Re: Accelerating memset/memcpy

On Tue, Oct 01, 2002 at 06:20:15PM +0100, Nigel Stephens wrote:

> Good point. So there's at least three alternatives to start with: "pref 
> 30", "create_dirty_exclusive" and "none". Let's hope that covers most of 
> them - now we just have to do a survey! But at least "pref 30" should be 
> a safe initial assumption for any MIPS32/MIPS64 processor.

We have a table of CPUs and their capabilities - so selection of
routines can be made CPU dependant without any hassles.  We just
need to decide what to add to the table.  :-)

We also need to decide how to call them. ie, is an indirect function
call overhead acceptable, or do we want to copy the code we want over
a stub (similar to the way we copy the exception vectors)?   [The last
bit doesn't read well but I hope you get the gist of what I mean!]

Simon.
--
Simon Burge                                   <[email protected]>
NetBSD Development, Support and Service:   http://www.wasabisystems.com/