Re: 3gb/1gb (fwd)

On the flip side, a smaller kernel space would mean a smaller RAM disk
size limit (for ones which store the disk in kernel memory).  Or more
contortions to copy data from one memory area to another.  It would
also reduce the space for file system cache memory, I presume.

By the way, is there a kernel call for copying between mutually
inaccessible user memory areas?  Then the RAM disk could be a
separate user level program rather than taking up kernel space.
Wish I had time to look into it myself...

- Alex

Ingo Weinhold wrote on Thu, 04 Sep 2008 22:21:46 +0200:
> On 2008-09-04 at 21:54:23 [+0200], Rene Gollent <[email protected]> wrote:
> > Just wondering how difficult would it be for our VM to support doing a
> > 3GB/1GB address space split instead of 2/2? I realize it might not be
> > so important right now, but as it becomes more common for machines to
> > have 2GB+ it'd probably be helpful in the short term, at least until
> > we support 64-bit natively.
> 
> I don't think it's particularly complicated. I believe it would mainly be 
> changing a few macros (specifying user/kernel address space, and the location 
> of a few other things) and maybe a hardcoded value here and there.
> 
> I might misunderstand your argument, but having more than 2 GB of physical 
> RAM has not much to do with how the address space is split. The only reason 
> for changing the split would be applications using more than 2 GB of virtual 
> memory. ATM I don't see any candidates, though.


-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/