minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

[Bugfix] Fix incorrect dp-attention parallel info in bench_one_batch (#21519)

Browse Source
This commit is contained in:
lviy
2026-04-02 20:13:53 -07:00
committed by GitHub
parent 69e89a1fcc
commit 5f0df1e2ad
1 changed files with 3 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
python/sglang/bench_one_batch.py
+3 -1
View File
@@ -66,6 +66,7 @@ import torch.distributed as dist
from sglang.srt.configs.model_config import ModelConfig from sglang.srt.configs.model_config import ModelConfig
from sglang.srt.distributed.parallel_state import destroy_distributed_environment from sglang.srt.distributed.parallel_state import destroy_distributed_environment
from sglang.srt.entrypoints.engine import _set_envs_and_config from sglang.srt.entrypoints.engine import _set_envs_and_config
from sglang.srt.layers.dp_attention import get_attention_tp_size
from sglang.srt.layers.moe import initialize_moe_config from sglang.srt.layers.moe import initialize_moe_config
from sglang.srt.layers.quantization.fp4_utils import initialize_fp4_gemm_config from sglang.srt.layers.quantization.fp4_utils import initialize_fp4_gemm_config
from sglang.srt.layers.quantization.fp8_utils import initialize_fp8_gemm_config from sglang.srt.layers.quantization.fp8_utils import initialize_fp8_gemm_config
@@ -453,7 +454,8 @@ def _maybe_prepare_mlp_sync_batch(batch: ScheduleBatch, model_runner):
prepare_mlp_sync_batch_raw( prepare_mlp_sync_batch_raw(
batch, batch,
dp_size=model_runner.server_args.dp_size, dp_size=model_runner.server_args.dp_size,
attn_tp_size=1, attn_tp_size=get_attention_tp_size(),
attn_cp_size=model_runner.attn_cp_size,
tp_group=model_runner.tp_group, tp_group=model_runner.tp_group,
get_idle_batch=None, get_idle_batch=None,
disable_cuda_graph=model_runner.server_args.disable_cuda_graph, disable_cuda_graph=model_runner.server_args.disable_cuda_graph,