svn commit: r340038 - in head: lib/libc/sys sys/kern

[Gary Jennejohn]

On Thu, 1 Nov 2018 20:29:27 -0700
Conrad Meyer <cem at freebsd.org> wrote:

> On Thu, Nov 1, 2018 at 5:46 PM Konstantin Belousov <kostikbel at gmail.com> wrote:
> >
> > On Thu, Nov 01, 2018 at 11:46:23PM +0000, Conrad Meyer wrote:  
> > > ...
> > > @@ -232,7 +232,15 @@ points outside the process's allocated address space.
> > >  A signal was delivered before the time limit expired and
> > >  before any of the selected events occurred.
> > >  .It Bq Er EINVAL
> > > -The specified time limit is invalid. One of its components is negative or too large.
> > > +The specified time limit is invalid.
> > > +One of its components is negative or too large.
> > > +.It Bq Er EINVAL
> > > +The number of pollfd structures specified by
> > > +.Fa nfds
> > > +exceeds the system tunable
> > > +.Va kern.maxfilesperproc
> > > +and
> > > +.Dv FD_SETSIZE .  
> >
> > This is somewhat confusing.  Looking only at the man page text, most
> > straight reading of it is that nfds cannot exceed min(maxfilesperproc,
> > FD_SETSIZE).  But in fact nfds cannot exceed max of it, which is
> > significantly different, i.e. poll(2) does not suffer from the (userspace)
> > FD_SETSIZE limitation of select(2).  
> 
> min(maxfilesperproc, FD_SETSIZE) would be written as "nfds exceeds ...
> maxfilesperproc OR FD_SETSIZE," rather than "and."  I think the
> sentence is unambiguous.
> 

Based on Konstantin's comment I'd have to say that it is not
unambiguous.

As a native English speaker I had to read it several times to
understand it.

I think something like "exceeds both" might be clearer.

-- 
Gary Jennejohn