fix conflicts

This commit is contained in:
Gaohan123 committed 2025-10-24 16:49:19 +08:00
1 parent 22f36de34e
commit 171f3c1ea7
42 files changed
+2930 -1842

No files matched your search

+27 -18
View File
@@ -1,13 +1,14 @@
from vllm.v1.worker.gpu_worker import Worker as GPUWorker
import torch
import os
import gc
from vllm.utils import GiB_bytes, MemorySnapshot
from vllm.platforms import current_platform
from vllm.worker.worker import _check_if_gpu_supports_dtype
from vllm.v1.worker.gpu_worker import init_worker_distributed_environment
import os
import torch
from vllm.model_executor import set_random_seed
from vllm.platforms import current_platform
from vllm.utils import GiB_bytes, MemorySnapshot
from vllm.v1.utils import report_usage_stats
from vllm.v1.worker.gpu_worker import Worker as GPUWorker
from vllm.v1.worker.gpu_worker import init_worker_distributed_environment
from vllm.worker.worker import _check_if_gpu_supports_dtype
from vllm_omni.worker.gpu_diffusion_model_runner import GPUDiffusionModelRunner
@@ -34,10 +35,15 @@ class GPUDiffusionWorker(GPUWorker):
# take current memory snapshot
self.init_snapshot = MemorySnapshot()
self.requested_memory = (self.init_snapshot.total_memory *
self.cache_config.gpu_memory_utilization)
self.requested_memory = (
self.init_snapshot.total_memory
* self.cache_config.gpu_memory_utilization
)
if self.init_snapshot.free_memory < self.requested_memory:
GiB = lambda b: round(b / GiB_bytes, 2)
def GiB(b):
return round(b / GiB_bytes, 2)
raise ValueError(
f"Free memory on device "
f"({GiB(self.init_snapshot.free_memory)}/"
@@ -48,20 +54,23 @@ class GPUDiffusionWorker(GPUWorker):
f"utilization or reduce GPU memory used by other processes."
)
else:
raise RuntimeError(
f"Not support device type: {self.device_config.device}")
raise RuntimeError(f"Not support device type: {self.device_config.device}")
# Initialize the distributed environment.
init_worker_distributed_environment(self.vllm_config, self.rank,
self.distributed_init_method,
self.local_rank,
current_platform.dist_backend)
init_worker_distributed_environment(
self.vllm_config,
self.rank,
self.distributed_init_method,
self.local_rank,
current_platform.dist_backend,
)
# Set random seed.
set_random_seed(self.model_config.seed)
# Construct the model runner
self.model_runner: GPUDiffusionModelRunner = GPUDiffusionModelRunner(
self.vllm_config, self.device)
self.vllm_config, self.device
)
if self.rank == 0:
# If usage stat is enabled, collect relevant info.
report_usage_stats(self.vllm_config)
report_usage_stats(self.vllm_config)