[rollout, perf, cfg] fix: Add global step info and support more profile control params for rollout profiling (sglang backend)

[bithighrr]

What does this PR do?

Address key functionality gaps in rollout discrete profiling for the sglang backend by adding global step awareness and expanding support for flexible profile control parameters:

1. Missing global step information resulted in disorganized profiling files;
2. Lack of support for specifying critical sglang backend profile control parameters (including num_steps, profile_by_stage, and merge_profiles) led to overly large profile files and hindered convenient profiling analysis.

Solutions implemented:

• Pass kwargs parameters to the rollout server to enable global step awareness, and generate an independent folder for each global step to organize profiling data in a structured way;
• Extend the profiling config by adding optional parameters to the content field, and remove unnecessary restrictions on content values in TorchProfilerToolConfig to support specifying sglang-specific profile control parameters (e.g., profile_by_stage, merge_profiles).

Checklist Before Starting

• Search for similar PRs. Paste at least one query link here: PR #4320: [perf] feat: verl profiler system support Agent Loop scenario and integrate torch.profiler
• Format the PR title as [{modules}] {type}: {description} (This will be checked by the CI)
  • {modules} include fsdp, megatron, veomni, sglang, vllm, rollout, trainer, ci, training_utils, recipe, hardware, deployment, ray, worker, single_controller, misc, perf, model, algo, env, tool, ckpt, doc, data, cfg, reward
  • If this PR involves multiple modules, separate them with , like [megatron, fsdp, doc]
  • {type} is in feat, fix, refactor, chore, test
  • If this PR breaks any API (CLI arguments, config, function signature, etc.), add [BREAKING] to the beginning of the title.
  • Example: [BREAKING][fsdp, megatron] feat: dynamic batching

Test

To validate the functionality of each profile control parameter for the sglang rollout backend, orthogonal test cases were designed (minimizing redundant combinations while covering all key parameter values). The test results are as follows:

step_start	step_end	profile-by-stage	merge-profiles	stack	shapes	cpu/cuda	Test Result
default	default	default (False)	default (False)	default (False)	default (False)	default	Pass
0 (≥0)	1 (≥1)	default	default	default	default	default	Pass
0 (≥0)	default	default	default	default	default	default	Pass
default	1 (≥1)	default	default	default	default	default	Pass
default	default	True	default	default	default	default	Pass
default	default	default	True	default	default	default	Pass
default	default	default	default	True	default	default	Pass
default	default	default	default	default	True	default	Pass
default	default	default	default	default	default	[] / [cpu] / [cuda] / [cpu,cuda]	Pass

API and Usage Example

  rollout:
    quantization: null
    profiler:
      enable: True
      all_ranks: False
      ranks: [0, 1, 2] 
      tool_config:
        torch:
          discrete: True
          step_start: 0
          step_end: 1
          contents: [cpu, cuda, stack, shapes, profile_by_stage, merge_profiles]

Design & Code Changes

Demonstrate the high-level design if this PR is complex, and list the specific changes.

Checklist Before Submitting

Important

Please check all the following items before requesting a review, otherwise the reviewer might deprioritize this PR for review.

• Read the Contribute Guide.
• Apply pre-commit checks: pre-commit install && pre-commit run --all-files --show-diff-on-failure --color=always
• Add / Update the documentation.
• Add unit or end-to-end test(s) to the CI workflow to cover all the code. If not feasible, explain why: ...
• Once your PR is ready for CI, send a message in the ci-request channel in the verl Slack workspace. (If not accessible, please try the Feishu group (飞书群).)
• If your PR is related to the recipe submodule, please also update the reference to the submodule commit via git submodule update --remote or cd recipe && git pull origin main.

[gemini-code-assist]

Code Review

This pull request introduces global step awareness for rollout profiling and expands support for sglang-specific profile control parameters. The changes correctly integrate global_step into the profiling process and add new configuration options for TorchProfilerToolConfig. However, there are a couple of high-severity issues related to the profiling logic in async_sglang_server.py that need to be addressed to ensure correct and predictable behavior.

[bithighrr]

The 'Repo id error' in NPU unit tests appears to be unrelated to this PR, could you please verify?

[bithighrr]

Local Test Status Summary

Test Suite	Commit 74f3224	PR c68dd6c
test_reward_model_manager	Failed (same root cause)	Failed (same root cause)
e2e_ascend/E2E_Ascend_testing_for_RL_training_scenarios_of_LLM_models	Passed	Passed
e2e_ppo_trainer_megatron_vllm_2/e2e_ppo_trainer_fsdp_vllm(run_function_reward.sh)	Passed	Passed

Can you check this for me? Thanks!

[tardis-key]

lgtm

[bithighrr]

The test case e2e_ascend/E2E Ascend testing for RL training scenarios of LLM models nearly failed during the execution of other CI jobs.
The test case reward_model_sglang passed in the previous CI job of this PR.

The failure of these two test cases is likely due to the impact of other CI jobs. A potential solution would be to increase the container shared memory (shm-size) for both test cases.