From: Pavel Tatashin <pasha.tatashin@soleen.com>
To: Joao Martins <joao.m.martins@oracle.com>
Cc: Jason Gunthorpe <jgg@ziepe.ca>,
LKML <linux-kernel@vger.kernel.org>,
linux-mm <linux-mm@kvack.org>,
Andrew Morton <akpm@linux-foundation.org>,
Vlastimil Babka <vbabka@suse.cz>, Michal Hocko <mhocko@suse.com>,
David Hildenbrand <david@redhat.com>,
Oscar Salvador <osalvador@suse.de>,
Dan Williams <dan.j.williams@intel.com>,
Sasha Levin <sashal@kernel.org>,
Tyler Hicks <tyhicks@linux.microsoft.com>,
Joonsoo Kim <iamjoonsoo.kim@lge.com>,
Mike Kravetz <mike.kravetz@oracle.com>,
Steven Rostedt <rostedt@goodmis.org>,
Ingo Molnar <mingo@redhat.com>,
Peter Zijlstra <peterz@infradead.org>,
Mel Gorman <mgorman@suse.de>,
Matthew Wilcox <willy@infradead.org>,
David Rientjes <rientjes@google.com>,
John Hubbard <jhubbard@nvidia.com>,
Linux Doc Mailing List <linux-doc@vger.kernel.org>,
Ira Weiny <ira.weiny@intel.com>,
linux-kselftest@vger.kernel.org, James Morris <jmorris@namei.org>
Subject: Re: [PATCH v8 02/14] mm/gup: check every subpage of a compound page during isolation
Date: Wed, 3 Feb 2021 09:51:32 -0500 [thread overview]
Message-ID: <CA+CK2bBSJ7T=jsukntQGqO0DoWE_MnhDwtHv-6rfXAPvznKh0Q@mail.gmail.com> (raw)
In-Reply-To: <05a66361-214c-2afe-22e4-12862ea1e4e2@oracle.com>
On Wed, Feb 3, 2021 at 8:23 AM Joao Martins <joao.m.martins@oracle.com> wrote:
>
> On 1/25/21 7:47 PM, Pavel Tatashin wrote:
> > When pages are isolated in check_and_migrate_movable_pages() we skip
> > compound number of pages at a time. However, as Jason noted, it is
> > not necessary correct that pages[i] corresponds to the pages that
> > we skipped. This is because it is possible that the addresses in
> > this range had split_huge_pmd()/split_huge_pud(), and these functions
> > do not update the compound page metadata.
> >
> > The problem can be reproduced if something like this occurs:
> >
> > 1. User faulted huge pages.
> > 2. split_huge_pmd() was called for some reason
> > 3. User has unmapped some sub-pages in the range
> > 4. User tries to longterm pin the addresses.
> >
> > The resulting pages[i] might end-up having pages which are not compound
> > size page aligned.
> >
> > Fixes: aa712399c1e8 ("mm/gup: speed up check_and_migrate_cma_pages() on huge page")
> > Reported-by: Jason Gunthorpe <jgg@nvidia.com>
> > Signed-off-by: Pavel Tatashin <pasha.tatashin@soleen.com>
> > Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
> > ---
>
> [...]
>
> > /*
> > * If we get a page from the CMA zone, since we are going to
> > * be pinning these entries, we might as well move them out
> > @@ -1599,8 +1596,6 @@ static long check_and_migrate_cma_pages(struct mm_struct *mm,
> > }
> > }
> > }
> > -
> > - i += step;
> > }
> >
>
Hi Joao,
> With this, longterm gup will 'regress' for hugetlbfs e.g. from ~6k -> 32k usecs when
> pinning a 16G hugetlb file.
Estimate or you actually measured?
>
> Splitting can only occur on THP right? If so, perhaps we could retain the @step increment
Yes, I do not think we can split HugePage, only THP.
