On Mon, Jan 31, 2022 at 03:27:02PM +0100, David Hildenbrand wrote: > On 11.01.22 00:15, Matthew Wilcox (Oracle) wrote: > > There isn't enough information to make this a useful check any more; > > the useful parts of it were moved in earlier patches, so remove this > > set of checks now. > > > > Signed-off-by: Matthew Wilcox (Oracle) <willy@xxxxxxxxxxxxx> > > > Very nice cleanup IMHO > > Reviewed-by: David Hildenbrand <david@xxxxxxxxxx> Thanks. I'm thinking about adding another check. If the order-0 page has a non-NULL ->mapping, I believe that is enough to determine that it's a page cache / KSM / Anon page, and as such is guaranteed to actually be order-0 (and not part of a postive-order allocation without __GFP_COMP set). But that would require a bit more auditing, or just throwing it in and seeing what breaks *whistles*.
