From: Michal Hocko <mhocko@kernel.org>
To: Oscar Salvador <osalvador@suse.de>
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org,
david@redhat.com, mike.kravetz@oracle.com
Subject: Re: [RFC PATCH] mm,memory_hotplug: Unlock 1GB-hugetlb on x86_64
Date: Thu, 28 Feb 2019 10:55:35 +0100 [thread overview]
Message-ID: <20190228095535.GX10588@dhcp22.suse.cz> (raw)
In-Reply-To: <20190228094104.wbeaowsx25ckpcc7@d104.suse.de>
On Thu 28-02-19 10:41:08, Oscar Salvador wrote:
> On Thu, Feb 28, 2019 at 10:21:54AM +0100, Michal Hocko wrote:
> > On Thu 21-02-19 10:42:12, Oscar Salvador wrote:
> > [...]
> > > diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
> > > index d5f7afda67db..04f6695b648c 100644
> > > --- a/mm/memory_hotplug.c
> > > +++ b/mm/memory_hotplug.c
> > > @@ -1337,8 +1337,7 @@ static unsigned long scan_movable_pages(unsigned long start, unsigned long end)
> > > if (!PageHuge(page))
> > > continue;
> > > head = compound_head(page);
> > > - if (hugepage_migration_supported(page_hstate(head)) &&
> > > - page_huge_active(head))
> > > + if (page_huge_active(head))
> > > return pfn;
> > > skip = (1 << compound_order(head)) - (page - head);
> > > pfn += skip - 1;
> >
> > Is this part correct? Say we have a gigantic page which is migrateable.
> > Now scan_movable_pages would skip it and we will not migrate it, no?
>
> All non-migrateable hugepages should have been caught in has_unmovable_pages:
>
> <--
> if (PageHuge(page)) {
> struct page *head = compound_head(page);
> unsigned int skip_pages;
>
> if (!hugepage_migration_supported(page_hstate(head)))
> goto unmovable;
> -->
>
> So, there is no need to check again for migrateability here, as it is something
> that does not change.
> To put it in another way, all huge pages found in scan_movable_pages() should be
> migrateable.
> In scan_movable_pages() we just need to check whether the hugepage, gigantic or not, is
> in use (aka active) to migrate it.
You seemed to miss my point or I am wrong here. If scan_movable_pages
skips over a hugetlb page then there is nothing to migrate it and it
will stay in the pfn range and the range will not become idle.
--
Michal Hocko
SUSE Labs
next prev parent reply other threads:[~2019-02-28 9:55 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-02-21 9:42 [RFC PATCH] mm,memory_hotplug: Unlock 1GB-hugetlb on x86_64 Oscar Salvador
2019-02-21 22:12 ` Mike Kravetz
2019-02-22 8:24 ` Oscar Salvador
2019-02-27 21:51 ` Oscar Salvador
2019-02-27 22:00 ` Mike Kravetz
2019-02-28 7:38 ` David Hildenbrand
2019-02-28 9:16 ` Michal Hocko
2019-02-28 9:21 ` Michal Hocko
2019-02-28 9:41 ` Oscar Salvador
2019-02-28 9:55 ` Michal Hocko [this message]
2019-02-28 10:19 ` Oscar Salvador
2019-02-28 12:11 ` Michal Hocko
2019-02-28 13:40 ` Oscar Salvador
2019-02-28 14:08 ` Michal Hocko
2019-02-28 21:01 ` Oscar Salvador
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190228095535.GX10588@dhcp22.suse.cz \
--to=mhocko@kernel.org \
--cc=david@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mike.kravetz@oracle.com \
--cc=osalvador@suse.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).