On Wed, 25 Mar 2015, Kirill A. Shutemov wrote: > > We only need tail refcounting for THP, so I think this should fix the issue: > > diff --git a/include/linux/mm.h b/include/linux/mm.h > index 4a3a38522ab4..9ab432660adb 100644 > --- a/include/linux/mm.h > +++ b/include/linux/mm.h > @@ -456,7 +456,7 @@ static inline int page_count(struct page *page) > > static inline bool __compound_tail_refcounted(struct page *page) > { > - return !PageSlab(page) && !PageHeadHuge(page); > + return !PageSlab(page) && !PageHeadHuge(page) && PageAnon(page); > } > > /* Yes, that should be a good fix for the mapcount issue. And no coincidence that it's just what I needed too, when reusing the PG_compound_lock bit: see my 10/24 (which had to rearrange mm.h. not having your 1/16). Hugh -- 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>