rapidsai / gpuci-build-environment Goto Github PK
View Code? Open in Web Editor NEWCommon build environment used by gpuCI for building RAPIDS
License: Apache License 2.0
gpuci-build-environment's People
Contributors
Stargazers
Watchers
Forkers
quansight raydouglass kimsoohwan hubbucket-team vishalbelsare kkraus14 dillon-cullinan rlratzel shelly0-0c revans2 dantegd jakirkham shaunstoltz teju85 gmarkall shwina jasonatnvidia ajschmidt8 mike-wendt trxcllnt ayodeawe galipremsagar charlesbluca msadang johnzed bdice joglekaragpuci-build-environment's Issues
Add NUMA to build machines
For ucx-py tests we need numa packages: libnuma libnuma-dev . Would it be possible to add these to the build machines used for CI for ucx-py
cc @jakirkham
Should build RHEL 7 gcc toolchains with `--disable-libstdcxx-dual-abi`
Dockerfile.centos7-gcc7 and Dockerfile.centos7-gcc9 should pass the--disable-libstdcxx-dual-abi flag to use the pre-C++-11 ABI to align with the RHEL 6/7 toolchain ABI.
From https://gcc.gnu.org/onlinedocs/libstdc++/manual/configure.html:
--disable-libstdcxx-dual-abi
Disable support for the new, C++11-conforming implementations of std::string, std::list etc. so that the library only provides definitions of types using the old ABI (see Dual ABI). This option changes the library ABI.
Conda environment dir is owned by root
The conda environment dir is owned by root, but the GPU runners run as a non-root user. This is problematic when CMake tries to install things such as directories because it tries to set permissions on things which may already exist and can't because it doesn't own them.
Other CUDA Versions (11.7)
first, thanks for maintaining such a useful set of docker images. i have found it very useful.
i wanted to ask if I could help add images built on cuda 11.7 . I'd be happy to submit a PR if someone helps me understand what needs changing.
Conda-forge about to update to GCC 10
Conda-forge is about to move to GCC 10 for all of its non-CUDA package builds: conda-forge/conda-forge-pinning-feedstock#2010
This likely means we'll need to move to using GCC 10's libstdc++ in our build environment. Will need to replicate this (https://github.com/rapidsai/gpuci-build-environment/blob/branch-21.10/builds-gcc/Dockerfile.centos7-gcc7#L50-L54) into the GCC 9 images most likely.
Install conda-build and conda-verify in base
Would be good to move the installation of conda-build and conda-verify to the base environment. Currently these are installed in the gdf environment.
Add cudatoolkit to image
Pre-install cudatoolkit based on the version of CUDA for the container
conda-build version 3.18.3 is not found
I am trying to build this on my local box as part of addressing review comments from rapidsai/cudf#2132 but I am not able to build it after conda-build was pinned to 3.18.3. 9857270
Is there something I am missing?
$ sudo docker build -t test_image -f ./Dockerfile .
...
PackagesNotFoundError: The following packages are not available from current channels:
- conda-build=3.18.3
Current channels:
- https://conda.anaconda.org/rapidsai/label/cuda9.2/linux-64
- https://conda.anaconda.org/rapidsai/label/cuda9.2/noarch
- https://conda.anaconda.org/rapidsai-nightly/label/cuda9.2/linux-64
- https://conda.anaconda.org/rapidsai-nightly/label/cuda9.2/noarch
- https://conda.anaconda.org/nvidia/label/cuda9.2/linux-64
- https://conda.anaconda.org/nvidia/label/cuda9.2/noarch
- https://conda.anaconda.org/conda-forge/linux-64
- https://conda.anaconda.org/conda-forge/noarch
- https://repo.anaconda.com/pkgs/main/linux-64
- https://repo.anaconda.com/pkgs/main/noarch
- https://repo.anaconda.com/pkgs/free/linux-64
- https://repo.anaconda.com/pkgs/free/noarch
- https://repo.anaconda.com/pkgs/r/linux-64
- https://repo.anaconda.com/pkgs/r/noarch
To search for alternate channels that may provide the conda package you're
looking for, navigate to
https://anaconda.org
and use the search bar at the top of the page.
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.