From: Jason Gunthorpe <jgg@ziepe.ca>
To: Jason Wang <jasowang@redhat.com>
Cc: Alex Williamson <alex.williamson@redhat.com>,
"Michael S. Tsirkin" <mst@redhat.com>,
Parav Pandit <parav@mellanox.com>,
Jeff Kirsher <jeffrey.t.kirsher@intel.com>,
davem@davemloft.net, gregkh@linuxfoundation.org,
Dave Ertman <david.m.ertman@intel.com>,
netdev@vger.kernel.org, linux-rdma@vger.kernel.org,
nhorman@redhat.com, sassmann@redhat.com,
Kiran Patil <kiran.patil@intel.com>,
Tiwei Bie <tiwei.bie@intel.com>
Subject: Re: [net-next v2 1/1] virtual-bus: Implementation of Virtual Bus
Date: Fri, 22 Nov 2019 14:02:14 -0400 [thread overview]
Message-ID: <20191122180214.GD7448@ziepe.ca> (raw)
In-Reply-To: <721e49c2-a2e1-853f-298b-9601c32fcf9e@redhat.com>
On Fri, Nov 22, 2019 at 04:45:38PM +0800, Jason Wang wrote:
>
> On 2019/11/21 下午10:17, Jason Gunthorpe wrote:
> > On Thu, Nov 21, 2019 at 03:21:29PM +0800, Jason Wang wrote:
> > > > The role of vfio has traditionally been around secure device
> > > > assignment of a HW resource to a VM. I'm not totally clear on what the
> > > > role if mdev is seen to be, but all the mdev drivers in the tree seem
> > > > to make 'and pass it to KVM' a big part of their description.
> > > >
> > > > So, looking at the virtio patches, I see some intended use is to map
> > > > some BAR pages into the VM.
> > > Nope, at least not for the current stage. It still depends on the
> > > virtio-net-pci emulatio in qemu to work. In the future, we will allow such
> > > mapping only for dorbell.
> > There has been a lot of emails today, but I think this is the main
> > point I want to respond to.
> >
> > Using vfio when you don't even assign any part of the device BAR to
> > the VM is, frankly, a gigantic misuse, IMHO.
>
> That's not a compelling point.
Well, this discussion is going nowhere.
> > Just needing userspace DMA is not, in any way, a justification to use
> > vfio.
> >
> > We have extensive library interfaces in the kernel to do userspace DMA
> > and subsystems like GPU and RDMA are full of example uses of this kind
> > of stuff.
>
> I'm not sure which library did you mean here. Is any of those library used
> by qemu? If not, what's the reason?
I mean the library functions in the kernel that vfio uses to implement
all the user dma stuff. Other subsystems use them too, it is not
exclusive to vfio.
> > Further, I do not think it is wise to design the userspace ABI around
> > a simplistict implementation that can't do BAR assignment,
>
> Again, the vhost-mdev follow the VFIO ABI, no new ABI is invented, and
> mmap() was kept their for mapping device regions.
The patches have a new file in include/uapi.
Everything in include/api is considered new user ABI.
> > My advice is to proceed as a proper subsystem with your own chardev,
> > own bus type, etc and maybe live in staging for a bit until 2-3
> > drivers are implementing the ABI (or at the very least agreeing with),
> > as is the typical process for Linux.
>
> I'm open to comments for sure, but looking at all the requirement for vDPA,
> most of the requirement could be settled through existed modules, that's not
> only a simplification for developing but also for management layer or
> userspace drivers.
We've already got disagreement that the GUID based mdev approach is
desirable for management.
Performing userspace DMA from a kernel driver is absolutely not a
reason to use VFIO.
Is there any technical justification?
