cvs commit: src/lib/libc/i386/net htonl.S ntohl.S

[John Baldwin]

On Friday 22 October 2004 09:05 pm, David O'Brien wrote:
> On Thu, Oct 21, 2004 at 03:57:23PM -0400, John Baldwin wrote:
> > I waded though many arch@ archives but couldn't find where I had
> > brought this issue up.  I did find one instance where it was discussed
> > prior to the SMPng commit back in April/May 2000 (old, yes) where cp@
> > wanted to drop 386 and 486 support for 5, and the ideas there were to
> > allow for separate kernels.  At this point, I guess I don't care/have
> > enough time to burn on this.  I would think you of all people would
> > care about sticking to previously agreed to decisions though.
>
> I'm trying to.  The problem is we don't seem to have a consensis on what
> the "previously agreed to decision" was.  And we didn't formally document
> it.

As best as I can tell, the only discussions in the public lists centered on 
the kernel and not really userland.  The closest thing to mentioning userland 
is that at the time the discussions always assumed that only a different 
kernel would be needed for 80386 support, but that doesn't directly address 
userland, at best it weakly implies that the userland would work on both.  
However, as I said earlier, at this point I no longer care what happens in 
RELENG_5 on this topic.  I would be fine with dropping 80386 for 5.4 
personally.

-- 

John Baldwin <jhb at FreeBSD.org>  <><  http://www.FreeBSD.org/~jhb/
"Power Users Use the Power to Serve!"  -  http://www.FreeBSD.org/