[CI] [ROCm] Fix tests/entrypoints/test_grpc_server.py on ROCm (#31970)

Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>

requirements/common.txt

@@ -53,4 +53,4 @@ anthropic == 0.71.0
 model-hosting-container-standards >= 0.1.10, < 1.0.0
 mcp
 grpcio>=1.76.0
-grpcio-reflection>=1.76.0
+grpcio-reflection>=1.76.0

requirements/rocm-test.txt

@@ -89,4 +89,4 @@ perceptron==0.1.4
 # Required for the multi-modal models test
 timm==1.0.17
 # Required for plugins test
-albumentations==1.4.6
+albumentations==1.4.6

requirements/rocm.txt

@@ -16,3 +16,4 @@ runai-model-streamer[s3,gcs]==0.15.3
 conch-triton-kernels==1.2.1
 timm>=1.0.17
 fastsafetensors @ git+https://github.com/foundation-model-stack/fastsafetensors.git@d6f998a03432b2452f8de2bb5cefb5af9795d459
+grpcio-tools>=1.76.0

setup.py

@@ -19,6 +19,7 @@ from packaging.version import Version, parse
 from setuptools import Extension, setup
 from setuptools.command.build_ext import build_ext
 from setuptools.command.build_py import build_py
+from setuptools.command.develop import develop
 from setuptools_scm import get_version
 from torch.utils.cpp_extension import CUDA_HOME, ROCM_HOME
 
@@ -147,6 +148,14 @@ class BuildPyAndGenerateGrpc(build_py):
         super().run()
 
 
+class DevelopAndGenerateGrpc(develop):
+    """Develop mode that also generates gRPC stubs from proto files."""
+
+    def run(self):
+        compile_grpc_protos()
+        super().run()
+
+
 class CMakeExtension(Extension):
     def __init__(self, name: str, cmake_lists_dir: str = ".", **kwa) -> None:
         super().__init__(name, sources=[], py_limited_api=not is_freethreaded(), **kwa)
@@ -950,13 +959,17 @@ if _no_device():
     ext_modules = []
 
 if not ext_modules:
-    cmdclass = {"build_py": BuildPyAndGenerateGrpc}
+    cmdclass = {
+        "build_py": BuildPyAndGenerateGrpc,
+        "develop": DevelopAndGenerateGrpc,
+    }
 else:
     cmdclass = {
         "build_ext": precompiled_build_ext
         if envs.VLLM_USE_PRECOMPILED
         else cmake_build_ext,
         "build_py": BuildPyAndGenerateGrpc,
+        "develop": DevelopAndGenerateGrpc,
     }
 
 setup(

tests/entrypoints/test_grpc_server.py

@@ -29,7 +29,7 @@ def find_free_port() -> int:
     return port
 
 
-async def wait_for_server(port: int, timeout: float = 30.0) -> bool:
+async def wait_for_server(port: int, timeout: float = 60.0) -> bool:
     """Wait for the gRPC server to be ready by trying health checks."""
     start_time = time.time()
     print("waiting for server to start...")