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
d3c84297
Unverified
Commit
d3c84297
authored
Oct 06, 2025
by
Thomas Parnell
Committed by
GitHub
Oct 06, 2025
Browse files
[CI] Add comment about the single cudagraph capture size that is used (#26252)
parent
f509a208
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
4 additions
and
0 deletions
+4
-0
tests/conftest.py
tests/conftest.py
+4
-0
No files found.
tests/conftest.py
View file @
d3c84297
...
@@ -740,6 +740,10 @@ class VllmRunner:
...
@@ -740,6 +740,10 @@ class VllmRunner:
)
)
if
not
kwargs
.
get
(
"compilation_config"
,
None
):
if
not
kwargs
.
get
(
"compilation_config"
,
None
):
# Note(@tdoublep): This is set to 4 because some tests (e.g., hybrid
# model tests) may set max_num_seqs=4. If min cudagraph_capture_size is
# set to larger than max_num_seqs, then it will lead to *no* graphs
# being captured which can trigger edge cases that we don't handle yet.
kwargs
[
"compilation_config"
]
=
{
"cudagraph_capture_sizes"
:
[
4
]}
kwargs
[
"compilation_config"
]
=
{
"cudagraph_capture_sizes"
:
[
4
]}
with
init_ctx
:
with
init_ctx
:
...
...
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