[None] [fix] Fix nsys in slurm scripts #7409
Conversation
Signed-off-by: Yiqing Yan <yiqingy@nvidia.com>
Signed-off-by: Kaiyu Xie <26294424+kaiyux@users.noreply.github.com>
kaiyux
requested review from
Shixiaowei02,
dongxuy04,
laikhtewari,
pcastonguay,
qiaoxj07 and
schetlur-nv
📝 WalkthroughWalkthroughNSYS profiling defaults were altered in Slurm benchmark scripts: Changes
Sequence Diagram(s)sequenceDiagram
autonumber
participant Slurm as Slurm Job (disaggr_torch.slurm)
participant Worker as start_worker.sh
participant GEN as GEN Process
participant CTX as CTX Process
participant NSYS as NSYS Wrapper
Slurm->>Worker: Launch worker with nsys_on=${full_logdir}
alt Role == GEN and SLURM_PROCID == 0
Worker->>NSYS: Prefix command (nsys profile ... -o ${full_logdir}/gen_0)
NSYS->>GEN: Execute trtllm-llmapi-launch
else Other roles or PROCIDs
Worker->>GEN: Execute trtllm-llmapi-launch (no NSYS)
Worker->>CTX: Execute trtllm-llmapi-launch (no NSYS)
end
note over Slurm,Worker: nsys_on now computed after full_logdir and non-empty
Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes Tip 🔌 Remote MCP (Model Context Protocol) integration is now available!Pro plan users can now connect to remote MCP servers from the Integrations page. Connect with popular remote MCPs such as Notion and Linear to add more context to your reviews and chats. ✨ Finishing Touches
🧪 Generate unit tests
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. 🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
SupportNeed help? Create a ticket on our support page for assistance with any issues or questions. CodeRabbit Commands (Invoked using PR/Issue comments)Type Other keywords and placeholders
Status, Documentation and Community
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 0
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
examples/disaggregated/slurm/benchmark/start_worker.sh (1)
59-79: Fix NSYS prefix quoting/invocation (SC2089/SC2090) and quote paths.The string-form prefix loses quoting on expansion and can break the
-evalue; use an array and quote paths. Also quote command args to avoid word-splitting.@@ -if [ -z "${nsys_folder:-}" ]; then +if [ -z "${nsys_folder:-}" ]; then echo "nsys is not enabled, start normal flow" - trtllm-llmapi-launch ${numa_bind_cmd} trtllm-serve ${model_path} --host $(hostname) --port ${port} --extra_llm_api_options ${config_file} + trtllm-llmapi-launch ${numa_bind_cmd} trtllm-serve "${model_path}" --host "$(hostname)" --port "${port}" --extra_llm_api_options "${config_file}" else - nsys_prefix="" - nsys_file=${nsys_folder}/nsys_worker_proc_${instance_id}_${SLURM_PROCID} + nsys_prefix=() + nsys_file="${nsys_folder}/nsys_worker_proc_${instance_id}_${SLURM_PROCID}" export TLLM_PROFILE_RECORD_GC=1 export TLLM_NVTX_DEBUG=1 - if [ "${role}" = "GEN" ] && [ "$SLURM_PROCID" = "0" ]; then - export TLLM_PROFILE_START_STOP=200-250 - nsys_prefix="nsys profile -e \"NSYS_MPI_STORE_TEAMS_PER_RANK=1\" -o ${nsys_file} -f true -t cuda,nvtx,python-gil -c cudaProfilerApi --cuda-graph-trace node --capture-range-end=stop --gpu-metrics-devices=none" - echo "nsys_prefix: ${nsys_prefix}" + if [ "${role}" = "GEN" ] && [ "${SLURM_PROCID}" = "0" ]; then + export TLLM_PROFILE_START_STOP="${TLLM_PROFILE_START_STOP:-200-250}" + nsys_prefix=(nsys profile -e "NSYS_MPI_STORE_TEAMS_PER_RANK=1" -o "${nsys_file}" -f true -t cuda,nvtx,python-gil -c cudaProfilerApi --cuda-graph-trace node --capture-range-end=stop --gpu-metrics-devices=none) + echo "nsys_prefix: ${nsys_prefix[*]}" elif [ "${role}" = "CTX" ]; then echo "nsys is not enabled on ctx_gpus" fi - ${nsys_prefix} trtllm-llmapi-launch ${numa_bind_cmd} \ - trtllm-serve ${model_path} \ - --host $(hostname) --port ${port} \ - --extra_llm_api_options ${config_file} + "${nsys_prefix[@]}" trtllm-llmapi-launch ${numa_bind_cmd} \ + trtllm-serve "${model_path}" \ + --host "$(hostname)" --port "${port}" \ + --extra_llm_api_options "${config_file}" fi
🧹 Nitpick comments (3)
examples/disaggregated/slurm/benchmark/start_worker.sh (1)
68-71: Make profiling iter-range configurable.Use env override with a sensible default instead of a hard-coded span.
