From: Kirill A. Korinsky <kirill@korins.ky>
Subject: Re: relayd: fix GCC16 warnings
To: Theo Buehler <tb@theobuehler.org>
Cc: rafael@sizeofvoid.org, tech@openbsd.org
Date: Sat, 24 Oct 2026 01:06:32 +0200

On Sat, 24 Oct 2026 00:20:58 +0200,
Theo Buehler <tb@theobuehler.org> wrote:
> 
> On Fri, Oct 23, 2026 at 11:41:47PM +0200, Kirill A. Korinsky wrote:
> > On Fri, 23 Oct 2026 21:58:19 +0200,
> > Rafael Sadowski <rafael@sizeofvoid.org> wrote:
> > > 
> > > Make GCC16 happy with relayd code
> > > 
> > > ca.c:348:50: warning: operand of '?:' changes signedness from 'int' to 'unsigned int' due to unsignedness of other operand [-Wsign-compare]
> > >   348 |                 cko.cko_tlen = (ret > 0) ? len : -1;
> > > 
> > > 
> > ...
> > > -		cko.cko_tlen = (ret > 0) ? len : -1;
> > > +		cko.cko_tlen = (ret > 0) ? (int)len : -1;
> > 
> > Not sure that I understand this warning.
> > 
> > len is int, -1 is int, cko_tlen is int.
> >
> > Why gcc things what something here is unsigned int?
> 
> gcc is often dumb, but here it is right (even if it's a harmless
> warning)
> 
> len is an unsigned int. Thus the return type of the conditional
> expression is unsigned int and -1 ets promoted.
> 

Thanks, probably I need some sleep, indeed.

-- 
wbr, Kirill