[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAJ3xEMhMpDqWrn8h_3jmMQaEz_Wt6OAVkh0BNkM6-Re6avhWXQ@mail.gmail.com>
Date: Wed, 23 Oct 2019 18:48:09 +0300
From: Or Gerlitz <gerlitz.or@...il.com>
To: Yuval Avnery <yuvalav@...lanox.com>
Cc: Linux Netdev List <netdev@...r.kernel.org>,
Jiri Pirko <jiri@...lanox.com>,
Saeed Mahameed <saeedm@...lanox.com>,
Leon Romanovsky <leon@...nel.org>,
David Miller <davem@...emloft.net>,
Jakub Kicinski <jakub.kicinski@...ronome.com>, shuah@...nel.org
Subject: Re: [PATCH net-next 0/9] devlink vdev
On Tue, Oct 22, 2019 at 8:46 PM Yuval Avnery <yuvalav@...lanox.com> wrote:
> This patchset introduces devlink vdev.
> Currently, legacy tools do not provide a comprehensive solution that can
> be used in both SmartNic and non-SmartNic mode.
> Vdev represents a device that exists on the ASIC but is not necessarily
> visible to the kernel.
I assume you refer to a function (e.g Eth VF, NVME) or mediated device (e.g the
sub-functions work by mellanox we presented in netdev 0x13) residing on the
host but not visible to the smart nic, smart... would be good to
clarify that here.
FWIW whatever SW you have on the smart NIC, it will still not be able
to actually provision / PT / map this device to a guest/container before/after
this patch set, and a host SW agent is still needed, agree?
worth explaining what delta/progress this series gets us in this context
(of inability to provision from the smartnic ).
Or.
Powered by blists - more mailing lists