biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Refactor] Remove redundant TP gather/split in split_qkv in QwenVL (#28271)

Browse Source 
Signed-off-by: gcanlin <canlinguosdu@gmail.com>
This commit is contained in:
Canlin Guo
2025-11-12 23:56:47 +08:00
committed by GitHub
parent f76e85c299
commit bc5bd45c7d
2 changed files with 1 additions and 42 deletions
Download Patch File Download Diff File Expand all files Collapse all files

30
vllm/model_executor/models/qwen2_5_vl.py
View File

@@ -291,25 +291,6 @@ class Qwen2_5_VisionMLP(nn.Module):
return x_down return x_down
def all_gather_interleave(local_tensor, hidden_size: int, tp_size: int):
"""All-gather the input tensor interleavely across model parallel group."""
import torch.distributed as dist
gathered_tensors = [torch.zeros_like(local_tensor) for _ in range(tp_size)]
dist.all_gather(
gathered_tensors, local_tensor, group=parallel_state.get_tp_group().device_group
)
gathered_tensors_split = [
torch.split(tensor, hidden_size // tp_size, -1) for tensor in gathered_tensors
]
ordered_tensors = [
tensor for pair in zip(*gathered_tensors_split) for tensor in pair
]
result_tensor = torch.cat(ordered_tensors, dim=-1)
return result_tensor
class Qwen2_5_VisionAttention(nn.Module): class Qwen2_5_VisionAttention(nn.Module):
def __init__( def __init__(
self, self,
@@ -383,21 +364,10 @@ class Qwen2_5_VisionAttention(nn.Module):
def split_qkv(self, qkv: torch.Tensor) -> tuple[torch.Tensor, ...]: def split_qkv(self, qkv: torch.Tensor) -> tuple[torch.Tensor, ...]:
# [s, b, 3 * head * head_dim] # [s, b, 3 * head * head_dim]
seq_len, bs, _ = qkv.shape seq_len, bs, _ = qkv.shape
if self.tp_size > 1:
qkv = all_gather_interleave(qkv, self.qkv.hidden_size, self.tp_size)
# [s, b, 3 * head * head_dim] -> 3 * [s, b, head * head_dim] # [s, b, 3 * head * head_dim] -> 3 * [s, b, head * head_dim]
q, k, v = qkv.chunk(3, dim=2) q, k, v = qkv.chunk(3, dim=2)
# 3 * [s, b, head * head_dim]
if self.tp_size > 1:
splitter = partial(
dist_utils.split_tensor_along_last_dim, num_partitions=self.tp_size
)
q = splitter(q)[self.tp_rank]
k = splitter(k)[self.tp_rank]
v = splitter(v)[self.tp_rank]
# 3 * [s, b, head * head_dim] -> 3 * [s, b, head, head_dim] # 3 * [s, b, head * head_dim] -> 3 * [s, b, head, head_dim]
new_shape = ( new_shape = (
seq_len, seq_len,

13
vllm/model_executor/models/qwen2_vl.py
View File

@@ -50,7 +50,7 @@ from vllm.attention.layer import (
) )
from vllm.config import VllmConfig from vllm.config import VllmConfig
from vllm.config.multimodal import BaseDummyOptions from vllm.config.multimodal import BaseDummyOptions
from vllm.distributed import parallel_state, tensor_model_parallel_all_gather from vllm.distributed import parallel_state
from vllm.distributed import utils as dist_utils from vllm.distributed import utils as dist_utils
from vllm.logger import init_logger from vllm.logger import init_logger
from vllm.model_executor.layers.activation import QuickGELU from vllm.model_executor.layers.activation import QuickGELU
@@ -396,21 +396,10 @@ class Qwen2VisionAttention(nn.Module):
def split_qkv(self, qkv: torch.Tensor) -> tuple[torch.Tensor, ...]: def split_qkv(self, qkv: torch.Tensor) -> tuple[torch.Tensor, ...]:
# [s, b, 3 * head * head_dim] # [s, b, 3 * head * head_dim]
seq_len, bs, _ = qkv.shape seq_len, bs, _ = qkv.shape
if self.tp_size > 1:
qkv = tensor_model_parallel_all_gather(qkv)
# [s, b, 3 * head * head_dim] -> 3 * [s, b, head * head_dim] # [s, b, 3 * head * head_dim] -> 3 * [s, b, head * head_dim]
q, k, v = qkv.chunk(3, dim=2) q, k, v = qkv.chunk(3, dim=2)
# 3 * [s, b, head * head_dim]
if self.tp_size > 1:
splitter = partial(
dist_utils.split_tensor_along_last_dim, num_partitions=self.tp_size
)
q = splitter(q)[self.tp_rank]
k = splitter(k)[self.tp_rank]
v = splitter(v)[self.tp_rank]
# 3 * [s, b, head * head_dim] -> 3 * [s, b, head, head_dim] # 3 * [s, b, head * head_dim] -> 3 * [s, b, head, head_dim]
new_shape = ( new_shape = (
seq_len, seq_len,