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
1cbcfb94
Unverified
Commit
1cbcfb94
authored
Sep 24, 2025
by
Li, Jiang
Committed by
GitHub
Sep 24, 2025
Browse files
[Bugfix][CPU] Skip unsupported custom op register on CPU (#25534)
Signed-off-by:
jiang1.li
<
jiang1.li@intel.com
>
parent
fed8a9b1
Changes
1
Show whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
8 additions
and
6 deletions
+8
-6
vllm/model_executor/layers/quantization/utils/fp8_utils.py
vllm/model_executor/layers/quantization/utils/fp8_utils.py
+8
-6
No files found.
vllm/model_executor/layers/quantization/utils/fp8_utils.py
View file @
1cbcfb94
...
...
@@ -130,12 +130,14 @@ def _w8a8_triton_block_scaled_mm_fake(
device
=
qx
.
device
)
direct_register_custom_op
(
# Note: the check can be removed when CPU torch > 2.7
if
not
current_platform
.
is_cpu
():
direct_register_custom_op
(
"w8a8_triton_block_scaled_mm_func"
,
_w8a8_triton_block_scaled_mm_func
,
fake_impl
=
_w8a8_triton_block_scaled_mm_fake
,
dispatch_key
=
"CUDA"
,
)
)
# TODO fix ROCm->Triton custom path:
...
...
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