Re: [PATCH v11 4/6] mm: function to offer a page block on the free list

From	Christian Borntraeger <borntraeger@de.ibm.com>
Newsgroups	linux.kernel
Subject	Re: [PATCH v11 4/6] mm: function to offer a page block on the free list
Date	2017-06-21 15:00 +0200
Message-ID	<tUNoK-6da-19@gated-at.bofh.it> (permalink)
References	<tQpEl-qv-1@gated-at.bofh.it> <tQpEl-qv-7@gated-at.bofh.it> <tRyme-3aL-9@gated-at.bofh.it> <tUuvM-2D0-1@gated-at.bofh.it> <tUuvM-2D0-15@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On 06/20/2017 06:49 PM, David Hildenbrand wrote:
> On 20.06.2017 18:44, Rik van Riel wrote:
>> On Mon, 2017-06-12 at 07:10 -0700, Dave Hansen wrote:
>>
>>> The hypervisor is going to throw away the contents of these pages,
>>> right?  As soon as the spinlock is released, someone can allocate a
>>> page, and put good data in it.  What keeps the hypervisor from
>>> throwing
>>> away good data?
>>
>> That looks like it may be the wrong API, then?
>>
>> We already have hooks called arch_free_page and
>> arch_alloc_page in the VM, which are called when
>> pages are freed, and allocated, respectively.
>>
>> Nitesh Lal (on the CC list) is working on a way
>> to efficiently batch recently freed pages for
>> free page hinting to the hypervisor.
>>
>> If that is done efficiently enough (eg. with
>> MADV_FREE on the hypervisor side for lazy freeing,
>> and lazy later re-use of the pages), do we still
>> need the harder to use batch interface from this
>> patch?
>>
> David's opinion incoming:
> 
> No, I think proper free page hinting would be the optimum solution, if
> done right. This would avoid the batch interface and even turn
> virtio-balloon in some sense useless.
> 
Two reasons why I disagree:
- virtio-balloon is often used as memory hotplug. (e.g. libvirts current/max memory
uses virtio ballon)
- free page hinting will not allow to shrink the page cache of guests (like a ballooner does)

Thread

Re: [PATCH v11 4/6] mm: function to offer a page block on the free  list Rik van Riel <riel@redhat.com> - 2017-06-20 18:50 +0200
  Re: [PATCH v11 4/6] mm: function to offer a page block on the free  list David Hildenbrand <david@redhat.com> - 2017-06-20 18:50 +0200
    Re: [PATCH v11 4/6] mm: function to offer a page block on the free  list Rik van Riel <riel@redhat.com> - 2017-06-20 19:30 +0200
      Re: [PATCH v11 4/6] mm: function to offer a page block on the free  list "Michael S. Tsirkin" <mst@redhat.com> - 2017-06-20 20:30 +0200
        Re: [PATCH v11 4/6] mm: function to offer a page block on the free  list Rik van Riel <riel@redhat.com> - 2017-06-20 22:00 +0200
          Re: [PATCH v11 4/6] mm: function to offer a page block on the free  list "Michael S. Tsirkin" <mst@redhat.com> - 2017-06-21 14:50 +0200
      Re: [Qemu-devel] [PATCH v11 4/6] mm: function to offer a page block  on the free list Wei Wang <wei.w.wang@intel.com> - 2017-06-21 10:40 +0200
    Re: [PATCH v11 4/6] mm: function to offer a page block on the free  list "Michael S. Tsirkin" <mst@redhat.com> - 2017-06-20 20:20 +0200
      Re: [PATCH v11 4/6] mm: function to offer a page block on the free  list "Michael S. Tsirkin" <mst@redhat.com> - 2017-06-20 21:00 +0200
        Re: [PATCH v11 4/6] mm: function to offer a page block on the free  list David Hildenbrand <david@redhat.com> - 2017-06-20 21:10 +0200
      Re: [PATCH v11 4/6] mm: function to offer a page block on the free  list David Hildenbrand <david@redhat.com> - 2017-06-20 21:00 +0200
    Re: [PATCH v11 4/6] mm: function to offer a page block on the free  list Christian Borntraeger <borntraeger@de.ibm.com> - 2017-06-21 15:00 +0200
      Re: [PATCH v11 4/6] mm: function to offer a page block on the free  list David Hildenbrand <david@redhat.com> - 2017-06-21 15:50 +0200

csiph-web