[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH v1 0/6] Extend vhost-user to support VFIO based
From: |
Liang, Cunming |
Subject: |
Re: [Qemu-devel] [PATCH v1 0/6] Extend vhost-user to support VFIO based accelerators |
Date: |
Thu, 25 Jan 2018 16:10:56 +0000 |
> -----Original Message-----
> From: Stefan Hajnoczi [mailto:address@hidden
> Sent: Thursday, January 25, 2018 10:22 PM
> To: Bie, Tiwei <address@hidden>
> Cc: address@hidden; address@hidden;
> address@hidden; address@hidden; address@hidden;
> address@hidden; address@hidden; Tan, Jianfeng
> <address@hidden>; Liang, Cunming <address@hidden>;
> Wang, Xiao W <address@hidden>; Wang, Zhihong
> <address@hidden>; Daly, Dan <address@hidden>
> Subject: Re: [Qemu-devel] [PATCH v1 0/6] Extend vhost-user to support
> VFIO based accelerators
>
> On Thu, Jan 25, 2018 at 12:03:22PM +0800, Tiwei Bie wrote:
> > Why extend vhost-user for vDPA
> > ==============================
> >
> > We have already implemented various virtual switches (e.g. OVS-DPDK)
> > based on vhost-user for VMs in the Cloud. They are purely software
> > running on CPU cores. When we have accelerators for such NFVi
> > applications, it's ideal if the applications could keep using the
> > original interface (i.e. vhost-user netdev) with QEMU, and
> > infrastructure is able to decide when and how to switch between CPU
> > and
> accelerators within the interface.
> > And the switching (i.e. switch between CPU and accelerators) can be
> > done flexibly and quickly inside the applications.
> >
> > More details about this can be found from the Cunming's discussions
> > on the RFC patch set.
> >
> > The previous links:
> > RFC:
> > http://lists.nongnu.org/archive/html/qemu-devel/2017-12/msg04844.htm
> > l
>
> Is vDPA also useful in the simpler use case where there is no NFVi
> application?
There're separate patches to have non-DPDK vDPA(aka. vhost-vfio, a new vhost
backend proposed) support, which is the scope of stage-II as we mentioned in
community call.
Stay tuned for the RFC patch. Main idea is that the virtio compatible device
driver register as a mdev to talk with qemu vhost-vfio.
>
> In other words, you have virtio-net inside the guest and vhost-net on the
> host.
> You'd like to accelerate the virtio-net NIC using the hardware's vDPA support.
>
> DPDK requires dedicated logical cores and hugepages. Those resources
> will not be used (wasted) if you just want to enable vDPA but have no
> DPDK packet processing application.
>
> How can this use case be supported without wasting resources?
>
> Stefan
- [Qemu-devel] [PATCH v1 3/6] virtio: support adding sub-regions for notify region, (continued)
- [Qemu-devel] [PATCH v1 3/6] virtio: support adding sub-regions for notify region, Tiwei Bie, 2018/01/24
- [Qemu-devel] [PATCH v1 1/6] vhost-user: support receiving file descriptors in slave_read, Tiwei Bie, 2018/01/24
- [Qemu-devel] [PATCH v1 2/6] vhost-user: introduce shared vhost-user state, Tiwei Bie, 2018/01/24
- [Qemu-devel] [PATCH v1 4/6] vfio: support getting VFIOGroup from groupfd, Tiwei Bie, 2018/01/24
- [Qemu-devel] [PATCH v1 5/6] vfio: remove DPRINTF() definition from vfio-common.h, Tiwei Bie, 2018/01/24
- [Qemu-devel] [PATCH v1 6/6] vhost-user: add VFIO based accelerators support, Tiwei Bie, 2018/01/24
- Re: [Qemu-devel] [PATCH v1 0/6] Extend vhost-user to support VFIO based accelerators, Stefan Hajnoczi, 2018/01/25
- Re: [Qemu-devel] [PATCH v1 0/6] Extend vhost-user to support VFIO based accelerators,
Liang, Cunming <=