Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in
Toggle navigation
Menu
Open sidebar
OpenDAS
vllm_cscc
Commits
ad971af8
Unverified
Commit
ad971af8
authored
Apr 08, 2025
by
zxfan-cpu
Committed by
GitHub
Apr 07, 2025
Browse files
[Bugfix] fix use-ep bug to enable ep by dp/tp size > 1 (#16161)
parent
f2ebb6f5
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
1 addition
and
1 deletion
+1
-1
vllm/model_executor/layers/fused_moe/layer.py
vllm/model_executor/layers/fused_moe/layer.py
+1
-1
No files found.
vllm/model_executor/layers/fused_moe/layer.py
View file @
ad971af8
...
@@ -437,7 +437,7 @@ class FusedMoE(torch.nn.Module):
...
@@ -437,7 +437,7 @@ class FusedMoE(torch.nn.Module):
# Use expert parallelism instead of tensor parallelism?
# Use expert parallelism instead of tensor parallelism?
vllm_config
=
get_current_vllm_config
()
vllm_config
=
get_current_vllm_config
()
use_ep
=
(
vllm_config
.
parallel_config
.
enable_expert_parallel
use_ep
=
(
vllm_config
.
parallel_config
.
enable_expert_parallel
and
self
.
tp_size
>
1
)
and
self
.
tp_size
*
self
.
dp_size
>
1
)
# For smuggling this layer into the fused moe custom op
# For smuggling this layer into the fused moe custom op
self
.
use_direct_call
=
self
.
dp_size
==
1
self
.
use_direct_call
=
self
.
dp_size
==
1
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment