Discussion of the implementations of VIRTIO specification
 help / color / mirror / Atom feed
From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
To: Gerd Hoffmann <kraxel@redhat.com>
Cc: Frank Yang <lfy@google.com>,
	"Michael S. Tsirkin" <mst@redhat.com>,
	Roman Kiryanov <rkir@google.com>,
	Stefan Hajnoczi <stefanha@redhat.com>,
	virtio-dev@lists.oasis-open.org
Subject: Re: [virtio-dev] Memory sharing device
Date: Thu, 21 Feb 2019 09:24:55 +0000	[thread overview]
Message-ID: <20190221092455.GB2605@work-vm> (raw)
In-Reply-To: <20190221072950.tcii5np5dsk4wvyd@sirius.home.kraxel.org>

* Gerd Hoffmann (kraxel@redhat.com) wrote:
>   Hi,
> 
> > > For communication between guest processes within the same VM I don't
> > > really see a need to involve the hypervisor ...
> > >
> > Right, once the host memory is set up we can rely on purely guest side stuff
> > map sub-regions of it.
> 
> Or just use guest ram ...
> 
> > > > Yes, also, other devices of the same VM.
> > >
> > > So why involve the hypervisor here?  The guest can handle that on its
> > > own.  Passing an image data buffer from the usb webcam to the intel gpu
> > > for display (on bare metal) isn't fundamentally different from passing a
> > > buffer from virtio-camera to virtio-gpu (in a VM).  Linux guests will
> > > use dma-bufs for that, other OSes probably something else.
> > 
> > That's true that it can be handled purely in the guest layers,
> > if there is an existing interface in the guest
> > to pass the proposed host memory id's / offsets / sizes
> > between them.
> 
> Note:  I think using a pci memory bar (aka host memory mapped into the
> guest) as backing storage for dma-bufs isn't going to work.

(Not knowing dma-bufs) but could you explain why?
Note in my spec the pci-bar isn't necessarily one chunk of host memory;
it's a chunk of host VMA into which multiple mmaps go.

Dave

> > However, for the proposed host memory sharing spec,
> > would there be a standard way to share the host memory across
> > different virtio devices without relying on Linux dmabufs?
> 
> I think with the current draft for each device (virtio-fs, virtio-gpu,
> ...) has its own device-specific memory, and there is no mechanism to
> exchange buffers between devices.
> 
> Stefan?
> 
> I'm also not convinced that explicitly avoiding dmabufs is a good idea
> here.  That would put virtio into its own universe and sharing buffers
> with non-virtio devices will not work.  Think about a intel vgpu as
> display device, or a usb camera attached to the guest using usb
> pass-through.
> 
> Experience shows that using virtualization-specific features /
> optimizations / short-cuts often turns out to have drawbacks in the long
> run, even if it looked like a good idea initially.  Just look at the
> mess we had with virtio-pci dma after iommu emulation landed in qemu.
> And this is only one example, we have more of this ...
> 
> cheers,
>   Gerd
> 
--
Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK

---------------------------------------------------------------------
To unsubscribe, e-mail: virtio-dev-unsubscribe@lists.oasis-open.org
For additional commands, e-mail: virtio-dev-help@lists.oasis-open.org


  reply	other threads:[~2019-02-21  9:25 UTC|newest]

