend B
Subprocess-->>Benchmark: Return stdout/stderr logs
Benchmark->>Parser: Parse KV cache metrics from logs
Parser-->>Benchmark: Return metrics
Benchmark->>Validator: Validate metrics against expectations
Validator-->>Benchmark: Validation result
TestRunner->>Comparator: Compare backend A vs backend B performance
Comparator-->>TestRunner: Assert within tolerance or fail
Estimated code review effort🎯 4 (Complex) | ⏱️ ~45 minutes Suggested reviewers
Note ⚡️ Unit Test Generation is now available in beta!Learn more here, or try it out under "Finishing Touches" below. 📜 Recent review detailsConfiguration used: .coderabbit.yaml 📒 Files selected for processing (4)
✅ Files skipped from review due to trivial changes (1)
🚧 Files skipped from review as they are similar to previous changes (2)
🧰 Additional context used📓 Path-based instructions (2)**/*.py📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
Files:
**/*.{cpp,h,hpp,cc,cxx,cu,py}📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
Files:
🧠 Learnings (6)📓 Common learnings📚 Learning: in tensorrt-llm testing, it's common to have both cli flow tests (test_cli_flow.py) and pytorch api ...Applied to files:
📚 Learning: in tensorrt-llm, test files (files under tests/ directories) do not require nvidia copyright headers...Applied to files:
📚 Learning: applies to **/*.{cpp,h,hpp,cc,cxx,cu,py} : all tensorrt-llm open source software code should contain...Applied to files:
📚 Learning: applies to **/*.{h,hpp} : use a preprocessor guard in header files. the guard name must have prefix ...Applied to files:
📚 Learning: in tensorrt-llm, examples directory can have different dependency versions than the root requirement...Applied to files:
⏰ 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)
🔇 Additional comments (7)
✨ 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 comments)
Other keywords and placeholders
Documentation and Community
|
coderabbitai
bot
requested review from
kaiyux,
litaotju,
nv-guomingz and
yilin-void
MrGeva
force-pushed
the
user/egeva/h100_test
branch
from
c668f54 to
c4a11d5
Compare
|
/bot run |
|
PR_Github #13555 [ 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: 4
🧹 Nitpick comments (3)
tests/integration/test_lists/test-db/l0_a30.yml (1)
21-21: Document the reason for test exclusion on A30 GPUs.The exclusion of
test_trtllm_bench_backend_comparisonfrom A30 runs should be documented to explain why this test is not suitable for A30 hardware. Consider adding a comment explaining whether this is due to:
- H100-specific performance baselines
- Hardware capability requirements
- Resource constraints
# ------------- PyTorch tests --------------- - unittest/_torch/modeling -k "modeling_nemotron_nas" - unittest/_torch/modeling -k "modeling_phi3" - unittest/_torch/modeling -k "modeling_qwen" - unittest/_torch/modeling -k "modeling_qwen_moe" + # test_trtllm_bench_backend_comparison excluded: H100-specific performance baselines - unittest/_torch/auto_deploy/unit/singlegpu -k "not test_trtllm_bench_backend_comparison" - unittest/_torch/test_beam_search.pytests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py (2)
94-96: Document the use of environment variable for PyTorch backend.The
TLLM_OVERRIDE_LAYER_NUMenvironment variable is used specifically for the PyTorch backend, but its purpose and effect are not documented.if backend == "pytorch": + # Override the number of hidden layers for PyTorch backend testing + # This allows testing with a reduced model size for faster execution env["TLLM_OVERRIDE_LAYER_NUM"] = str(num_hidden_layers) print(f"📋 Using TLLM_OVERRIDE_LAYER_NUM from env: {env['TLLM_OVERRIDE_LAYER_NUM']}") cmd.extend(["--kv_cache_free_gpu_mem_fraction", str(free_mem_ratio)])
448-582: Consider breaking down this complex function.This function is quite long (130+ lines) and handles multiple responsibilities. Consider extracting helper functions for better maintainability.
