Re: Active Memory Defragmentation: Our implementation & problems

[Dave Hansen]

On Tue, 2004-02-03 at 22:57, IWAMOTO Toshihiro wrote:
> At 03 Feb 2004 21:54:34 -0800,
> Dave Hansen wrote:
> > Moving file-backed pages is mostly handled already.  You can do a
> > regular page-cache lookup with find_get_page(), make your copy,
> > invalidate the old one, then readd the new one.  The invalidation can be
> > done in the same style as shrink_list().
> 
> Actually, it is a bit more complicated.
> I have implemented similar functionality for memory hotremoval.
> 
> See my post about memory hotremoval
> http://marc.theaimsgroup.com/?l=linux-kernel&m=107354781130941&w=2
> for details.
> remap_onepage() and remapd() in the patch are the main functions.

remap_onepage() is quite a function.  300 lines.  It sure does cover a
lot of ground. :)

Defragmentation is a bit easier than removal because it isn't as
mandatory.  Instead of having to worry about waiting on things like
writeback, the defrag code can just bail.  

--dave

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/