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
e983c804
Commit
e983c804
authored
Mar 19, 2025
by
zhuwenwen
Browse files
add --gpu-max-threads-per-block=1024 to fix rms_norm_kernel accuracy
parent
0dc4c8e9
Changes
1
Show whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
4 additions
and
3 deletions
+4
-3
cmake/utils.cmake
cmake/utils.cmake
+4
-3
No files found.
cmake/utils.cmake
View file @
e983c804
...
@@ -119,10 +119,11 @@ function (get_torch_gpu_compiler_flags OUT_GPU_FLAGS GPU_LANG)
...
@@ -119,10 +119,11 @@ function (get_torch_gpu_compiler_flags OUT_GPU_FLAGS GPU_LANG)
list
(
APPEND GPU_FLAGS
list
(
APPEND GPU_FLAGS
"-DUSE_ROCM"
"-DUSE_ROCM"
#"-DENABLE_FP8"
#
"-DENABLE_FP8"
"-U__HIP_NO_HALF_CONVERSIONS__"
"-U__HIP_NO_HALF_CONVERSIONS__"
"-U__HIP_NO_HALF_OPERATORS__"
"-U__HIP_NO_HALF_OPERATORS__"
"-fno-gpu-rdc"
)
"-fno-gpu-rdc"
"--gpu-max-threads-per-block=1024"
)
endif
()
endif
()
set
(
${
OUT_GPU_FLAGS
}
${
GPU_FLAGS
}
PARENT_SCOPE
)
set
(
${
OUT_GPU_FLAGS
}
${
GPU_FLAGS
}
PARENT_SCOPE
)
...
...
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