On Tue 08-11-16 16:03:07, Kirill A. Shutemov wrote: > On Tue, Nov 08, 2016 at 11:35:41AM +0100, Jan Kara wrote: > > Hi Kirill, > > > > I've noticed that your commit 83929372f629 (filemap: prepare find and > > delete operations for huge pages) added to page_cache_tree_delete(): > > > > int i, nr = PageHuge(page) ? 1 : hpage_nr_pages(page); > > > > Isn't the logic computing 'nr' inverted? I'd expect that if page is > > !PageHuge, we want to delete just one page... OTOH I'm surprised this would > > not blow up anywhere if it was really wrong so maybe I just miss something. > > No, that's not bug. > > PageHuge() detects hugetlb pages (we probably should rename the helper) > which represented by one entry on radix-tree per huge page. Ah, right. Thanks for explanation! Maybe that line would deserve a comment? Honza -- Jan Kara <jack@xxxxxxxx> SUSE Labs, CR -- 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>