Subject: Re: CVS commit: src/sys
To: Jonathan Stone <jonathan@dsg.stanford.edu>
From: Robert Elz <kre@munnari.OZ.AU>
List: tech-kern
Date: 12/07/2004 04:26:52
    Date:        Mon, 06 Dec 2004 12:58:27 -0800
    From:        Jonathan Stone <jonathan@dsg.stanford.edu>
    Message-ID:  <E1CbPwB-0008B5-00@smeg.dsg.stanford.edu>

  | I cannot agree.  There's still been no case presented that we ever
  | *need* more than a single loopback device. (It could be that the
  | ``needs_count'' was merely a hangover, from before we had a separate
  | ``needs_flag''.

No, you're debating a different question.

We've always had the ability to have multiple loopback devices, that
hasn't been changed here.

(I quote from a fairly old if_loop.c ...

	struct  ifnet loif[NLOOP];

which is from ...

__KERNEL_RCSID(0, "$NetBSD: if_loop.c,v 1.48 2003/08/15 19:22:08 jonathan Exp $"
)

All that has been altered is the way that is achieved.

kre