Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
change
sglang
Commits
6b8f66ef
Unverified
Commit
6b8f66ef
authored
Aug 02, 2024
by
Yineng Zhang
Committed by
GitHub
Aug 03, 2024
Browse files
misc: update cuda graph capture exception log (#894)
parent
7937a886
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
6 additions
and
4 deletions
+6
-4
python/sglang/srt/model_executor/model_runner.py
python/sglang/srt/model_executor/model_runner.py
+6
-4
No files found.
python/sglang/srt/model_executor/model_runner.py
View file @
6b8f66ef
...
@@ -312,10 +312,12 @@ class ModelRunner:
...
@@ -312,10 +312,12 @@ class ModelRunner:
self
.
cuda_graph_runner
.
capture
(
batch_size_list
)
self
.
cuda_graph_runner
.
capture
(
batch_size_list
)
except
RuntimeError
as
e
:
except
RuntimeError
as
e
:
raise
Exception
(
raise
Exception
(
f
"Capture cuda graph failed:
{
e
}
. Possible solutions:
\n
"
f
"Capture cuda graph failed:
{
e
}
\n
"
f
"1. disable cuda graph by --disable-cuda-graph
\n
"
"Possible solutions:
\n
"
f
"2. set --mem-fraction-static to a smaller value
\n
"
"1. disable torch compile by not using --enable-torch-compile
\n
"
f
"Open an issue on GitHub with reproducible scripts if you need help.
\n
"
"2. disable cuda graph by --disable-cuda-graph
\n
"
"3. set --mem-fraction-static to a smaller value
\n
"
"Open an issue on GitHub https://github.com/sgl-project/sglang/issues/new/choose
\n
"
)
)
@
torch
.
inference_mode
()
@
torch
.
inference_mode
()
...
...
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