Re: [PATCH v15 6/7] virtio-balloon: Add support for providing free page reports to host

[Alexander Duyck <alexander.h.duyck@linux.intel.com>]

On Fri, 2019-12-13 at 02:08 -0500, Michael S. Tsirkin wrote:
> On Thu, Dec 05, 2019 at 08:22:55AM -0800, Alexander Duyck wrote:
> > From: Alexander Duyck <alexander.h.duyck@linux.intel.com>
> > 
> > Add support for the page reporting feature provided by virtio-balloon.
> > Reporting differs from the regular balloon functionality in that is is
> > much less durable than a standard memory balloon. Instead of creating a
> > list of pages that cannot be accessed the pages are only inaccessible
> > while they are being indicated to the virtio interface. Once the
> > interface has acknowledged them they are placed back into their respective
> > free lists and are once again accessible by the guest system.
> > 
> > Unlike a standard balloon we don't inflate and deflate the pages. Instead
> > we perform the reporting, and once the reporting is completed it is
> > assumed that the page has been dropped from the guest and will be faulted
> > back in the next time the page is accessed.
> > 
> > For this reason when I had originally introduced the patch set I referred
> > to this behavior as a "bubble" instead of a "balloon" since the duration
> > is short lived, and when the page is touched the "bubble" is popped and
> > the page is faulted back in.
> > 
> > Signed-off-by: Alexander Duyck <alexander.h.duyck@linux.intel.com>
> 
> virtio POV is fine here:
> 
> Acked-by: Michael S. Tsirkin <mst@redhat.com>
> 
> However please copy virtio-comment on UAPI changes.

So I have been avoiding copying virtio-dev on the kernel changes as I had
gotten feedback that it was annoying some people as they were getting
bounces since they were not subscribed. Will the same type of thing happen
with virtio-comment?

> If possible isolate the last chunk in a patch by itself
> to make it easier for non-kernel developers to review.

Are you talking about the change in "include/uapi/linux/virtio_balloon.h"?

I have it as a standalone patch in the QEMU set, and for the QEMU set I
had included virtio-dev. Would you prefer I include virtio-comment instead
or in addition to virtio-dev? My thought is that I would prefer to keep
the virtio people focused on the QEMU code since they are probably more
comfortable with that, and the kernel people focused on the kernel code.

> > ---
> >  drivers/virtio/Kconfig              |    1 +
> >  drivers/virtio/virtio_balloon.c     |   64 +++++++++++++++++++++++++++++++++++
> >  include/uapi/linux/virtio_balloon.h |    1 +
> >  3 files changed, 66 insertions(+)

<snip>

> > 
> > diff --git a/include/uapi/linux/virtio_balloon.h b/include/uapi/linux/virtio_balloon.h
> > index a1966cd7b677..19974392d324 100644
> > --- a/include/uapi/linux/virtio_balloon.h
> > +++ b/include/uapi/linux/virtio_balloon.h
> > @@ -36,6 +36,7 @@
> >  #define VIRTIO_BALLOON_F_DEFLATE_ON_OOM	2 /* Deflate balloon on OOM */
> >  #define VIRTIO_BALLOON_F_FREE_PAGE_HINT	3 /* VQ to report free pages */
> >  #define VIRTIO_BALLOON_F_PAGE_POISON	4 /* Guest is using page poisoning */
> > +#define VIRTIO_BALLOON_F_REPORTING	5 /* Page reporting virtqueue */
> >  
> >  /* Size of a PFN in the balloon interface. */
> >  #define VIRTIO_BALLOON_PFN_SHIFT 12

If this is the bit we are talking about I have it split out already into a
QEMU specific patch as well, it can be found here:
https://lore.kernel.org/lkml/20191205162422.19737.57728.stgit@localhost.localdomain/

If needed I could probably add a cover page and/or update the comments in
that patch if that is needed to better explain the change.

Thanks.

- Alex