OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

virtio-comment message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: Re: [virtio-comment] [PATCH] virtio-net: Add support for the flexible driver notification

>-----Original Message-----
>From: Michael S. Tsirkin <mst@redhat.com>
>Sent: Monday, 3 February, 2020 11:52
>To: Vitaly Mireyno <vmireyno@marvell.com>
>Cc: Stefan Hajnoczi <stefanha@redhat.com>; virtio-comment@lists.oasis-
>open.org; Jason Wang <jasowang@redhat.com>; Ariel Elior
>Subject: [EXT] Re: [virtio-comment] [PATCH] virtio-net: Add support for the
>flexible driver notification
>On Thu, Jan 30, 2020 at 04:56:15PM +0000, Vitaly Mireyno wrote:
>> The main motivation is to support virtio on the existing silicon,
>> where this is the only way. Hopefully more use-cases would arise that
>> would find this useful.
>So there's some silicon out there that just happened to use exactly the virtio
>format, except for a couple of bits in the notification? Interesting.

Yes, exactly. The virtio notifications format is quite generic, so it's probably not that surprising that it's almost suitable for some HW devices.

>And what are these bits actually set to in this hardware? Just curious ...

At the low level, the notification data contains an encoded information of where to store the producer, and where to send the notification to. In case of the virtio, it will enqueue the request for the networking fast-path processing, and update the new ring producer in the internal location, specific to the L2/virtio. In other cases, it may just update the data in the internal memory (at a protocol specific location), without triggering any further action.


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]