> for compound pages but when !is_transparent_hugepage(head) or just PageHuge(head) like:
>
> + if (!is_transparent_hugepage(head) && PageCompound(page))
> + i += (compound_nr(head) - (pages[i] - head));
>
> Or making specific to hugetlbfs:
>
> + if (PageHuge(head))
> + i += (compound_nr(head) - (pages[i] - head));
Yes, this is reasonable optimization. I will submit a follow up patch
against linux-next.
Thank you,
Pasha
next prev parent reply other threads:[~2021-02-03 14:52 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-01-25 19:47 [PATCH v8 00/14] prohibit pinning pages in ZONE_MOVABLE Pavel Tatashin
2021-01-25 19:47 ` [PATCH v8 01/14] mm/gup: don't pin migrated cma pages in movable zone Pavel Tatashin
2021-01-25 19:47 ` [PATCH v8 02/14] mm/gup: check every subpage of a compound page during isolation Pavel Tatashin
2021-02-03 13:22 ` Joao Martins
2021-02-03 14:51 ` Pavel Tatashin [this message]
2021-02-03 15:32 ` Joao Martins
2021-02-03 16:46 ` Joao Martins
2021-02-03 14:53 ` Jason Gunthorpe
2021-02-03 16:13 ` Joao Martins
2021-02-03 18:00 ` Jason Gunthorpe
2021-01-25 19:47 ` [PATCH v8 03/14] mm/gup: return an error on migration failure Pavel Tatashin
2021-01-25 19:47 ` [PATCH v8 04/14] mm/gup: check for isolation errors Pavel Tatashin
2021-01-25 19:47 ` [PATCH v8 05/14] mm cma: rename PF_MEMALLOC_NOCMA to PF_MEMALLOC_PIN Pavel Tatashin
2021-01-25 19:47 ` [PATCH v8 06/14] mm: apply per-task gfp constraints in fast path Pavel Tatashin
2021-01-25 19:47 ` [PATCH v8 07/14] mm: honor PF_MEMALLOC_PIN for all movable pages Pavel Tatashin
[not found] ` <20210131130903.4875-1-lecopzer@gmail.com>
2021-01-31 16:03 ` Pavel Tatashin
2021-01-25 19:47 ` [PATCH v8 08/14] mm/gup: do not migrate zero page Pavel Tatashin
2021-01-25 19:47 ` [PATCH v8 09/14] mm/gup: migrate pinned pages out of movable zone Pavel Tatashin
2021-01-25 19:47 ` [PATCH v8 10/14] memory-hotplug.rst: add a note about ZONE_MOVABLE and page pinning Pavel Tatashin
2021-01-25 19:47 ` [PATCH v8 11/14] mm/gup: change index type to long as it counts pages Pavel Tatashin
2021-01-25 19:47 ` [PATCH v8 12/14] mm/gup: longterm pin migration cleanup Pavel Tatashin
2021-01-25 19:47 ` [PATCH v8 13/14] selftests/vm: gup_test: fix test flag Pavel Tatashin
2021-01-25 19:47 ` [PATCH v8 14/14] selftests/vm: gup_test: test faulting in kernel, and verify pinnable pages Pavel Tatashin
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='CA+CK2bBSJ7T=jsukntQGqO0DoWE_MnhDwtHv-6rfXAPvznKh0Q@mail.gmail.com' \
--to=pasha.tatashin@soleen.com \
--cc=akpm@linux-foundation.org \
--cc=dan.j.williams@intel.com \
--cc=david@redhat.com \
--cc=iamjoonsoo.kim@lge.com \
--cc=ira.weiny@intel.com \
--cc=jgg@ziepe.ca \
--cc=jhubbard@nvidia.com \
--cc=jmorris@namei.org \
--cc=joao.m.martins@oracle.com \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-kselftest@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@suse.de \
--cc=mhocko@suse.com \
--cc=mike.kravetz@oracle.com \
--cc=mingo@redhat.com \
--cc=osalvador@suse.de \
--cc=peterz@infradead.org \
--cc=rientjes@google.com \
--cc=rostedt@goodmis.org \
--cc=sashal@kernel.org \
--cc=tyhicks@linux.microsoft.com \
--cc=vbabka@suse.cz \
--cc=willy@infradead.org \
/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).