search for: must_tell_host

VIRTIO_BALLOON_F_MUST_TELL_HOST cannot be used properly because it is a "negative" feature: it tells you that silent defalte is not supported. Right now, QEMU refuses migration if the target does not support all the features that were negotiated. But then: - a migration from non-MUST_TELL_HOST to MUST_TELL_HOST will s...

VIRTIO_BALLOON_F_MUST_TELL_HOST cannot be used properly because it is a "negative" feature: it tells you that silent defalte is not supported. Right now, QEMU refuses migration if the target does not support all the features that were negotiated. But then: - a migration from non-MUST_TELL_HOST to MUST_TELL_HOST will s...

Drivers treat MUST_TELL_HOST as optional: windows drivers do not ack it and expect this means they can tell host *after* deflate. This was not the intent but the documentation was not very clear on this point. Luckily hyprevisors did not implement this feature yet so to provide guidance for future devices make spec match dri...

Drivers treat MUST_TELL_HOST as optional: windows drivers do not ack it and expect this means they can tell host *after* deflate. This was not the intent but the documentation was not very clear on this point. Luckily hyprevisors did not implement this feature yet so to provide guidance for future devices make spec match dri...

...:10AM +0200, Paolo Bonzini wrote: > Il 28/05/2013 20:15, Michael S. Tsirkin ha scritto: > > What you write in spec below and what you write above seems to > > contradict. > > > > Look, how about the simpler patch that I sent: > > "[PATCH] virtio-spec: balloon: MUST_TELL_HOST is optional" > > > > does it functionally do everything you want in this patch? > > No, though it is a step forward. > > But let's take a step back instead; here are the requirements: > > 1) old drivers work unmodified on new hosts > > 2) new driver...

On Tue, May 28, 2013 at 07:40:17PM +0200, Paolo Bonzini wrote: > The idea of the VIRTIO_BALLOON_F_MUST_TELL_HOST feature was to let drivers > skip usage of the deflate queue when leaking the balloon ("silent > deflation"). Guests may benefit from silent deflate by aggressively > inflating the balloon; they know that they will be able to use ballooned > pages without issuing a (blocking)...

Il 28/05/2013 20:15, Michael S. Tsirkin ha scritto: > What you write in spec below and what you write above seems to > contradict. > > Look, how about the simpler patch that I sent: > "[PATCH] virtio-spec: balloon: MUST_TELL_HOST is optional" > > does it functionally do everything you want in this patch? No, though it is a step forward. But let's take a step back instead; here are the requirements: 1) old drivers work unmodified on new hosts 2) new drivers work unmodified on old guests, though it is okay...

...can crash it. So we need to know what the guest is going to do, in order to enable/disable the optimization. > If guest cooperates we don't have to require anything, > just go with what guest tells us it will do. Yes. >>> Please see >>> [PATCH] virtio-spec: balloon: MUST_TELL_HOST is optional >>> that does exactly this. >> >> That patch mandates a change in guest behavior that is not compatible >> with the existing Windows driver. Mine doesn't. >> >> Paolo > > Hmm I don't see it. > In fact the goal was to document the...

The idea of the VIRTIO_BALLOON_F_MUST_TELL_HOST feature is to let drivers skip usage of the deflate queue when leaking the balloon ("silent deflation"). Guests may benefit from silent deflate by aggressively inflating the balloon; they know that they will be able to use ballooned pages without issuing a (blocking) request to the devic...

The idea of the VIRTIO_BALLOON_F_MUST_TELL_HOST feature is to let drivers skip usage of the deflate queue when leaking the balloon ("silent deflation"). Guests may benefit from silent deflate by aggressively inflating the balloon; they know that they will be able to use ballooned pages without issuing a (blocking) request to the devic...

...3 at 04:06:02PM +0200, Paolo Bonzini wrote: > >>>> Il 28/05/2013 15:32, Michael S. Tsirkin ha scritto: > >>>>> At this point I am confused. I think there are two changes in your patch: > >>>>> > >>>>> 1. Handling of VIRTIO_F_GUEST_MUST_TELL_HOST > >>>>> Is this functionally identical to what I proposed? > >>>>> If yes, I am fine with either change being applied. > >>>> > >>>> Yes. > >>>> > >>>>> 2. New SILENT_DEFLATE feature > >>&g...

