On Tue, Jun 27, 2023 at 11:17 AM Parav Pandit wrote:
>
>
>
> > From: Jason Wang
> > Sent: Monday, June 26, 2023 10:38 PM
>
> > > >
> > > Because there is no point of failing it later when actual rw occurs.
> >
> > Even with admin virtqueue, the hypervisor should be ready for any admin
> > command
> From: Jason Wang
> Sent: Monday, June 26, 2023 10:38 PM
> > >
> > Because there is no point of failing it later when actual rw occurs.
>
> Even with admin virtqueue, the hypervisor should be ready for any admin
> command failures somehow.
>
Yes, device failures can happen regardless.
> And
On 6/26/2023 6:50 PM, Xuan Zhuo wrote:
On Mon, 26 Jun 2023 17:56:01 +0800, "Zhu, Lingshan"
wrote:
On 6/26/2023 5:16 PM, Xuan Zhuo wrote:
On Mon, 26 Jun 2023 16:59:48 +0800, "Zhu, Lingshan"
wrote:
On 6/26/2023 4:09 PM, Xuan Zhuo wrote:
On Mon, 26 Jun 2023 15:57:33 +0800, "Zhu, Lingshan
On Mon, Jun 26, 2023 at 12:04 PM Parav Pandit wrote:
>
>
>
> > From: virtio-comm...@lists.oasis-open.org > open.org> On Behalf Of Jason Wang
> > Sent: Monday, June 26, 2023 12:00 AM
>
> > > This way we just say "make legacy guests work" and
> > > this is the problem of the hardware vendor not
On Mon, Jun 26, 2023 at 11:52 AM Parav Pandit wrote:
>
>
>
> > From: Jason Wang
> > Sent: Sunday, June 25, 2023 11:32 PM
>
> > > Hypervisor flow without involving guest; first sanity round to figure out
> > > things
> > can work:
> >
> > Why is this sanity round a must?
> >
> Because there is no
> From: Michael S. Tsirkin
> Sent: Monday, June 26, 2023 8:35 AM
> > Yet to catch up on the thread, so likely I am missing something.
> >
> > The flow is for one OS (Linux) is:
> > 1. user enable SR-IOV on the PF device in a host, which creates SR-IOV VFs
> > in
> the device.
> > (num_vfs an
> From: Xuan Zhuo
> Sent: Monday, June 26, 2023 8:33 AM
[..]
> > The flow is for one OS (Linux) is:
> > 1. user enable SR-IOV on the PF device in a host, which creates SR-IOV VFs
> > in
> the device.
> > (num_vfs and vf enable bit in the PCI capability)
> >
> > 2. VFs are created at the PCI
On Mon, Jun 26, 2023 at 12:19:04PM +, Parav Pandit wrote:
>
>
> > From: Xuan Zhuo
> > Sent: Monday, June 26, 2023 6:51 AM
> >
> > On Mon, 26 Jun 2023 17:56:01 +0800, "Zhu, Lingshan"
> > wrote:
> > >
> > >
> > > On 6/26/2023 5:16 PM, Xuan Zhuo wrote:
> > > > On Mon, 26 Jun 2023 16:59:48 +08
> From: Xuan Zhuo
> Sent: Monday, June 26, 2023 6:51 AM
>
> On Mon, 26 Jun 2023 17:56:01 +0800, "Zhu, Lingshan"
> wrote:
> >
> >
> > On 6/26/2023 5:16 PM, Xuan Zhuo wrote:
> > > On Mon, 26 Jun 2023 16:59:48 +0800, "Zhu, Lingshan"
> wrote:
> > >>
> > >> On 6/26/2023 4:09 PM, Xuan Zhuo wrote:
On Mon, Jun 26, 2023 at 11:50:24AM +0800, Jason Wang wrote:
> > But hardware
> > vendors will do that quickly if they can't sell hardware.
>
> It looks to me that producing a modern only device will be more quick and
> easy?
>
> Thanks
Exactly, that's the point.
--
MST
-
On 6/26/2023 5:16 PM, Xuan Zhuo wrote:
On Mon, 26 Jun 2023 16:59:48 +0800, "Zhu, Lingshan"
wrote:
On 6/26/2023 4:09 PM, Xuan Zhuo wrote:
On Mon, 26 Jun 2023 15:57:33 +0800, "Zhu, Lingshan"
wrote:
On 6/26/2023 3:08 PM, Xuan Zhuo wrote:
On Mon, 26 Jun 2023 14:43:17 +0800, "Zhu, Lingshan
On Mon, Jun 26, 2023 at 02:22:10PM +0800, Xuan Zhuo wrote:
> The VFs of the SR-IOV are created by the user inside the guest OS, so the
> virtio
> devices don't know about these VFs. Because each VF may be assigned a
> different role
> by the user, the virtio device can not choose one VF to bind r
On 6/26/2023 4:09 PM, Xuan Zhuo wrote:
On Mon, 26 Jun 2023 15:57:33 +0800, "Zhu, Lingshan"
wrote:
On 6/26/2023 3:08 PM, Xuan Zhuo wrote:
On Mon, 26 Jun 2023 14:43:17 +0800, "Zhu, Lingshan"
wrote:
On 6/26/2023 2:22 PM, Xuan Zhuo wrote:
The VFs of the SR-IOV are created by the user insi
On 6/26/2023 3:08 PM, Xuan Zhuo wrote:
On Mon, 26 Jun 2023 14:43:17 +0800, "Zhu, Lingshan"
wrote:
On 6/26/2023 2:22 PM, Xuan Zhuo wrote:
The VFs of the SR-IOV are created by the user inside the guest OS, so the virtio
devices don't know about these VFs. Because each VF may be assigned a d
On Mon, Jun 26, 2023 at 11:59:42AM +0800, Jason Wang wrote:
> > - when some weird legacy
> > support requirement surfaces, it will be up to the vendor
> > to fix. HW vendors are also more agressive in deprecating
> > old hardware - they will just stop shipping new
> > hardware when there ar
15 matches
Mail list logo