[Bugfix][Spec Decode] Avoid fastsafetensors deadlock for PP draft models (#54416)

Signed-off-by: Canlin <canlinguosdu@gmail.com>
Co-authored-by: Codex <noreply@openai.com>
This commit is contained in:
Canlin Guo
2026-09-12 14:22:30 +08:00
committed by GitHub
co-authored by Codex
parent eed1f3d0c6
commit dca96bf97b
6 changed files with 42 additions and 2 deletions
@@ -6,11 +6,13 @@
assert on the config ``load_eagle_model`` hands to ``get_model``.
"""
from dataclasses import dataclass
from dataclasses import dataclass, field
from types import SimpleNamespace
from unittest.mock import patch
import pytest
from vllm.config import LoadConfig
from vllm.v1.worker.gpu.spec_decode.eagle.utils import load_eagle_model
@@ -43,6 +45,7 @@ class _VllmConfig:
kernel_config: _KernelConfig
cache_config: _CacheConfig
speculative_config: _SpeculativeConfig
load_config: LoadConfig = field(default_factory=LoadConfig)
def _config(target_backend: str, draft_backend: str | None) -> _VllmConfig:
@@ -65,6 +68,10 @@ def _capture_draft_config(cfg):
with (
patch("vllm.v1.worker.gpu.spec_decode.eagle.utils.get_model", _fake_get_model),
patch(
"vllm.v1.worker.gpu.spec_decode.utils.get_pp_group",
return_value=SimpleNamespace(world_size=1),
),
pytest.raises(_Captured) as exc,
):
load_eagle_model(object(), cfg)
@@ -8,11 +8,13 @@ target is typically unquantized, and quantized-only backends reject it, so the
server fails to start rather than falling back.
"""
from dataclasses import dataclass
from dataclasses import dataclass, field
from types import SimpleNamespace
from unittest.mock import patch
import pytest
from vllm.config import LoadConfig
from vllm.v1.worker.gpu.spec_decode.eagle.utils import load_eagle_model
@@ -39,6 +41,7 @@ class _VllmConfig:
kernel_config: _KernelConfig
cache_config: _CacheConfig
speculative_config: _SpeculativeConfig
load_config: LoadConfig = field(default_factory=LoadConfig)
def _config(target_moe: str, draft_moe: str | None) -> _VllmConfig:
@@ -62,6 +65,10 @@ def _capture_draft_config(cfg):
with (
patch("vllm.v1.worker.gpu.spec_decode.eagle.utils.get_model", _fake_get_model),
patch(
"vllm.v1.worker.gpu.spec_decode.utils.get_pp_group",
return_value=SimpleNamespace(world_size=1),
),
pytest.raises(_Captured) as exc,
):
load_eagle_model(object(), cfg)