-
Notifications
You must be signed in to change notification settings - Fork 754
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Integrate turbomind into sgl-kernel #2999
base: main
Are you sure you want to change the base?
Conversation
sgl-kernel/MANIFEST.in
Outdated
@@ -0,0 +1,2 @@ | |||
include src/sgl-kernel/turbomind/lib/_turbomind_ext.cpython-312-x86_64-linux-gnu.so |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Why is the Python version specified here?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
because pybind11 is used in building
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We can’t hardcode the Python version
@@ -0,0 +1 @@ | |||
import sgl_kernel.turbomind.lib._turbomind_ext as _turbomind_ext |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Please run make format
… into turbomind_integrate
BTW Can we only use the C++ and CUDA code from |
BTW we should ensure the PR Test (sgl-kernel) is successful. |
… into turbomind_integrate
sgl-kernel/build.sh
Outdated
@@ -22,6 +22,9 @@ docker run --rm \ | |||
export SGL_KERNEL_ENABLE_SM90A=${ENABLE_SM90A} && \ | |||
mkdir -p /usr/lib/x86_64-linux-gnu/ && \ | |||
ln -s /usr/local/cuda-${CUDA_VERSION}/targets/x86_64-linux/lib/stubs/libcuda.so /usr/lib/x86_64-linux-gnu/libcuda.so && \ | |||
cd /sgl-kernel && \ | |||
cd /sgl-kernel/3rdparty/turbomind && \ |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We should handle the build process internally within setup.py, not this way.
sgl-kernel/setup.py
Outdated
extra_link_args = [ | ||
"-Wl,-rpath,$ORIGIN/../../torch/lib", | ||
"-L/usr/lib/x86_64-linux-gnu", | ||
f"{str(root)}/3rdparty/turbomind/build/lib/libgemm2.a", |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
same as above
@zhyncs
Checklist