Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

[Doc] Add info on GPU compiler (#6204)

Browse Source 
* Add note about the required compiler version for CUDA. 
* Also added a link that gives a short explanation on compute capability version
This commit is contained in:
Igor Moura 2020-10-06 00:35:18 -03:00 committed by GitHub
parent 1013224888
commit 5908598666
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 10 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
doc/build.rst
View File

@ -168,6 +168,12 @@ XGBoost can be built with GPU support for both Linux and Windows using CMake. GP
An up-to-date version of the CUDA toolkit is required.
.. note:: Checking your compiler version
CUDA is really picky about supported compilers, a table for the compatible compilers for the latests CUDA version on Linux can be seen `here <https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html>`_.
Some distros package a compatible ``gcc`` version with CUDA. If you run into compiler errors with ``nvcc``, try specifying the correct compiler with ``-DCMAKE_CXX_COMPILER=/path/to/correct/g++ -DCMAKE_C_COMPILER=/path/to/correct/gcc``. On Arch Linux, for example, both binaries can be found under ``/opt/cuda/bin/``.
From the command line on Linux starting from the XGBoost directory:
.. code-block:: bash
@ -177,6 +183,10 @@ From the command line on Linux starting from the XGBoost directory:
cmake .. -DUSE_CUDA=ON
make -j4
.. note:: Specifying compute capability
To speed up compilation, the compute version specific to your GPU could be passed to cmake as, e.g., ``-DGPU_COMPUTE_VER=50``. A quick explanation and numbers for some architectures can be found `here <https://arnon.dk/matching-sm-architectures-arch-and-gencode-for-various-nvidia-cards/>`_.
.. note:: Enabling distributed GPU training
By default, distributed GPU training is disabled and only a single GPU will be used. To enable distributed GPU training, set the option ``USE_NCCL=ON``. Distributed GPU training depends on NCCL2, available at `this link <https://developer.nvidia.com/nccl>`_. Since NCCL2 is only available for Linux machines, **distributed GPU training is available only for Linux**.
@ -206,7 +216,6 @@ On Windows, run CMake as follows:
cmake .. -G"Visual Studio 15 2017 Win64" -T v140,cuda=8.0 -DUSE_CUDA=ON
To speed up compilation, the compute version specific to your GPU could be passed to cmake as, e.g., ``-DGPU_COMPUTE_VER=50``.
The above cmake configuration run will create an ``xgboost.sln`` solution file in the build directory. Build this solution in release mode as a x64 build, either from Visual studio or from command line:
.. code-block:: bash