1 << 31 redux

[Rodney W. Grimes]

> On 11 January 2018 at 07:47, Warner Losh <imp at bsdimp.com> wrote:
> >
> >
> > If we can't get people to fix the warnings we have in the tree now
> > (especially the kernel), why enable new warnings that will just be ignored?
> >
> 
> We've been doing a reasonable job with warnings until now. Either way, at
> higher warnings levels we should add what we can.

Is not this 1U<<31 -> signed value really just sweeping the bigger
issue that we are using signed values in unsigned ways?

-- 
Rod Grimes                                                 rgrimes at freebsd.org