Re: Introducing non-RAM managed page ("device page")

[Jason Thorpe <thorpej%shagadelic.org@localhost>]

On Sep 15, 2009, at 9:04 AM, Eduardo Horvath wrote:

> We've handled mmap()ing device memory forever without getting VM  
> involved.
> Not to imply that the current device mmap() interface could not be
> improved.  What features of the VM subsystem do you think is needed  
> to map
> device pages into processes?

That's done with a special pager, the device pager.  We need to allow  
the vnode pager to talk to device memory for XIP.

-- thorpej