Hi,
I am looking to build CBC from source, for our HPC cluster. A golden rule of building for science is to be able to reproduce builds. This implies not fetching source files while building, but rather fetching all source tarballs and using checksums to make sure they don't change over time.
This does not seem to be the way that coinbrew works. To be honest, I am unsure why coinbrew is needed at all, considering that all subprojects have configure scripts.
I installed all subprojects by downloading tarballs from https://www.coin-or.org/download/source/, recording their checksums, extracting, running configure, make, make install.
It all builds fine. However, our user wants the equivalent of
coinbrew build Cbc --enable-cbc-parallel
and there seems to be no corresponding option in the configure scripts.
Is there any way to document what is needed to build your software using standard tools ?
Hi, I am looking to build CBC from source, for our HPC cluster. A golden rule of building for science is to be able to reproduce builds. This implies not fetching source files while building, but rather fetching all source tarballs and using checksums to make sure they don't change over time.
This does not seem to be the way that coinbrew works. To be honest, I am unsure why coinbrew is needed at all, considering that all subprojects have configure scripts.
I installed all subprojects by downloading tarballs from https://www.coin-or.org/download/source/, recording their checksums, extracting, running configure, make, make install.
It all builds fine. However, our user wants the equivalent of
and there seems to be no corresponding option in the configure scripts.
Is there any way to document what is needed to build your software using standard tools ?