Subject: Re: more closely scoped source changes for the 'hanging' mount problem
To: None <mjacob@feral.com>
From: Jason Thorpe <thorpej@nas.nasa.gov>
List: port-alpha
Date: 04/07/1999 19:36:27
On Wed, 7 Apr 1999 17:52:01 -0700 (PWT) 
 Matthew Jacob <mjacob@feral.com> wrote:

 > (another in a list of deletable messages about this particular issue)
 > 
 > I started with a working CVS tree tagged as
 > 
 > 	D99.03.26.20.00.00
 > 
 > (that's 'March 26 1999 20:00 GMT') which does not hang for me and a tree
 > tagged as
 > 
 > 	D99.03.27.04.00.00
 > 
 > (that's 'March 27 1999 04:00 GMT') which hangs coming up for me.

Well, there is one way you can tell if these changes actually cause the
problem.  The extra arguments to pmap_enter() make no difference if you
make pmap_enter() ignore them... so you could certainly #if 0 out the
part that pays attention to access_type, build a kernel, and try it...

However, since you updated your userland and the problem no longer
occurs, I guess we'll never know.  *shrug*

        -- Jason R. Thorpe <thorpej@nas.nasa.gov>