diff --git a/docs/contributing/profiling.md b/docs/contributing/profiling.md index 952b602aa..4e61d2531 100644 --- a/docs/contributing/profiling.md +++ b/docs/contributing/profiling.md @@ -22,7 +22,7 @@ profiler_config: Supported fields: | Field | Description | -|---|---| +| --- | --- | | `profiler` | Profiler backend. Supported values: `torch`, `cuda`. Use `torch` for `trace.json`, Excel operator tables, and optional memory snapshots. Use `cuda` for Nsight Systems only. | | `torch_profiler_dir` | Output directory for torch-profiler artifacts. Required when `profiler: torch`. | | `torch_profiler_use_gzip` | Compress `trace_rank*.json` into `trace_rank*.json.gz`. | @@ -360,7 +360,7 @@ If unset, the monitor writes `./vllm_omni_orch_monitor_.json` at shutd Each 1-second window records: | Series | Description | -|---|---| +| --- | --- | | `windows.duration_s` | Wall time covered by the window | | `windows.loop_idle` / `windows.loop_active` | Orchestrator poll-loop iterations with no work vs. active forwarding | | `windows.dispatch_queue_size` | Ready-queue size sampled at each monitor window boundary | @@ -388,7 +388,7 @@ before the window closes, hiding a transient ready-queue backlog in the latter. This monitor is intentionally separate from the existing profiling tools: | Tool | Scope | Output | -|---|---|---| +| --- | --- | --- | | `--enable-diffusion-pipeline-profiler` | Diffusion stage functions (`vae.decode`, `diffuse`, …) | Per-request `stage_durations` / logs | | `--enable-ar-profiler` | AR stage generation time | Per-request `stage_durations` | | `profiler_config` (`torch` / `cuda`) | GPU/CPU kernels inside a stage worker | `trace.json`, Nsight, operator tables | diff --git a/tests/engine/test_orchestrator_event_driven.py b/tests/engine/test_orchestrator_event_driven.py index abfa05933..109721446 100644 --- a/tests/engine/test_orchestrator_event_driven.py +++ b/tests/engine/test_orchestrator_event_driven.py @@ -1,5 +1,5 @@ # SPDX-License-Identifier: Apache-2.0 -# SPDX-FileCopyrightText: Copyright contributors to the vLLM project +# SPDX-FileCopyrightText: Copyright contributors to the vLLM-Omni project """Event-driven orchestration loop (``VLLM_OMNI_EVENT_DRIVEN_ORCH=1``) tests. Parity suite: re-runs the legacy orchestration scenarios from diff --git a/tests/engine/test_orchestrator_monitor.py b/tests/engine/test_orchestrator_monitor.py index 1961f0ddf..615bb0e89 100644 --- a/tests/engine/test_orchestrator_monitor.py +++ b/tests/engine/test_orchestrator_monitor.py @@ -1,3 +1,6 @@ +# SPDX-License-Identifier: Apache-2.0 +# SPDX-FileCopyrightText: Copyright contributors to the vLLM-Omni project + import json import time diff --git a/tests/worker/test_worker_base.py b/tests/worker/test_worker_base.py index 4570e6eda..a2779f937 100644 --- a/tests/worker/test_worker_base.py +++ b/tests/worker/test_worker_base.py @@ -1,5 +1,5 @@ # SPDX-License-Identifier: Apache-2.0 -# SPDX-FileCopyrightText: Copyright contributors to the vLLM project +# SPDX-FileCopyrightText: Copyright contributors to the vLLM-Omni project """Characterization tests for ``vllm_omni.worker.base.OmniGPUWorkerBase``. Pins the CURRENT behaviour of ``determine_available_memory`` (device-level diff --git a/vllm_omni/engine/orchestrator_monitor.py b/vllm_omni/engine/orchestrator_monitor.py index 132e55190..e18287762 100644 --- a/vllm_omni/engine/orchestrator_monitor.py +++ b/vllm_omni/engine/orchestrator_monitor.py @@ -1,3 +1,6 @@ +# SPDX-License-Identifier: Apache-2.0 +# SPDX-FileCopyrightText: Copyright contributors to the vLLM-Omni project + """Optional orchestrator window monitor (diagnostic only). Enable with: diff --git a/vllm_omni/worker/base.py b/vllm_omni/worker/base.py index 01a08bd72..312d8c828 100644 --- a/vllm_omni/worker/base.py +++ b/vllm_omni/worker/base.py @@ -1,3 +1,6 @@ +# SPDX-License-Identifier: Apache-2.0 +# SPDX-FileCopyrightText: Copyright contributors to the vLLM-Omni project + """Base worker class for vLLM-Omni with device-level GPU memory profiling.""" from __future__ import annotations