Subject: Re: cpu_switch (was Re: 1.5 Release documentation ...)
To: Chris Gilbert <chris@buzzbee.freeserve.co.uk>
From: Neil A. Carson <neil@causality.com>
List: port-arm32
Date: 11/07/2000 10:46:10
Chris Gilbert wrote:
> 
> On Tue, 07 Nov 2000, Neil A. Carson wrote:
> > I'll bet compared to the cache invalidates you'll never notice the 12
> > loops...
> 
> Where would you recommend working to get better performance?  At one time the
> pmap_* stuff was a mess, and that was expected to be improved.  And I note
> that you did some work on it, so is it safe to assume that pmap is not a
> problem any more?

I had some lazy context switch stuff that postponed context switch to
userret a long time ago, forcing the issue if any copyin/copyouts etc.
happened. It was quite hacky but made an improvement.

Not sure what happened to that code.

I think that was the penultimate thing that would have made a noticable
difference...

	Neil