From: Joao Martins <joao.m.martins@oracle.com>
To: Jason Gunthorpe <jgg@ziepe.ca>
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org,
linux-rdma@vger.kernel.org,
Andrew Morton <akpm@linux-foundation.org>,
Doug Ledford <dledford@redhat.com>,
John Hubbard <jhubbard@nvidia.com>,
Matthew Wilcox <willy@infradead.org>
Subject: Re: [PATCH v3 2/4] mm/gup: decrement head page once for group of subpages
Date: Thu, 11 Feb 2021 10:14:59 +0000 [thread overview]
Message-ID: <10989552-b759-bd5c-985c-fbb1753731a2@oracle.com> (raw)
In-Reply-To: <20210210210242.GQ4718@ziepe.ca>
On 2/10/21 9:02 PM, Jason Gunthorpe wrote:
> On Fri, Feb 05, 2021 at 08:41:25PM +0000, Joao Martins wrote:
>> Rather than decrementing the head page refcount one by one, we
>> walk the page array and checking which belong to the same
>> compound_head. Later on we decrement the calculated amount
>> of references in a single write to the head page. To that
>> end switch to for_each_compound_head() does most of the work.
>>
>> set_page_dirty() needs no adjustment as it's a nop for
>> non-dirty head pages and it doesn't operate on tail pages.
>>
>> This considerably improves unpinning of pages with THP and
>> hugetlbfs:
>>
>> - THP
>> gup_test -t -m 16384 -r 10 [-L|-a] -S -n 512 -w
>> PIN_LONGTERM_BENCHMARK (put values): ~87.6k us -> ~23.2k us
>>
>> - 16G with 1G huge page size
>> gup_test -f /mnt/huge/file -m 16384 -r 10 [-L|-a] -S -n 512 -w
>> PIN_LONGTERM_BENCHMARK: (put values): ~87.6k us -> ~27.5k us
>>
>> Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
>> Reviewed-by: John Hubbard <jhubbard@nvidia.com>
>> ---
>> mm/gup.c | 29 +++++++++++------------------
>> 1 file changed, 11 insertions(+), 18 deletions(-)
>
> Looks fine
>
> Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
>
Thanks!
> I was wondering why this only touches the FOLL_PIN path,
That's just because I was looking at pinning mostly.
> it would make
> sense to also use this same logic for release_pages()
Yeah, indeed -- any place tearing potentially consecutive sets of pages
are candidates.
>
> for (i = 0; i < nr; i++) {
> struct page *page = pages[i];
> page = compound_head(page);
> if (is_huge_zero_page(page))
> continue;
>
> Jason
>
next prev parent reply other threads:[~2021-02-11 10:15 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-05 20:41 [PATCH v3 0/4] mm/gup: page unpining improvements Joao Martins
2021-02-05 20:41 ` [PATCH v3 1/4] mm/gup: add compound page list iterator Joao Martins
2021-02-10 23:20 ` Jason Gunthorpe
2021-02-11 10:45 ` Joao Martins
2021-02-05 20:41 ` [PATCH v3 2/4] mm/gup: decrement head page once for group of subpages Joao Martins
2021-02-10 21:02 ` Jason Gunthorpe
2021-02-11 10:14 ` Joao Martins [this message]
2021-02-05 20:41 ` [PATCH v3 3/4] mm/gup: add a range variant of unpin_user_pages_dirty_lock() Joao Martins
2021-02-10 23:15 ` Jason Gunthorpe
2021-02-11 10:24 ` Joao Martins
2021-02-10 23:19 ` John Hubbard
2021-02-11 10:52 ` Joao Martins
2021-02-05 20:41 ` [PATCH v3 4/4] RDMA/umem: batch page unpin in __ib_umem_release() Joao Martins
2021-02-10 23:17 ` Jason Gunthorpe
2021-02-11 10:35 ` Joao Martins
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=10989552-b759-bd5c-985c-fbb1753731a2@oracle.com \
--to=joao.m.martins@oracle.com \
--cc=akpm@linux-foundation.org \
--cc=dledford@redhat.com \
--cc=jgg@ziepe.ca \
--cc=jhubbard@nvidia.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-rdma@vger.kernel.org \
--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).