[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ZInalCeSNmAiG2K4@x1n>
Date: Wed, 14 Jun 2023 11:19:48 -0400
From: Peter Xu <peterx@...hat.com>
To: Matthew Wilcox <willy@...radead.org>
Cc: linux-kernel@...r.kernel.org, linux-mm@...ck.org,
Andrea Arcangeli <aarcange@...hat.com>,
John Hubbard <jhubbard@...dia.com>,
Mike Rapoport <rppt@...nel.org>,
David Hildenbrand <david@...hat.com>,
Vlastimil Babka <vbabka@...e.cz>,
"Kirill A . Shutemov" <kirill@...temov.name>,
Andrew Morton <akpm@...ux-foundation.org>,
Mike Kravetz <mike.kravetz@...cle.com>,
James Houghton <jthoughton@...gle.com>,
Hugh Dickins <hughd@...gle.com>
Subject: Re: [PATCH 6/7] mm/gup: Accelerate thp gup even for "pages != NULL"
On Wed, Jun 14, 2023 at 03:58:34PM +0100, Matthew Wilcox wrote:
> On Tue, Jun 13, 2023 at 05:53:45PM -0400, Peter Xu wrote:
> > + if (page_increm > 1)
> > + WARN_ON_ONCE(
> > + try_grab_folio(compound_head(page),
>
> You don't need to call compound_head() here; try_grab_folio() works
> on tail pages just fine.
I did it with caution because two things I'm not sure: either
is_pci_p2pdma_page() or is_longterm_pinnable_page() inside, both calls
is_zone_device_page() on the page*.
But I just noticed try_grab_folio() is also used in gup_pte_range() where
the thp can be pte mapped, so I assume we at least need that to handle tail
page well.
Do we perhaps need the compound_head() in try_grab_folio() as a separate
patch? Or maybe I was wrong on is_zone_device_page()?
>
> > + page_increm - 1,
> > + foll_flags) == NULL);
> > +
> > + for (j = 0; j < page_increm; j++) {
> > + subpage = nth_page(page, j);
> > + pages[i+j] = subpage;
> > + flush_anon_page(vma, subpage, start + j * PAGE_SIZE);
> > + flush_dcache_page(subpage);
>
> You're better off calling flush_dcache_folio() right at the end.
Will do.
Thanks,
--
Peter Xu
Powered by blists - more mailing lists