...On Tue, May 28, 2013 at 04:06:02PM +0200, Paolo Bonzini wrote: >>>> Il 28/05/2013 15:32, Michael S. Tsirkin ha scritto: >>>>> At this point I am confused. I think there are two changes in your patch: >>>>> >>>>> 1. Handling of VIRTIO_F_GUEST_MUST_TELL_HOST >>>>> Is this functionally identical to what I proposed? >>>>> If yes, I am fine with either change being applied. >>>> >>>> Yes. >>>> >>>>> 2. New SILENT_DEFLATE feature >>>>> Since guest can get sam...

...G_MAX); > + > /* Now we reset the device so we can clean up the queues. */ > vb->vdev->config->reset(vb->vdev); Right, thanks! > > > Issue 2: When called via the shrinker, (but also to fix Issue 1), it could be that > we do have VIRTIO_BALLOON_F_MUST_TELL_HOST. I don't think it is an issue here. MUST_TELL_HOST is for the ballooning pages, where pages are offered to host to _USE_. For free page hint, as the name already suggests, it's just a _HINT_ , so in whatever use case, the host should not take the page to use. So the guest doesn't need...

...ose the feature, > > Ack feature but don't tell host? That would be a clear guest bug. > AFAIK that's not what windows drivers do. > Am I wrong? Yes. I think we are in agreement on this part. >>>>> Please see >>>>> [PATCH] virtio-spec: balloon: MUST_TELL_HOST is optional >>>>> that does exactly this. >>>> >>>> That patch mandates a change in guest behavior that is not compatible >>>> with the existing Windows driver. Mine doesn't. >>> >>> Hmm I don't see it. >>> In fa...

...est is going to do, in order to > enable/disable the optimization. > > > If guest cooperates we don't have to require anything, > > just go with what guest tells us it will do. > > Yes. > > >>> Please see > >>> [PATCH] virtio-spec: balloon: MUST_TELL_HOST is optional > >>> that does exactly this. > >> > >> That patch mandates a change in guest behavior that is not compatible > >> with the existing Windows driver. Mine doesn't. > >> > >> Paolo > > > > Hmm I don't see it. &...

...>>>> (everything on free_page_list). >>>> >>>> So instead of only reverting, I think we should split it up and always >>>> register the shrinker for VIRTIO_BALLOON_F_FREE_PAGE_HINT and the OOM >>>> notifier (as before) for VIRTIO_BALLOON_F_MUST_TELL_HOST. >> >> s/VIRTIO_BALLOON_F_MUST_TELL_HOST/VIRTIO_BALLOON_F_DEFLATE_ON_OOM/ >> >> :) > > Well VIRTIO_BALLOON_F_MUST_TELL_HOST is also broken by shrinker > with VIRTIO_BALLOON_F_FREE_PAGE_HINT as that code adds buffers > but does not wait for them to be used even w...

...>>>> (everything on free_page_list). >>>> >>>> So instead of only reverting, I think we should split it up and always >>>> register the shrinker for VIRTIO_BALLOON_F_FREE_PAGE_HINT and the OOM >>>> notifier (as before) for VIRTIO_BALLOON_F_MUST_TELL_HOST. >> >> s/VIRTIO_BALLOON_F_MUST_TELL_HOST/VIRTIO_BALLOON_F_DEFLATE_ON_OOM/ >> >> :) > > Well VIRTIO_BALLOON_F_MUST_TELL_HOST is also broken by shrinker > with VIRTIO_BALLOON_F_FREE_PAGE_HINT as that code adds buffers > but does not wait for them to be used even w...

Here is the series, split in two patches, with small edits and new commit messages. Paolo Bonzini (2): virtio-balloon spec: rewrite description of VIRTIO_BALLOON_F_MUST_TELL_HOST virtio-balloon spec: reintroduce "silent deflation" feature virtio-spec.lyx | 238 ++++++++++++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 233 insertions(+), 5 deletions(-) -- 1.8.2.1

The idea of the VIRTIO_BALLOON_F_MUST_TELL_HOST feature was to let drivers skip usage of the deflate queue when leaking the balloon ("silent deflation"). Guests may benefit from silent deflate by aggressively inflating the balloon; they know that they will be able to use ballooned pages without issuing a (blocking) request to the devi...

...pose POISON with HINT. Then this configuration (HINT && !POISON) allows us to detect old QEMU (pre your changes). I am also interested in the following question: Is this good for anything? In particular are there any bugs in HINT we can fix using this hack? E.g. I know HINT does not obey MUST_TELL_HOST under OOM. But that unfortunately is a guest not a host bug, so this hack does not seem useful. > > We don't know what hypervisor uses the hints for. > > I agree, but at the same time the way the feature was originally coded > it was only checked if the FREE_PAGE_HINT feature...