Re: [virtio-dev] Re: [PATCH v7 5/6] virtio-balloon: Pull page poisoning config out of free page hinting

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

On Wed, 2019-09-04 at 15:28 -0400, Michael S. Tsirkin wrote:
> On Wed, Sep 04, 2019 at 08:10:55AM -0700, Alexander Duyck wrote:
> > From: Alexander Duyck <alexander.h.duyck@linux.intel.com>
> > 
> > Currently the page poisoning setting wasn't being enabled unless free page
> > hinting was enabled. However we will need the page poisoning tracking logic
> > as well for unused page reporting. As such pull it out and make it a
> > separate bit of config in the probe function.
> > 
> > In addition we can actually wrap the code in a check for NO_SANITY. If we
> > don't care what is actually in the page we can just default to 0 and leave
> > it there.
> > 
> > Signed-off-by: Alexander Duyck <alexander.h.duyck@linux.intel.com>
> > ---
> >  drivers/virtio/virtio_balloon.c |   19 +++++++++++++------
> >  mm/page_reporting.c             |    4 ++++
> >  2 files changed, 17 insertions(+), 6 deletions(-)
> > 
> > diff --git a/drivers/virtio/virtio_balloon.c b/drivers/virtio/virtio_balloon.c
> > index 226fbb995fb0..2c19457ab573 100644
> > --- a/drivers/virtio/virtio_balloon.c
> > +++ b/drivers/virtio/virtio_balloon.c
> > @@ -842,7 +842,6 @@ static int virtio_balloon_register_shrinker(struct virtio_balloon *vb)
> >  static int virtballoon_probe(struct virtio_device *vdev)
> >  {
> >  	struct virtio_balloon *vb;
> > -	__u32 poison_val;
> >  	int err;
> >  
> >  	if (!vdev->config->get) {
> > @@ -909,11 +908,19 @@ static int virtballoon_probe(struct virtio_device *vdev)
> >  						  VIRTIO_BALLOON_CMD_ID_STOP);
> >  		spin_lock_init(&vb->free_page_list_lock);
> >  		INIT_LIST_HEAD(&vb->free_page_list);
> > -		if (virtio_has_feature(vdev, VIRTIO_BALLOON_F_PAGE_POISON)) {
> > -			memset(&poison_val, PAGE_POISON, sizeof(poison_val));
> > -			virtio_cwrite(vb->vdev, struct virtio_balloon_config,
> > -				      poison_val, &poison_val);
> > -		}
> > +	}
> > +	if (virtio_has_feature(vdev, VIRTIO_BALLOON_F_PAGE_POISON)) {
> > +		__u32 poison_val = 0;
> > +
> > +#if !defined(CONFIG_PAGE_POISONING_NO_SANITY)
> > +		/*
> > +		 * Let hypervisor know that we are expecting a specific
> > +		 * value to be written back in unused pages.
> > +		 */
> > +		memset(&poison_val, PAGE_POISON, sizeof(poison_val));
> > +#endif
> > +		virtio_cwrite(vb->vdev, struct virtio_balloon_config,
> > +			      poison_val, &poison_val);
> >  	}
> >  	/*
> >  	 * We continue to use VIRTIO_BALLOON_F_DEFLATE_ON_OOM to decide if a
> 
> I'm a bit confused by this part. Should we not just clear
> VIRTIO_BALLOON_F_PAGE_POISON completely?
> 
> In my mind the value written should be what guest puts in
> free pages - and possibly what it expects to find there later.

I thought it better to err on the side of more information rather than
less. With this the host knows that page poisoning is enabled, but that 0
is an acceptable value.

> If it doesn't expect anything there then it makes sense
> to clear VIRTIO_BALLOON_F_PAGE_POISON so that host does
> not try to put the poison value there.

That makes sense.

> But I think that it does not make sense to lie to host about the poison
> value - I think that if we do send poison value to
> host it's reasonable for host to expect free pages
> have that value - and even possibly to validate that.
> 
> So I think that the hack belongs in virtballoon_validate,
> near the page_poisoning_enabled check.

Yeah, I will move that for v8. I will just add an IS_ENABLED check to the
!page_poisoning_enabled check in virtballoon_validate and can drop the
#ifdef check.

Thanks.

- Alex