From: Nitesh Narayan Lal <nitesh@redhat.com>
To: Alexander Duyck <alexander.duyck@gmail.com>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org,
kvm@vger.kernel.org
Cc: rkrcmar@redhat.com, alexander.h.duyck@linux.intel.com,
x86@kernel.org, mingo@redhat.com, bp@alien8.de, hpa@zytor.com,
pbonzini@redhat.com, tglx@linutronix.de,
akpm@linux-foundation.org,
Luiz Capitulino <lcapitulino@redhat.com>,
David Hildenbrand <david@redhat.com>
Subject: Re: [RFC PATCH 0/4] kvm: Report unused guest pages to host
Date: Thu, 7 Feb 2019 09:48:27 -0500 [thread overview]
Message-ID: <e0bf61d2-c315-ae4f-6ddb-93b7882fd13f@redhat.com> (raw)
In-Reply-To: <20190204181118.12095.38300.stgit@localhost.localdomain>
[-- Attachment #1.1: Type: text/plain, Size: 3764 bytes --]
On 2/4/19 1:15 PM, Alexander Duyck wrote:
> This patch set provides a mechanism by which guests can notify the host of
> pages that are not currently in use. Using this data a KVM host can more
> easily balance memory workloads between guests and improve overall system
> performance by avoiding unnecessary writing of unused pages to swap.
>
> In order to support this I have added a new hypercall to provided unused
> page hints and made use of mechanisms currently used by PowerPC and s390
> architectures to provide those hints. To reduce the overhead of this call
> I am only using it per huge page instead of of doing a notification per 4K
> page. By doing this we can avoid the expense of fragmenting higher order
> pages, and reduce overall cost for the hypercall as it will only be
> performed once per huge page.
>
> Because we are limiting this to huge pages it was necessary to add a
> secondary location where we make the call as the buddy allocator can merge
> smaller pages into a higher order huge page.
>
> This approach is not usable in all cases. Specifically, when KVM direct
> device assignment is used, the memory for a guest is permanently assigned
> to physical pages in order to support DMA from the assigned device. In
> this case we cannot give the pages back, so the hypercall is disabled by
> the host.
>
> Another situation that can lead to issues is if the page were accessed
> immediately after free. For example, if page poisoning is enabled the
> guest will populate the page *after* freeing it. In this case it does not
> make sense to provide a hint about the page being freed so we do not
> perform the hypercalls from the guest if this functionality is enabled.
Hi Alexander,
Did you get a chance to look at my v8 posting of Guest Free Page Hinting
[1]?
Considering both the solutions are trying to solve the same problem. It
will be great if we can collaborate and come up with a unified solution.
[1] https://lkml.org/lkml/2019/2/4/993
>
> My testing up till now has consisted of setting up 4 8GB VMs on a system
> with 32GB of memory and 4GB of swap. To stress the memory on the system I
> would run "memhog 8G" sequentially on each of the guests and observe how
> long it took to complete the run. The observed behavior is that on the
> systems with these patches applied in both the guest and on the host I was
> able to complete the test with a time of 5 to 7 seconds per guest. On a
> system without these patches the time ranged from 7 to 49 seconds per
> guest. I am assuming the variability is due to time being spent writing
> pages out to disk in order to free up space for the guest.
