rebase: align vllm-omni with vLLM d2906cc19586

Signed-off-by: tzhouam <tzhouam@connect.ust.hk>

docker/Dockerfile.ci

@@ -1,4 +1,4 @@
-# Target vllm version: releases/v0.29.0 @ f5c3cc240bc1, installed on top of
+# Target vllm version: releases/v0.29.0 @ d2906cc19586, installed on top of
 # the v0.28.0 base image via the precompiled wheel pinned below. The pin and
 # base intentionally differ until an official v0.29.0 base image is published.
 #
@@ -57,7 +57,7 @@ RUN uv pip install --system -e ".[dev]"
 # (e.g. +gaeee7ef93.cu130) which pip/uv refuse to install from a PEP 503 package index.
 # The cu130 wheel index lives at /<commit>/cu130/vllm/ but the actual wheel files
 # are stored at the top level /<commit>/<wheel>.
-ENV VLLM_PRECOMPILED_WHEEL_COMMIT=f5c3cc240bc1fc0519694fe689edb87dee8dfd92
+ENV VLLM_PRECOMPILED_WHEEL_COMMIT=d2906cc1958658f296aeee8b248deea684f56add
 # Keep the base image's CUDA toolkit libs on the loader path. This does NOT by
 # itself guarantee libnvJitLink.so.13 resolves — the .so.13 JIT linker is
 # handled separately via ldconfig after the cu130 wheels are installed (see the

tests/platforms/test_platform_detection.py

@@ -0,0 +1,38 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM-Omni project
+
+import logging
+
+import pytest
+
+import vllm_omni.platforms as platforms
+
+pytestmark = [pytest.mark.core_model, pytest.mark.cpu]
+
+
+def test_failing_oot_plugin_is_logged_and_falls_back(monkeypatch, caplog):
+    plugin_name = "raising_oot_plugin"
+
+    def raising_plugin():
+        raise RuntimeError("injected platform detection failure")
+
+    monkeypatch.setattr(platforms, "builtin_omni_platform_plugins", {})
+    monkeypatch.setattr(
+        platforms,
+        "load_omni_plugins_by_group",
+        lambda _group: {plugin_name: raising_plugin},
+    )
+
+    target_logger = logging.getLogger(platforms.__name__)
+    target_logger.addHandler(caplog.handler)
+    previous_level = target_logger.level
+    target_logger.setLevel(logging.DEBUG)
+    try:
+        platform_cls = platforms.resolve_current_omni_platform_cls_qualname()
+    finally:
+        target_logger.removeHandler(caplog.handler)
+        target_logger.setLevel(previous_level)
+
+    assert platform_cls == "vllm_omni.platforms.interface.UnspecifiedOmniPlatform"
+    assert plugin_name in caplog.text
+    assert "RuntimeError: injected platform detection failure" in caplog.text

vllm_omni/engine/arg_utils.py

@@ -443,8 +443,8 @@ class OmniEngineArgs(EngineArgs):
 @dataclass
 class OmniAsyncEngineArgs(AsyncEngineArgs, OmniEngineArgs):
     @classmethod
-    def add_cli_args(cls, parser: argparse.ArgumentParser) -> argparse.ArgumentParser:
-        parser = AsyncEngineArgs.add_cli_args(parser)
+    def add_cli_args(cls, parser: argparse.ArgumentParser, async_args_only: bool = False) -> argparse.ArgumentParser:
+        parser = AsyncEngineArgs.add_cli_args(parser, async_args_only=async_args_only)
         parser = OmniEngineArgs._add_omni_specific_args(parser)
         return parser
 

vllm_omni/platforms/__init__.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
 
 import logging
 import traceback
@@ -142,7 +142,11 @@ def resolve_current_omni_platform_cls_qualname() -> str:
             if platform_cls_qualname is not None:
                 activated_plugins.append(name)
         except Exception:
-            pass
+            logger.debug(
+                "OmniPlatform plugin %s failed during detection.",
+                name,
+                exc_info=True,
+            )
 
     activated_builtin_plugins = list(set(activated_plugins) & set(builtin_omni_platform_plugins.keys()))
     activated_oot_plugins = list(set(activated_plugins) & set(platform_plugins.keys()))