Bill Sommerfeld <sommerfeld@orchard.arlington.ma.us> writes:
> > It would be dangerous just to have a flag to the linker to say "don't
> > print the warnings" for the tree because people could add new unsafe
> > uses of mktemp. Therefore, we need *both* lint to be beefed up and
> > then after that a way of avoiding the linker warnings for in-tree code
> > because lint will subsume them.
> 
> If we do this, we also need to make all in-tree code lint-clean.  Do
> we have the resources for this?

Yes, and I believe we should be aiming for that long term. However, I
suspect it will take the next couple of years to finish that as a
project. (I also want us building a much more complete regression
testing environment with every API unit tested, but I believe that may
also take years. However, the march of a thousand miles starts with
one step...)

Perry
--
Perry E. Metzger		perry@wasabisystems.com
--
NetBSD Development, Support & CDs. http://www.wasabisystems.com/