Upgrading self CUDA 7.0

Its been a while since I did any CUDA development.
I last used 7.0
Are there any gotchas that people who have transition to the latest CUDA
suggest I should watch for?
I am not intending to use TensorFlow etc. just traditional parallel GPGPU programming.
Would you recommend learning how to use any of the tools?

Any help or guidance you can give would be most welcome
Bill

    Prof. W. B. Langdon
    Department of Computer Science
    University College London
    Gower Street, London WC1E 6BT, UK
    http://www.cs.ucl.ac.uk/staff/W.Langdon/

GI-2018 http://geneticimprovementofsoftware.com/ deadline 5th February 2018
2018 Humies http://www.human-competitive.org/call-for-entries
EuroGP 2018 http://www.evostar.org/2018/cfp_eurogp.php
barracuda_0.7.107h http://seqbarracuda.sourceforge.net/
choose your background http://web4.cs.ucl.ac.uk/staff/W.Langdon/colour_telephone/bgcolor.html
A Field Guide to Genetic Programming
http://www.gp-field-guide.org.uk/
GP EM http://www.springer.com/10710
GP Bibliography http://www.cs.bham.ac.uk/~wbl/biblio/

CUDA 9 has some breaking API changes, in part even for older architectures. It’s mostly about the new cooperative thread groups feature and new synchronization primitives that go along with it.

In particular you can no longer rely on warp synchronous execution on the new Volta architecture. So you need a new __syncwarp() primitive to ensure thread reconvergence.

It’s advisable to learn about the changes if you plan to execute your software on Volta chips at any point in time.

Many thanks.
I am running on compute level 5.0, so it sounds like I should be ok
Bill