svn commit: r349233 - head/sys/sys

[Alan Somers]

On Thu, Jun 20, 2019 at 9:14 AM Ian Lepore <ian at freebsd.org> wrote:
>
> On Thu, 2019-06-20 at 14:35 +0000, Alan Somers wrote:
> > Author: asomers
> > Date: Thu Jun 20 14:35:28 2019
> > New Revision: 349233
> > URL: https://svnweb.freebsd.org/changeset/base/349233
> >
> > Log:
> >   #include <sys/types.h> from sys/filio.h
> >
> >   This fixes world build after r349231
> >
> >   Reported by:        Jenkins
> >   MFC after:  2 weeks
> >   MFC-With:   349231
> >   Sponsored by:       The FreeBSD Foundation
> >
> > Modified:
> >   head/sys/sys/filio.h
> >
> > Modified: head/sys/sys/filio.h
> > =====================================================================
> > =========
> > --- head/sys/sys/filio.h      Thu Jun 20 14:34:45 2019        (r349232)
> > +++ head/sys/sys/filio.h      Thu Jun 20 14:35:28 2019        (r349233)
> > @@ -40,6 +40,7 @@
> >  #ifndef      _SYS_FILIO_H_
> >  #define      _SYS_FILIO_H_
> >
> > +#include <sys/types.h>
> >  #include <sys/ioccom.h>
> >
> >  /* Generic file-descriptor ioctl's. */
> >
>
> I wonder... is this one of those situations where it is better to use
> __int64_t in the struct, then #include <sys/_types.h>?  I think the net
> effect there would be less pollution with other types?  I've never seen
> written guidance about when to use the __names and _types.h, but I've
> always had the general impression that if you have to include a header
> from another system header, it's better to use the _header.h if it
> exists.
>
> -- Ian

Good question.  grep shows almost equal numbers of each (37 types.h
and 33 _types.h) in sys/sys.  Do you think I should change it?
-Alan