Messages in this thread | | | From | KOSAKI Motohiro <> | Subject | Re: [PATCH 1/4] mm: m(un)lock avoid ZERO_PAGE | Date | Wed, 16 Sep 2009 09:08:15 +0900 (JST) |
| |
> I'm still reluctant to clutter __get_user_pages() with another flag, > just to avoid touching ZERO_PAGE count in mlock(); though we can add > that later if it shows up as an issue in practice. > > But when mlocking, we can test page->mapping slightly earlier, to avoid > the potentially bouncy rescheduling of lock_page on ZERO_PAGE - mlock > didn't lock_page in olden ZERO_PAGE days, so we might have regressed. > > And when munlocking, it turns out that FOLL_DUMP coincidentally does > what's needed to avoid all updates to ZERO_PAGE, so use that here also. > Plus add comment suggested by KAMEZAWA Hiroyuki. > > Signed-off-by: Hugh Dickins <hugh.dickins@tiscali.co.uk> > --- > > mm/mlock.c | 49 ++++++++++++++++++++++++++++++++++++------------- > 1 file changed, 36 insertions(+), 13 deletions(-) > > --- mm0/mm/mlock.c 2009-09-14 16:34:37.000000000 +0100 > +++ mm1/mm/mlock.c 2009-09-15 17:32:03.000000000 +0100 > @@ -198,17 +198,26 @@ static long __mlock_vma_pages_range(stru > for (i = 0; i < ret; i++) { > struct page *page = pages[i]; > > - lock_page(page); > - /* > - * Because we lock page here and migration is blocked > - * by the elevated reference, we need only check for > - * file-cache page truncation. This page->mapping > - * check also neatly skips over the ZERO_PAGE(), > - * though if that's common we'd prefer not to lock it. > - */ > - if (page->mapping) > - mlock_vma_page(page); > - unlock_page(page); > + if (page->mapping) { > + /* > + * That preliminary check is mainly to avoid > + * the pointless overhead of lock_page on the > + * ZERO_PAGE: which might bounce very badly if > + * there is contention. However, we're still > + * dirtying its cacheline with get/put_page: > + * we'll add another __get_user_pages flag to > + * avoid it if that case turns out to matter. > + */ > + lock_page(page); > + /* > + * Because we lock page here and migration is > + * blocked by the elevated reference, we need > + * only check for file-cache page truncation. > + */ > + if (page->mapping) > + mlock_vma_page(page); > + unlock_page(page); > + } > put_page(page); /* ref from get_user_pages() */ > }
Yes, I have similar patch :-) Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
| |