cvs commit: src/lib/libufs Makefile

[Tom Rhodes]

On Tue, 30 Aug 2005 16:30:36 -1000
juli mallett <jmallett at FreeBSD.org> wrote:

> * Tom Rhodes <trhodes at FreeBSD.org> [ Date: 2005-08-30 ]
> 	[ w.r.t. Re: cvs commit: src/lib/libufs Makefile ]
> > On Sun, 28 Aug 2005 12:11:42 -1000
> > juli mallett <jmallett at FreeBSD.org> wrote:
> > 
> > > * Tom Rhodes <trhodes at FreeBSD.org> [ Date: 2005-08-26 ]
> > > 	[ w.r.t. Re: cvs commit: src/lib/libufs Makefile ]
> > > > On Thu, 25 Aug 2005 09:54:16 -1000
> > > > juli mallett <jmallett at FreeBSD.org> wrote:
> > > > 
> > > > > * Tom Rhodes <trhodes at FreeBSD.org> [ Date: 2005-08-25 ]
> > > > > 	[ w.r.t. cvs commit: src/lib/libufs Makefile ]
> > > > > > trhodes     2005-08-25 10:22:30 UTC
> > > > > > 
> > > > > >   FreeBSD src repository
> > > > > > 
> > > > > >   Modified files:
> > > > > >     lib/libufs           Makefile 
> > > > > >   Log:
> > > > > >   Hook getino.3 up to the build and link it to putino.3.
> > > > > >   
> > > > > >   PR:     83820
> > > > > 
> > > > > This is wrong.  As the PR states it was purposely not hooked
> > > > > up. inode.c doesn't have functions which look like the
> > > > > interfaces you've just published, and there isn't even a
> > > > > putino(3) in there.  If the cause of the confusion is the
> > > > > manpage being listed in Xr's or something, then comment out
> > > > > the Xr's.  This is very obviously wrong, if you try to use
> > > > > what's in the manpage I committed.  Probably it shoudln't be
> > > > > in CVS at all.  The reason I put it there at the time is that
> > > > > it was ready and inode.c was about ready to hit CVS, but that
> > > > > got dropped.  Now, if you want to help me remember what P4
> > > > > tree that inode.c was in, and pick up whereever I left off,
> > > > > that'd be awesome.
> > > > 
> > > > Let me understand real quick, it was purposely not hooked up
> > > > because it was the only file not updated?  I mean, I can back
> > > > this out, no problem; however, I fail to see why nothing is
> > > > going on with this code.  No clue on the branch, sorry.
> > > 
> > > It was purposely not hooked up because the code that it documents
> > > isn't in CVS.
> > > 
> > 
> > I know now why it wasn't hooked up to the build.  What I'm asking
> > is:
> > 
> > 1: Is this currently maintained?  and
> > 2: What utilities are currently using this?
> 
> I don't remember what uses getino() as implemented in libufs.  Maybe
> nothing, in which case it can just be removed.  In any event, the
> interface that is there is bad and shouldn't be documented.  That
> said, it isn't currently maintained very well, so the odds of me
> remembering where the new code is and what's wrong with it are low,
> so in any event, the right thing to do is for the manpage to not be
> installed, and to not be in CVS.
> 
> Thanx,
> juli.
> 

Backed out, as you know.  We'll discuss this offline.  :)

-- 
Tom Rhodes