"Dr. Thomas Orgis" <thomas.orgis%uni-hamburg.de@localhost> writes:
So, this is my plan. Any opinions? Is there a realistic option to push
some things into 2019Q1?
(Replying without really digging in.)
It's great that you are working on this! It sounds sensible at first
read to me. We will have to think about the static/dynamic library
issues.
I definitely would like to see review and feedback from other people who
pay attention to HPC/blas things. I know there's at least one :-)
2019Q1 is nominally due to be released by this coming Sunday (being the
last day of Q1). It's always up to the person running each freeze, but
as a rule we do not make big structural changes in the day or so before
the freeze, especially if they haven't been up for review for a while.
A change this big should have at least a week for review. So I do not
see any of this getting into 2019Q1. That's not criticism of what
you've done in any way, but merely a statement that big changes have
risk, and stable branches are about reducing risk.
Separately from review and consensus, soon after 2019Q1 is cut is a good
time for possibly disruptive changes, since there is then a few months
to sort any issues.