Jason
next prev parent reply other threads:[~2019-11-22 18:02 UTC|newest]
Thread overview: 86+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-11-15 22:33 [net-next v2 1/1] virtual-bus: Implementation of Virtual Bus Jeff Kirsher
2019-11-15 23:25 ` Parav Pandit
2019-11-19 3:58 ` Ertman, David M
2019-11-19 4:31 ` Parav Pandit
2019-11-19 4:39 ` Parav Pandit
2019-11-19 17:46 ` Ertman, David M
2019-11-19 18:39 ` Jason Gunthorpe
2019-11-19 17:44 ` Ertman, David M
2019-11-19 4:08 ` Jason Wang
2019-11-19 4:36 ` Parav Pandit
2019-11-19 6:51 ` Jason Wang
2019-11-19 7:13 ` Parav Pandit
2019-11-19 7:37 ` Jason Wang
2019-11-19 15:14 ` Parav Pandit
2019-11-20 3:15 ` Jason Wang
2019-11-20 3:38 ` Parav Pandit
2019-11-20 4:07 ` Jason Wang
2019-11-20 13:41 ` Jason Gunthorpe
2019-11-21 4:06 ` Jason Wang
2019-11-20 8:52 ` Michael S. Tsirkin
2019-11-20 12:03 ` Jiri Pirko
2019-11-19 16:46 ` Jason Gunthorpe
2019-11-19 18:58 ` Michael S. Tsirkin
2019-11-19 19:03 ` Jason Gunthorpe
2019-11-19 21:34 ` Michael S. Tsirkin
2019-11-19 19:15 ` Jason Gunthorpe
2019-11-19 21:33 ` Michael S. Tsirkin
2019-11-19 23:10 ` Jason Gunthorpe
2019-11-20 0:16 ` Michael S. Tsirkin
2019-11-20 1:46 ` Jason Gunthorpe
2019-11-20 3:59 ` Jason Wang
2019-11-20 5:34 ` Jason Wang
2019-11-20 13:38 ` Jason Gunthorpe
2019-11-20 14:15 ` Michael S. Tsirkin
2019-11-20 17:28 ` Alex Williamson
2019-11-20 18:11 ` Jason Gunthorpe
2019-11-20 22:07 ` Alex Williamson
2019-11-20 22:39 ` Parav Pandit
2019-11-21 8:17 ` Jason Wang
2019-11-21 3:03 ` Jason Gunthorpe
2019-11-21 4:24 ` Michael S. Tsirkin
2019-11-21 13:44 ` Jason Gunthorpe
2019-11-23 16:50 ` Michael S. Tsirkin
2019-11-21 7:21 ` Jason Wang
2019-11-21 14:17 ` Jason Gunthorpe
2019-11-22 8:45 ` Jason Wang
2019-11-22 18:02 ` Jason Gunthorpe [this message]
2019-11-23 4:39 ` Tiwei Bie
2019-11-23 23:09 ` Jason Gunthorpe
2019-11-24 11:00 ` Michael S. Tsirkin
2019-11-24 14:56 ` Tiwei Bie
2019-11-25 0:07 ` Jason Gunthorpe
2019-11-24 14:51 ` Tiwei Bie
2019-11-24 15:07 ` Michael S. Tsirkin
2019-11-25 0:09 ` Jason Gunthorpe
2019-11-25 12:59 ` Jason Wang
2019-11-23 16:48 ` Michael S. Tsirkin
2019-11-21 5:22 ` Jason Wang
2019-11-21 6:59 ` Jason Wang
2019-11-21 3:52 ` Jason Wang
2019-11-20 7:38 ` Michael S. Tsirkin
2019-11-20 13:03 ` Jason Gunthorpe
2019-11-20 13:43 ` Michael S. Tsirkin
2019-11-20 14:30 ` Jason Gunthorpe
2019-11-20 14:57 ` Michael S. Tsirkin
2019-11-20 16:45 ` Jason Gunthorpe
2019-11-20 22:05 ` Michael S. Tsirkin
2019-11-21 1:38 ` Jason Gunthorpe
2019-11-21 4:53 ` Jason Wang
2019-11-20 3:29 ` Jason Wang
2019-11-20 3:24 ` Jason Wang
2019-11-20 13:33 ` Jason Gunthorpe
2019-11-21 3:57 ` Jason Wang
2019-11-21 15:10 ` Martin Habets
2019-11-22 9:13 ` Jason Wang
2019-11-22 16:19 ` Parav Pandit
2019-11-26 12:26 ` Martin Habets
2019-11-27 10:58 ` Jason Wang
2019-11-27 11:03 ` Jason Wang
2019-11-15 23:42 ` Parav Pandit
2019-11-18 7:48 ` Greg KH
2019-11-18 22:57 ` Ertman, David M
2019-11-19 8:04 ` Jason Wang
2019-11-19 17:50 ` Ertman, David M
2019-11-18 7:49 ` Greg KH
2019-11-18 22:55 ` Ertman, David M
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=20191122180214.GD7448@ziepe.ca \
--to=jgg@ziepe.ca \
--cc=alex.williamson@redhat.com \
--cc=davem@davemloft.net \
--cc=david.m.ertman@intel.com \
--cc=gregkh@linuxfoundation.org \
--cc=jasowang@redhat.com \
--cc=jeffrey.t.kirsher@intel.com \
--cc=kiran.patil@intel.com \
--cc=linux-rdma@vger.kernel.org \
--cc=mst@redhat.com \
--cc=netdev@vger.kernel.org \
--cc=nhorman@redhat.com \
--cc=parav@mellanox.com \
--cc=sassmann@redhat.com \
--cc=tiwei.bie@intel.com \
/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;
as well as URLs for NNTP newsgroup(s).