-
Notifications
You must be signed in to change notification settings - Fork 111
Update openEuler dockerfile for COMPILE_CUSTOM_KERNELS=1 #664
New issue
Have a question about this project? No Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “No 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? No Sign in to your account
Conversation
b42c5fe
to
351314f
Compare
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.
And for main pls refer to #661
Especially, pip cache purge
and $LD_LIBRARY_PATH
order and layer squash
@@ -96,4 +96,4 @@ jobs: | |||
# TODO: support and enable custom ops build for openEuler |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
this todo can be removed.
Dockerfile.openEuler
Outdated
|
||
# Install vllm-ascend | ||
RUN source /usr/local/Ascend/ascend-toolkit/set_env.sh && \ | ||
source /usr/local/Ascend/nnal/atb/set_env.sh && \ | ||
export LD_LIBRARY_PATH=/usr/local/Ascend/ascend-toolkit/latest/`uname -i`-linux/devlib:$LD_LIBRARY_PATH && \ | ||
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/Ascend/ascend-toolkit/latest/`uname -i`-linux/devlib && \ | ||
export LIBRARY_PATH=/usr/local/Ascend/ascend-toolkit/latest/lib64:$LIBRARY_PATH && \ |
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.
LIBRARY_PATH can be removed IMO
### What this PR does / why we need it? vLLM Ascend side followup on: [Core] Remove prompt string from engine core data structures vllm-project/vllm@df6f3ce ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? CI passed Signed-off-by: Yikun Jiang <yikunkero@gmail.com>
### What this PR does / why we need it? This pr support minicpm in branch main. see vllm-project#164 ### How was this patch tested? test locally with minicpm --------- Signed-off-by: MengqingCao <cmq0113@163.com>
…ct#656) ### What this PR does / why we need it? Eliminated duplicate `block_table` tensor initialization and cleaned up unused code segments. This resolves an issue where the second creation was overwriting the first, potentially leading to unexpected behavior. Signed-off-by: Jade Zheng <zheng.shoujian@outlook.com>
The torch-npu 2.5.1 are published: https://pypi.org/project/torch-npu/2.5.1/ It's time to remove all torch-npu dev version from vllm-ascend code base Yes, using torch-npu 2.5.1 - [ ] CI passed - [ ] Manually test - [ ] Grep all `dev2025` --------- Signed-off-by: Yikun Jiang <yikunkero@gmail.com>
) ### What this PR does / why we need it? Fix pip install cmd in installation.md Followup on: vllm-project#661 ### Does this PR introduce _any_ user-facing change? No, doc only ### How was this patch tested? Preview Signed-off-by: Yikun Jiang <yikunkero@gmail.com>
### What this PR does / why we need it? `reshape_and_cache_siso` seems have some funcitonality issues, use torch op combination replace this custom op by default. --------- Signed-off-by: ganyi <pleaplusone.gy@gmail.com>
What this PR does / why we need it?
Update openEuler dockerfile for COMPILE_CUSTOM_KERNELS=1
Does this PR introduce any user-facing change?
None
How was this patch tested?