Consider extracting these logical blocks into separate functions:
- Config file creation (lines 482-492)
- Backend comparison logic (lines 518-551)
- Golden comparison logic (lines 552-578)
This would improve readability and make the code easier to test and maintain.
📜 Review details
Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (3)
tests/integration/test_lists/test-db/l0_a30.yml(1 hunks)tests/integration/test_lists/test-db/l0_h100.yml(1 hunks)tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py(4 hunks)
🧰 Additional context used
📓 Path-based instructions (2)
**/*.py
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
**/*.py: The 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 filenames should use snake_case (e.g., some_file.py).
Python classes should use PascalCase (e.g., class SomeClass).
Python functions and methods should use snake_case (e.g., def my_awesome_function():).
Python local variables should use snake_case. Prefix k for variable names that start with a number (e.g., k_99th_percentile).
Python global variables should use upper snake_case and prefix G (e.g., G_MY_GLOBAL).
Python constants should use upper snake_case (e.g., 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 interfaces that may be used outside a file, prefer docstrings over comments in Python.
Comments in Python should be reserved for code within a function, or interfaces that are local to a file.
Use Google style docstrings for Python classes and functions, which can be parsed by Sphinx.
Attributes and variables in Python can be documented inline; attribute docstrings will be rendered under the docstring for the class.
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 to handle multiple possible variable types in Python, keep the body of the try as small as possible, using the else block to implement the logic.
Files:
tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py
**/*.{cpp,h,cu,py}
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
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.
Files:
tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py
🧠 Learnings (3)
tests/integration/test_lists/test-db/l0_h100.yml (1)
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.
tests/integration/test_lists/test-db/l0_a30.yml (1)
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.
tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py (1)
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.
⏰ 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 (1)
tests/integration/test_lists/test-db/l0_h100.yml (1)
18-18: LGTM!The addition of
test_trtllm_bench_backend_comparisonto the H100 test suite is appropriate and aligns with the PR objectives for H100-specific testing.
tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py
Show resolved
Hide resolved
tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py
Show resolved
Hide resolved
tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py
Outdated
Show resolved
Hide resolved
tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py
Show resolved
Hide resolved
|
PR_Github #13555 [ run ] completed with state |
|
/bot run --reuse-test --test-backend "pytorch" --disable-multi-gpu-test |
|
PR_Github #13572 [ run ] triggered by Bot |
|
PR_Github #13572 [ run ] completed with state |
|
/bot run |
coderabbitai
bot
requested review from
LarryXFly,
chzblych,
niukuo and
pamelap-nvidia
There was a problem hiding this comment.
Actionable comments posted: 0
♻️ Duplicate comments (4)
tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py (4)
14-49: Replace print statements with proper logging.The function uses print statements for debugging output. Consider using Python's logging module for better control over log levels and output formatting.
91-92: Move import statement to module level.Import statements should be at the module level unless there's a specific reason for lazy loading.
99-105: Add timeout protection for subprocess execution.The subprocess run could hang indefinitely. Consider adding a timeout to prevent test suite hangs.
278-284: Parameterize or document hardcoded memory values.The function contains hardcoded values for model size and extra consumption that should be parameterized or at least better documented with their derivation.
🧹 Nitpick comments (1)
tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py (1)
449-583: Consider breaking down this large function for better maintainability.This function handles multiple responsibilities including configuration setup, benchmark execution, and result comparison. Consider extracting the mode-specific logic into separate helper functions to improve readability and testability.
For example:
def _run_backend_comparison(autodeploy_report, model_name, dataset_path, temp_dir, ...): """Handle backend comparison mode logic.""" # Backend-specific logic here def _run_golden_comparison(autodeploy_report, kv_cache_metrics, expected_metrics, ...): """Handle golden comparison mode logic.""" # Golden-specific logic here
📜 Review details
Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (2)
tests/integration/test_lists/test-db/l0_b200.yml(1 hunks)tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py(4 hunks)
🧰 Additional context used
📓 Path-based instructions (2)
**/*.py
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
**/*.py: The 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 filenames should use snake_case (e.g., some_file.py).
Python classes should use PascalCase (e.g., class SomeClass).