Thread overview: 72+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-02-01 20:34 [virtio-dev] Memory sharing device Roman Kiryanov
2019-02-04  5:40 ` Stefan Hajnoczi
2019-02-04 10:13   ` Gerd Hoffmann
2019-02-04 10:18     ` Roman Kiryanov
2019-02-05  7:42     ` Roman Kiryanov
2019-02-05 10:04       ` Dr. David Alan Gilbert
2019-02-05 15:17         ` Frank Yang
2019-02-05 15:21           ` Frank Yang
2019-02-05 21:06         ` Roman Kiryanov
2019-02-06  7:03           ` Gerd Hoffmann
2019-02-06 15:09             ` Frank Yang
2019-02-06 15:11               ` Frank Yang
2019-02-08  7:57               ` Stefan Hajnoczi
2019-02-08 14:46                 ` Frank Yang
2019-02-06 20:14           ` Dr. David Alan Gilbert
2019-02-06 20:27             ` Frank Yang
2019-02-07 12:10               ` Cornelia Huck
2019-02-11 14:49       ` Michael S. Tsirkin
2019-02-11 15:14         ` Frank Yang
2019-02-11 15:25           ` Frank Yang
2019-02-12 13:01             ` Michael S. Tsirkin
2019-02-12 13:16             ` Dr. David Alan Gilbert
2019-02-12 13:27               ` Michael S. Tsirkin
2019-02-12 16:17                 ` Frank Yang
2019-02-19  7:17                   ` Gerd Hoffmann
2019-02-19 15:59                     ` Frank Yang
2019-02-20  6:51                       ` Gerd Hoffmann
2019-02-20 15:31                         ` Frank Yang
2019-02-21  6:55                           ` Gerd Hoffmann
2019-02-19  7:12             ` Gerd Hoffmann
2019-02-19 16:02               ` Frank Yang
2019-02-20  7:02                 ` Gerd Hoffmann
2019-02-20 15:32                   ` Frank Yang
2019-02-21  7:29                     ` Gerd Hoffmann
2019-02-21  9:24                       ` Dr. David Alan Gilbert [this message]
2019-02-21  9:59                         ` Gerd Hoffmann
2019-02-21 10:03                           ` Dr. David Alan Gilbert
2019-02-22  6:15                           ` Michael S. Tsirkin
2019-02-22  6:42                             ` Gerd Hoffmann
2019-02-11 16:57           ` Michael S. Tsirkin
2019-02-12  8:27         ` Roman Kiryanov
2019-02-12 11:25           ` Dr. David Alan Gilbert
2019-02-12 13:47             ` Cornelia Huck
2019-02-12 14:03               ` Michael S. Tsirkin
2019-02-12 15:56                 ` Frank Yang
2019-02-12 16:46                   ` Dr. David Alan Gilbert
2019-02-12 17:20                     ` Frank Yang
2019-02-12 17:26                       ` Frank Yang
2019-02-12 19:06                         ` Michael S. Tsirkin
2019-02-13  2:50                           ` Frank Yang
2019-02-13  4:02                             ` Michael S. Tsirkin
2019-02-13  4:19                               ` Michael S. Tsirkin
2019-02-13  4:59                                 ` Frank Yang
2019-02-13 18:18                                   ` Frank Yang
2019-02-14  7:15                                     ` Frank Yang
2019-02-22 22:05                                       ` Michael S. Tsirkin
2019-02-24 21:19                                         ` Frank Yang
2019-02-13  4:59                               ` Frank Yang
2019-02-19  7:54                       ` Gerd Hoffmann
2019-02-19 15:54                         ` Frank Yang
2019-02-20  3:46                           ` Michael S. Tsirkin
2019-02-20 15:24                             ` Frank Yang
2019-02-20 19:29                               ` Michael S. Tsirkin
2019-02-20  6:25                           ` Gerd Hoffmann
2019-02-20 15:30                             ` Frank Yang
2019-02-20 15:35                               ` Frank Yang
2019-02-21  6:44                               ` Gerd Hoffmann
2019-02-12 18:22                   ` Michael S. Tsirkin
2019-02-12 19:01                     ` Frank Yang
2019-02-12 19:15                       ` Michael S. Tsirkin
2019-02-12 20:15                         ` Frank Yang
2019-02-12 13:00           ` 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=20190221092455.GB2605@work-vm \
    --to=dgilbert@redhat.com \
    --cc=kraxel@redhat.com \
    --cc=lfy@google.com \
    --cc=mst@redhat.com \
    --cc=rkir@google.com \
    --cc=stefanha@redhat.com \
    --cc=virtio-dev@lists.oasis-open.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