If AVP was an upstream device of Qemu or Linux kernel that would be very natural to have a DPDK PMD (setting aside my comments on a preferred single virtual device).
As far as I know this is not the case. Because of that, one could see the AVP PMD as a way to leverage open source to promote proprietary technology. That is the heart of the problem with the proposal. So I would recommend waiting for an upstream qemu support to consider AVP in DPDK . FF ________________________________ From: dev <dev-boun...@dpdk.org> on behalf of Michael S. Tsirkin <m...@redhat.com> Sent: Friday, March 17, 2017 2:52:06 PM To: Thomas Monjalon Cc: Wiles, Keith; Jason Wang; Vincent JARDIN; Stephen Hemminger; O'Driscoll, Tim; Legacy, Allain (Wind River); Yigit, Ferruh; dev@dpdk.org; Jolliffe, Ian (Wind River); Markus Armbruster; Stefan Hajnoczi Subject: Re: [dpdk-dev] [PATCH v4 00/17] Wind River Systems AVP PMD vs virtio? - ivshmem is back On Fri, Mar 17, 2017 at 09:48:38AM +0100, Thomas Monjalon wrote: > I think there is one interesting technological point in this thread. > We are discussing about IVSHMEM but its support by Qemu is confused. > This feature is not in the MAINTAINERS file of Qemu. > Please Qemu maintainers, what is the future of IVSHMEM? You should try asking this question on the qemu mailing list. Looking at archives, Jan Kiszka was the last one who expressed some interest in this device. -- MST