- export TLLM_PROFILE_START_STOP=200-250 + export TLLM_PROFILE_START_STOP="${TLLM_PROFILE_START_STOP:-200-250}"tensorrt_llm/_torch/pyexecutor/py_executor.py (1)
496-502: Unify stop-path logging and ensure trace dir exists.Match the split-log style in both paths and create the parent directory before exporting the trace.
@@ - if enable_torch_trace: - torch_profiler.stop() - torch_profiler.export_chrome_trace(torch_trace_path) - logger.info(f"Profiling stopped at iteration {it}, " - f"trace saved to {torch_trace_path}") - torch.cuda.cudart().cudaProfilerStop() + if enable_torch_trace: + torch_profiler.stop() + # Ensure parent directory exists + trace_dir = os.path.dirname(torch_trace_path) or "." + os.makedirs(trace_dir, exist_ok=True) + torch_profiler.export_chrome_trace(torch_trace_path) + logger.info(f"Torch profiler trace saved to {torch_trace_path}.") + torch.cuda.cudart().cudaProfilerStop() + logger.info(f"Profiling stopped at iteration {it}.") @@ - if enable_torch_trace: - torch_profiler.stop() - torch_profiler.export_chrome_trace(torch_trace_path) - logger.info(f"Torch profiler trace saved to {torch_trace_path}.") + if enable_torch_trace: + torch_profiler.stop() + trace_dir = os.path.dirname(torch_trace_path) or "." + os.makedirs(trace_dir, exist_ok=True) + torch_profiler.export_chrome_trace(torch_trace_path) + logger.info(f"Torch profiler trace saved to {torch_trace_path}.")Also applies to: 561-566
examples/disaggregated/slurm/benchmark/disaggr_torch.slurm (1)
97-99: Fix misleading comment; clarify default.Comment says “Uncomment to enable” but the line is already enabled.
-# nsys_on="" -nsys_on=${full_logdir} # Uncomment this line to enable Nsys profiling +# To disable NSYS profiling, set: nsys_on="" +nsys_on=${full_logdir} # NSYS profiling enabled by default; set empty to disable
📜 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)
examples/disaggregated/slurm/benchmark/disaggr_torch.slurm(1 hunks)examples/disaggregated/slurm/benchmark/start_worker.sh(1 hunks)tensorrt_llm/_torch/pyexecutor/py_executor.py(1 hunks)
🧰 Additional context used
📓 Path-based instructions (2)
**/*.{cpp,cc,cxx,cu,h,hpp,hh,hxx,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
**/*.{cpp,cc,cxx,cu,h,hpp,hh,hxx,cuh,py}: Use spaces only; no tabs; indent with 4 spaces
Prepend NVIDIA copyright header (current year) to all source files (.cpp, .h, .cu, .py, etc.)
Files:
tensorrt_llm/_torch/pyexecutor/py_executor.py
**/*.py
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
**/*.py: Python code must target Python 3.8+
Indent Python with 4 spaces; no tabs
Preserve module namespaces when importing: from package.subpackage import foo; then call foo.SomeClass() instead of importing the class directly
Python naming: files snake_case; classes PascalCase; functions/methods snake_case; locals snake_case (prefix k_ when starting with a number); globals UPPER_SNAKE_CASE with G_ prefix; constants UPPER_SNAKE_CASE
Avoid shadowing outer-scope variables; initialize all externally visible members in init
Prefer docstrings for interfaces used outside a file; limit comments to function-internal or file-local interfaces
Use Google-style docstrings for classes and functions; document attributes/variables inline so Sphinx can render them
Avoid reflection when simpler alternatives exist; prefer explicit parameters and return dicts over locals()/dynamic tricks
In try/except, catch the narrowest exceptions possible; keep try bodies minimal and use else for the main logic when doing duck-typing checks
Files:
tensorrt_llm/_torch/pyexecutor/py_executor.py
🪛 Shellcheck (0.10.0)
examples/disaggregated/slurm/benchmark/start_worker.sh
[warning] 70-70: Quotes/backslashes will be treated literally. Use an array.
(SC2089)
[warning] 75-75: Quotes/backslashes in this variable will not be respected.
(SC2090)
⏰ 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)
examples/disaggregated/slurm/benchmark/start_worker.sh (1)
68-74: Restricting NSYS to GEN rank0 looks good.Helps keep overhead low while still capturing representative traces.
tensorrt_llm/_torch/pyexecutor/py_executor.py (1)
564-567: Clearer, separate shutdown logs.Splitting the messages improves readability during teardown.
Signed-off-by: Kaiyu Xie <26294424+kaiyux@users.noreply.github.com>
This reverts commit ec595a8.
|
/bot skip --comment "Slurm scripts are not protected by CI pipelines" |
|
/bot skip --comment "Slurm scripts are not protected by CI pipelines" |
|
PR_Github #17168 [ skip ] triggered by Bot |
|
PR_Github #17169 [ skip ] triggered by Bot |
|
PR_Github #17168 [ skip ] completed with state |
|
PR_Github #17169 [ skip ] completed with state |
Signed-off-by: Kaiyu Xie <26294424+kaiyux@users.noreply.github.com>
6 participants
Summary by CodeRabbit
New Features
Chores
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
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.
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.