From: "Kirill A. Shutemov" <kirill@shutemov.name>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>,
Andrea Arcangeli <aarcange@redhat.com>,
linux-mm@kvack.org, Andi Kleen <ak@linux.intel.com>,
"H. Peter Anvin" <hpa@linux.intel.com>,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH v4 10/10] thp: implement refcounting for huge zero page
Date: Fri, 19 Oct 2012 02:59:41 +0300 [thread overview]
Message-ID: <20121018235941.GA32397@shutemov.name> (raw)
In-Reply-To: <20121018164502.b32791e7.akpm@linux-foundation.org>
On Thu, Oct 18, 2012 at 04:45:02PM -0700, Andrew Morton wrote:
> On Mon, 15 Oct 2012 09:00:59 +0300
> "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com> wrote:
>
> > H. Peter Anvin doesn't like huge zero page which sticks in memory forever
> > after the first allocation. Here's implementation of lockless refcounting
> > for huge zero page.
> >
> > We have two basic primitives: {get,put}_huge_zero_page(). They
> > manipulate reference counter.
> >
> > If counter is 0, get_huge_zero_page() allocates a new huge page and
> > takes two references: one for caller and one for shrinker. We free the
> > page only in shrinker callback if counter is 1 (only shrinker has the
> > reference).
> >
> > put_huge_zero_page() only decrements counter. Counter is never zero
> > in put_huge_zero_page() since shrinker holds on reference.
> >
> > Freeing huge zero page in shrinker callback helps to avoid frequent
> > allocate-free.
>
> I'd like more details on this please. The cost of freeing then
> reinstantiating that page is tremendous, because it has to be zeroed
> out again. If there is any way at all in which the kernel can be made
> to enter a high-frequency free/reinstantiate pattern then I expect the
> effects would be quite bad.
>
> Do we have sufficient mechanisms in there to prevent this from
> happening in all cases? If so, what are they, because I'm not seeing
> them?
We only free huge zero page in shrinker callback if nobody in the system
uses it. Never on put_huge_zero_page(). Shrinker runs only under memory
pressure or if user asks (drop_caches).
Do you think we need an additional protection mechanism?
>
> > Refcounting has cost. On 4 socket machine I observe ~1% slowdown on
> > parallel (40 processes) read page faulting comparing to lazy huge page
> > allocation. I think it's pretty reasonable for synthetic benchmark.
--
Kirill A. Shutemov
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
WARNING: multiple messages have this Message-ID (diff)
From: "Kirill A. Shutemov" <kirill@shutemov.name>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>,
Andrea Arcangeli <aarcange@redhat.com>,
linux-mm@kvack.org, Andi Kleen <ak@linux.intel.com>,
"H. Peter Anvin" <hpa@linux.intel.com>,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH v4 10/10] thp: implement refcounting for huge zero page
Date: Fri, 19 Oct 2012 02:59:41 +0300 [thread overview]
Message-ID: <20121018235941.GA32397@shutemov.name> (raw)
In-Reply-To: <20121018164502.b32791e7.akpm@linux-foundation.org>
On Thu, Oct 18, 2012 at 04:45:02PM -0700, Andrew Morton wrote:
> On Mon, 15 Oct 2012 09:00:59 +0300
> "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com> wrote:
>
> > H. Peter Anvin doesn't like huge zero page which sticks in memory forever
> > after the first allocation. Here's implementation of lockless refcounting
> > for huge zero page.
> >
> > We have two basic primitives: {get,put}_huge_zero_page(). They
> > manipulate reference counter.
> >
> > If counter is 0, get_huge_zero_page() allocates a new huge page and
> > takes two references: one for caller and one for shrinker. We free the
> > page only in shrinker callback if counter is 1 (only shrinker has the
> > reference).
> >
> > put_huge_zero_page() only decrements counter. Counter is never zero
> > in put_huge_zero_page() since shrinker holds on reference.
> >
> > Freeing huge zero page in shrinker callback helps to avoid frequent
> > allocate-free.
>
> I'd like more details on this please. The cost of freeing then
> reinstantiating that page is tremendous, because it has to be zeroed
> out again. If there is any way at all in which the kernel can be made
> to enter a high-frequency free/reinstantiate pattern then I expect the
> effects would be quite bad.
>
> Do we have sufficient mechanisms in there to prevent this from
> happening in all cases? If so, what are they, because I'm not seeing
> them?
We only free huge zero page in shrinker callback if nobody in the system
uses it. Never on put_huge_zero_page(). Shrinker runs only under memory
pressure or if user asks (drop_caches).
Do you think we need an additional protection mechanism?
>
> > Refcounting has cost. On 4 socket machine I observe ~1% slowdown on
> > parallel (40 processes) read page faulting comparing to lazy huge page
> > allocation. I think it's pretty reasonable for synthetic benchmark.