Python functions and methods should use snake_case (e.g., def my_awesome_function():).
Python local variables should use snake_case, and prefix k for variable names that start with a number (e.g., k_99th_percentile).
Python global variables should use upper snake_case and prefix G (e.g., G_MY_GLOBAL).
Python constants should use upper snake_case (e.g., 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 interfaces that may be used outside a file, prefer docstrings over comments in Python.
Comments in Python should be reserved for code within a function, or interfaces that are local to a file.
Use Google style docstrings for classes and functions in Python, which can be parsed by Sphinx.
Attributes and variables in Python can be documented inline; attribute docstrings will be rendered under the docstring for the class.
Avoid using reflection in Python when functionality can be easily achieved without it.
When using try-except blocks in Python, limit the except to the smallest set of errors possible.
When using try-except blocks to handle multiple possible variable types in Python, keep the body of the try as small as possible, using the else block to implement the logic.
Files:
tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py
**/*.{cpp,h,hpp,cc,cxx,cu,py}
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
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.
Files:
tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py
🧠 Learnings (2)
tests/integration/test_lists/test-db/l0_b200.yml (1)
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.
tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py (1)
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.
🔇 Additional comments (1)
tests/integration/test_lists/test-db/l0_b200.yml (1)
59-59: LGTM!The exclusion of
test_trtllm_bench_backend_comparisonfrom the B200 test suite is consistent with the selective test execution strategy across different GPU configurations.
|
PR_Github #13639 [ run ] triggered by Bot |
|
PR_Github #13639 [ run ] completed with state |
|
/bot run --stage-list "H100_PCIe-PyTorch-2" --reuse-test --disable-multi-gpu-test |
MrGeva
changed the title
|
PR_Github #13665 [ run ] triggered by Bot |
|
/bot run |
galagam
force-pushed
the
user/egeva/h100_test
branch
from
519900f to
39f9cac
Compare
Signed-off-by: Eran Geva <19514940+MrGeva@users.noreply.github.com> added memory checks Signed-off-by: Eran Geva <19514940+MrGeva@users.noreply.github.com> changed kv cache test to be hw agnostic Signed-off-by: Eran Geva <19514940+MrGeva@users.noreply.github.com> added test with backend comparison Signed-off-by: Eran Geva <19514940+MrGeva@users.noreply.github.com> both tests pass Signed-off-by: Eran Geva <19514940+MrGeva@users.noreply.github.com> cleanups and refactoring Signed-off-by: Eran Geva <19514940+MrGeva@users.noreply.github.com> fixed llm_root issue Signed-off-by: Eran Geva <19514940+MrGeva@users.noreply.github.com> shrunk the model, and fixes Signed-off-by: Eran Geva <19514940+MrGeva@users.noreply.github.com> fixed trtllm-bench test stability Signed-off-by: Eran Geva <19514940+MrGeva@users.noreply.github.com> preserved the old test Signed-off-by: Eran Geva <19514940+MrGeva@users.noreply.github.com> set mem ratio to 0.3, fixed bug in default params Signed-off-by: Eran Geva <19514940+MrGeva@users.noreply.github.com>
Signed-off-by: Eran Geva <19514940+MrGeva@users.noreply.github.com>
Signed-off-by: Eran Geva <19514940+MrGeva@users.noreply.github.com>
Signed-off-by: Eran Geva <19514940+MrGeva@users.noreply.github.com>
Signed-off-by: Eran Geva <19514940+MrGeva@users.noreply.github.com>
Signed-off-by: Gal Hubara Agam <96368689+galagam@users.noreply.github.com>
galagam
force-pushed
the
user/egeva/h100_test
branch
from
d552bfe to
806394c
Compare
|
/bot run |
|
PR_Github #14712 [ run ] triggered by Bot |
|
PR_Github #14712 [ run ] completed with state |
5 participants
Summary by CodeRabbit
New Features
Refactor
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.