Subject: Re: NetBSD master CVS tree commits
To: None <Chris_G_Demetriou@NIAGARA.NECTAR.CS.CMU.EDU>
From: Jason Thorpe <thorpej@nas.nasa.gov>
List: tech-kern
Date: 01/15/1996 18:43:01
On Mon, 15 Jan 96 17:05:54 EST 
 Chris_G_Demetriou@NIAGARA.NECTAR.CS.CMU.EDU wrote:

 > > thorpej
 > > Sun Jan 14 16:06:51 PST 1996
 > > Update of /a/cvsroot/src/sys/arch/sparc/sparc
 > > In directory pain.lcs.mit.edu:/b/tmp/cvs-serv11023/sparc
 > > 
 > > Modified Files:
 > > 	autoconf.c 
 > > Log Message:
 > > Add a "mountroot hook" mechanism so that devices can take care of
 > > special needs before being used in (*mountroot)().
 > 
 > why make this machine-dependent?  It seems that every port which, say,
 > can have root on a floppy (or other removable-media device) would
 > benefit from this...
 > 
 > There are some weird semantic problems here, though, for instance, how
 > do you know which device is being 'mountrooted'?
 > 
 > perhaps an extra flag should be added someplace, or something...

Yah, I agree that it should be machine-independent.  I was actually 
attempting to solve a specific problem at the time, hence the addition of 
it to the sparc port directly.

However, I specifically named and wrote the hook stuff such that it would 
be easily made into a machine-independent thing.  I've already talked 
about this (briefly, due to time constraints) with Paul.

"After USENIX."  :-)

--------------------------------------------------------------------------
Jason R. Thorpe                                       thorpej@nas.nasa.gov
NASA Ames Research Center                               Home: 408.866.1912
NAS: M/S 258-6                                          Work: 415.604.0935
Moffett Field, CA 94035                                Pager: 415.428.6939