--
Kirill A. Shutemov
next prev parent reply other threads:[~2012-10-18 23:58 UTC|newest]
Thread overview: 63+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-10-15 6:00 [PATCH v4 00/10, REBASED] Introduce huge zero page Kirill A. Shutemov
2012-10-15 6:00 ` Kirill A. Shutemov
2012-10-15 6:00 ` [PATCH v4 01/10] thp: huge zero page: basic preparation Kirill A. Shutemov
2012-10-15 6:00 ` Kirill A. Shutemov
2012-10-15 6:00 ` [PATCH v4 02/10] thp: zap_huge_pmd(): zap huge zero pmd Kirill A. Shutemov
2012-10-15 6:00 ` Kirill A. Shutemov
2012-10-15 6:00 ` [PATCH v4 03/10] thp: copy_huge_pmd(): copy huge zero page Kirill A. Shutemov
2012-10-15 6:00 ` Kirill A. Shutemov
2012-10-15 6:00 ` [PATCH v4 04/10] thp: do_huge_pmd_wp_page(): handle " Kirill A. Shutemov
2012-10-15 6:00 ` Kirill A. Shutemov
2012-10-15 6:00 ` [PATCH v4 05/10] thp: change_huge_pmd(): keep huge zero page write-protected Kirill A. Shutemov
2012-10-15 6:00 ` Kirill A. Shutemov
2012-10-15 6:00 ` [PATCH v4 06/10] thp: change split_huge_page_pmd() interface Kirill A. Shutemov
2012-10-15 6:00 ` Kirill A. Shutemov
2012-10-15 6:00 ` [PATCH v4 07/10] thp: implement splitting pmd for huge zero page Kirill A. Shutemov
2012-10-15 6:00 ` Kirill A. Shutemov
2012-10-15 6:00 ` [PATCH v4 08/10] thp: setup huge zero page on non-write page fault Kirill A. Shutemov
2012-10-15 6:00 ` Kirill A. Shutemov
2012-10-15 6:00 ` [PATCH v4 09/10] thp: lazy huge zero page allocation Kirill A. Shutemov
2012-10-15 6:00 ` Kirill A. Shutemov
2012-10-15 6:00 ` [PATCH v4 10/10] thp: implement refcounting for huge zero page Kirill A. Shutemov
2012-10-15 6:00 ` Kirill A. Shutemov
2012-10-18 23:45 ` Andrew Morton
2012-10-18 23:45 ` Andrew Morton
2012-10-18 23:59 ` Kirill A. Shutemov [this message]
2012-10-18 23:59 ` Kirill A. Shutemov
2012-10-23 6:35 ` Kirill A. Shutemov
2012-10-23 6:35 ` Kirill A. Shutemov
2012-10-23 6:43 ` Andrew Morton
2012-10-23 6:43 ` Andrew Morton
2012-10-23 7:00 ` Kirill A. Shutemov
2012-10-23 22:59 ` Andrew Morton
2012-10-23 22:59 ` Andrew Morton
2012-10-23 23:38 ` Kirill A. Shutemov
2012-10-23 23:38 ` Kirill A. Shutemov
2012-10-24 19:22 ` Andrew Morton
2012-10-24 19:22 ` Andrew Morton
2012-10-24 19:45 ` Kirill A. Shutemov
2012-10-24 20:25 ` Andrew Morton
2012-10-24 20:25 ` Andrew Morton
2012-10-24 20:33 ` Kirill A. Shutemov
2012-10-24 20:44 ` Andi Kleen
2012-10-24 20:44 ` Andi Kleen
2012-10-25 20:49 ` Kirill A. Shutemov
2012-10-25 21:05 ` Andrew Morton
2012-10-25 21:05 ` Andrew Morton
2012-10-25 21:22 ` Kirill A. Shutemov
2012-10-25 21:22 ` Kirill A. Shutemov
2012-10-25 21:37 ` Andrew Morton
2012-10-25 21:37 ` Andrew Morton
2012-10-25 22:10 ` Kirill A. Shutemov
2012-10-16 9:53 ` [PATCH v4 00/10, REBASED] Introduce " Ni zhan Chen
2012-10-16 9:53 ` Ni zhan Chen
2012-10-16 10:54 ` Kirill A. Shutemov
2012-10-16 10:54 ` Kirill A. Shutemov
2012-10-16 11:13 ` Ni zhan Chen
2012-10-16 11:13 ` Ni zhan Chen
2012-10-16 11:28 ` Kirill A. Shutemov
2012-10-16 11:28 ` Kirill A. Shutemov
2012-10-16 11:37 ` Ni zhan Chen
2012-10-16 11:37 ` Ni zhan Chen
2012-10-26 15:14 ` [PATCH] thp, vmstat: implement HZP_ALLOC and HZP_ALLOC_FAILED events Kirill A. Shutemov
2012-10-26 15:14 ` Kirill A. Shutemov
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=20121018235941.GA32397@shutemov.name \
--to=kirill@shutemov.name \
--cc=aarcange@redhat.com \
--cc=ak@linux.intel.com \
--cc=akpm@linux-foundation.org \
--cc=hpa@linux.intel.com \
--cc=kirill.shutemov@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.