>
> ---
>
> Alexander Duyck (4):
> madvise: Expose ability to set dontneed from kernel
> kvm: Add host side support for free memory hints
> kvm: Add guest side support for free memory hints
> mm: Add merge page notifier
>
>
> Documentation/virtual/kvm/cpuid.txt | 4 ++
> Documentation/virtual/kvm/hypercalls.txt | 14 ++++++++
> arch/x86/include/asm/page.h | 25 +++++++++++++++
> arch/x86/include/uapi/asm/kvm_para.h | 3 ++
> arch/x86/kernel/kvm.c | 51 ++++++++++++++++++++++++++++++
> arch/x86/kvm/cpuid.c | 6 +++-
> arch/x86/kvm/x86.c | 35 +++++++++++++++++++++
> include/linux/gfp.h | 4 ++
> include/linux/mm.h | 2 +
> include/uapi/linux/kvm_para.h | 1 +
> mm/madvise.c | 13 +++++++-
> mm/page_alloc.c | 2 +
> 12 files changed, 158 insertions(+), 2 deletions(-)
>
> --
--
Regards
Nitesh
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
next prev parent reply other threads:[~2019-02-07 14:48 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-02-04 18:15 [RFC PATCH 0/4] kvm: Report unused guest pages to host Alexander Duyck
2019-02-04 18:15 ` [RFC PATCH 1/4] madvise: Expose ability to set dontneed from kernel Alexander Duyck
2019-02-04 18:15 ` [RFC PATCH 2/4] kvm: Add host side support for free memory hints Alexander Duyck
2019-02-10 0:44 ` Michael S. Tsirkin
2019-02-11 17:34 ` Alexander Duyck
2019-02-11 17:36 ` Michael S. Tsirkin
2019-02-11 17:41 ` Dave Hansen
2019-02-11 17:48 ` Michael S. Tsirkin
2019-02-11 18:30 ` Alexander Duyck
2019-02-11 19:24 ` Michael S. Tsirkin
2019-02-04 18:15 ` [RFC PATCH 3/4] kvm: Add guest " Alexander Duyck
2019-02-04 19:44 ` Dave Hansen
2019-02-04 20:42 ` Alexander Duyck
2019-02-04 23:00 ` Nadav Amit
2019-02-04 23:37 ` Alexander Duyck
2019-02-05 0:03 ` Nadav Amit
2019-02-05 0:16 ` Alexander Duyck
2019-02-05 1:46 ` Nadav Amit
2019-02-05 18:09 ` Alexander Duyck
2019-02-07 18:21 ` Luiz Capitulino
2019-02-07 18:44 ` Alexander Duyck
2019-02-07 20:02 ` Luiz Capitulino
2019-02-08 21:05 ` Nitesh Narayan Lal
2019-02-08 21:31 ` Alexander Duyck
2019-02-10 0:49 ` Michael S. Tsirkin
2019-02-11 16:31 ` Alexander Duyck
2019-02-11 17:36 ` Michael S. Tsirkin
2019-02-11 18:10 ` Alexander Duyck
2019-02-11 19:54 ` Michael S. Tsirkin
2019-02-11 21:00 ` Alexander Duyck
2019-02-11 22:52 ` Michael S. Tsirkin
[not found] ` <94462313ccd927d25675f69de459456cf066c1a2.camel@linux.intel.com>
2019-02-12 0:34 ` Michael S. Tsirkin
2019-02-11 17:48 ` Dave Hansen
2019-02-11 17:58 ` Michael S. Tsirkin
2019-02-11 18:19 ` Dave Hansen
2019-02-11 19:56 ` Michael S. Tsirkin
2019-02-04 18:15 ` [RFC PATCH 4/4] mm: Add merge page notifier Alexander Duyck
2019-02-04 19:40 ` Dave Hansen
2019-02-04 19:51 ` Alexander Duyck
2019-02-10 0:57 ` Michael S. Tsirkin
2019-02-11 13:30 ` Nitesh Narayan Lal
2019-02-11 14:17 ` Michael S. Tsirkin
2019-02-11 16:24 ` Nitesh Narayan Lal
2019-02-11 17:41 ` Michael S. Tsirkin
2019-02-11 18:09 ` Nitesh Narayan Lal
2019-02-11 6:40 ` Aaron Lu
2019-02-11 15:58 ` Alexander Duyck
2019-02-12 2:09 ` Aaron Lu
2019-02-12 17:20 ` Alexander Duyck
2019-02-04 18:19 ` [RFC PATCH QEMU] i386/kvm: Enable paravirtual unused page hint mechanism Alexander Duyck
2019-02-05 17:25 ` [RFC PATCH 0/4] kvm: Report unused guest pages to host Nitesh Narayan Lal
2019-02-05 18:43 ` Alexander Duyck
2019-02-07 14:48 ` Nitesh Narayan Lal [this message]
2019-02-07 16:56 ` Alexander Duyck
2019-02-10 0:51 ` Michael S. Tsirkin
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=e0bf61d2-c315-ae4f-6ddb-93b7882fd13f@redhat.com \
--to=nitesh@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=alexander.duyck@gmail.com \
--cc=alexander.h.duyck@linux.intel.com \
--cc=bp@alien8.de \
--cc=david@redhat.com \
--cc=hpa@zytor.com \
--cc=kvm@vger.kernel.org \
--cc=lcapitulino@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mingo@redhat.com \
--cc=pbonzini@redhat.com \
--cc=rkrcmar@redhat.com \
--cc=tglx@linutronix.de \
--cc=x86@kernel.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).