On Fri, Feb 09, 2018 at 01:51:32PM +0300, Kirill A. Shutemov wrote:
> On Wed, Feb 07, 2018 at 01:30:47PM -0800, Matthew Wilcox wrote:
> > From: Matthew Wilcox <mawilcox@xxxxxxxxxxxxx>
> >
> > We're already using a union of many fields here, so stop abusing the
> > _map_count and make page_type its own field. That implies renaming some
> > of the machinery that creates PageBuddy, PageBalloon and PageKmemcg;
> > bring back the PG_buddy, PG_balloon and PG_kmemcg names.
>
> Sounds reasonable to me.
>
> > Also, the special values don't need to be (and probably shouldn't be) powers
> > of two, so renumber them to just start at -128.
>
> Are you sure about this? Is it guarantee that we would not need in the
> future PG_buddy|PG_kmemcg for instance?
>
> I guess we may want to make it a bitfield. In negative space it's kinda
> interesting. :)
Far too interesting ;-) We still wouldn't want it to be powers of two ...
We'd want to do something like:
#define PAGE_TYPE_BASE 0xfffff000
#define PG_buddy 0x00000080
#define PG_balloon 0x00000100
#define PG_kmemcg 0x00000200
#define PG_pte 0x00000400
... etc ...
I think that's a worthwhile improvement, particularly since I think
PG_kmemcg might want to be a completely independent flag from everything
else ("This is a page allocated for userspace PTEs, so it's both a kmemcg
page and a PTE page").
I'll respin the patch. Thanks!
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>
