[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20190916113729.GF10231@dhcp22.suse.cz>
Date: Mon, 16 Sep 2019 13:37:29 +0200
From: Michal Hocko <mhocko@...nel.org>
To: Vlastimil Babka <vbabka@...e.cz>
Cc: "Kirill A. Shutemov" <kirill@...temov.name>,
Andrew Morton <akpm@...ux-foundation.org>, linux-mm@...ck.org,
linux-kernel@...r.kernel.org,
"Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>
Subject: Re: [PATCH] mm, thp: Do not queue fully unmapped pages for deferred
split
On Mon 16-09-19 13:11:37, Vlastimil Babka wrote:
> On 9/16/19 12:36 PM, Michal Hocko wrote:
> > On Fri 13-09-19 12:18:49, Kirill A. Shutemov wrote:
> >> Adding fully unmapped pages into deferred split queue is not productive:
> >> these pages are about to be freed or they are pinned and cannot be split
> >> anyway.
> >>
> >> Signed-off-by: Kirill A. Shutemov <kirill.shutemov@...ux.intel.com>
> >> ---
> >> mm/rmap.c | 14 ++++++++++----
> >> 1 file changed, 10 insertions(+), 4 deletions(-)
> >>
> >> diff --git a/mm/rmap.c b/mm/rmap.c
> >> index 003377e24232..45388f1bf317 100644
> >> --- a/mm/rmap.c
> >> +++ b/mm/rmap.c
> >> @@ -1271,12 +1271,20 @@ static void page_remove_anon_compound_rmap(struct page *page)
> >> if (TestClearPageDoubleMap(page)) {
> >> /*
> >> * Subpages can be mapped with PTEs too. Check how many of
> >> - * themi are still mapped.
> >> + * them are still mapped.
> >> */
> >> for (i = 0, nr = 0; i < HPAGE_PMD_NR; i++) {
> >> if (atomic_add_negative(-1, &page[i]._mapcount))
> >> nr++;
> >> }
> >> +
> >> + /*
> >> + * Queue the page for deferred split if at least one small
> >> + * page of the compound page is unmapped, but at least one
> >> + * small page is still mapped.
> >> + */
> >> + if (nr && nr < HPAGE_PMD_NR)
> >> + deferred_split_huge_page(page);
> >
> > You've set nr to zero in the for loop so this cannot work AFAICS.
>
> The for loop then does nr++ for each subpage that's still mapped?
I am blind obviously. Sorry about the noise. Then the patch looks
correct.
--
Michal Hocko
SUSE Labs
Powered by blists - more mailing lists