[1/N] Elastic EP Milestone 2 (#34861)

Signed-off-by: Yongji Wu <wuyongji317@gmail.com>
Signed-off-by: Itay Alroy <ialroy@nvidia.com>
Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>
Signed-off-by: Ron Tourgeman <rtourgeman@nvidia.com>
Co-authored-by: Yongji Wu <wuyongji317@gmail.com>
Co-authored-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>
Co-authored-by: Ron Tourgeman <rtourgeman@nvidia.com>
This commit is contained in:
Itay Alroy
2026-02-28 04:46:42 +00:00
committed by GitHub
co-authored by Yongji Wu Tyler Michael Smith Ron Tourgeman
parent 90805ff464
commit dea268336f
53 changed files with 3612 additions and 1015 deletions
+5 -2
View File
@@ -789,8 +789,11 @@ def test_hybrid_attention_mamba_tensor_shapes():
"MASTER_PORT": "12345",
}
)
init_distributed_environment()
initialize_model_parallel(tensor_model_parallel_size=1)
from tests.utils import ensure_current_vllm_config
with ensure_current_vllm_config():
init_distributed_environment()
initialize_model_parallel(tensor_model_parallel_size=1)
torch.set_default_dtype(torch.float16)
model_config = ModelConfig(
@@ -10,6 +10,7 @@ from unittest.mock import patch
import pytest
import torch
from vllm.config import set_current_vllm_config
from vllm.engine.arg_utils import EngineArgs
from vllm.utils.mem_utils import MemorySnapshot
from vllm.v1.worker.gpu_worker import Worker, init_worker_distributed_environment
@@ -95,7 +96,12 @@ def worker_process(
side_effect=make_operation_tracker("nccl_all_reduce", original_all_reduce),
)
with init_patch, memory_patch, all_reduce_patch:
with (
init_patch,
memory_patch,
all_reduce_patch,
set_current_vllm_config(vllm_config),
):
# Initialize device (this is where we test the order)
worker.init_device()