[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAJ3xEMjbKQSiC0Uqw2vQoaK2VfNkvKgYOo=-aACyq9a8gghm5g@mail.gmail.com>
Date: Wed, 30 Nov 2016 19:10:56 +0200
From: Or Gerlitz <gerlitz.or@...il.com>
To: John Fastabend <john.fastabend@...il.com>
Cc: Hadar Hen Zion <hadarh@...lanox.com>,
"David S. Miller" <davem@...emloft.net>,
Linux Netdev List <netdev@...r.kernel.org>,
Saeed Mahameed <saeedm@...lanox.com>,
Jiri Pirko <jiri@...lanox.com>,
Amir Vadai <amirva@...lanox.com>,
Or Gerlitz <ogerlitz@...lanox.com>,
Roi Dayan <roid@...lanox.com>
Subject: Re: [PATCH net-next 8/8] net/mlx5e: Support adding ingress tc rule
when egress device flag is set
On Wed, Nov 30, 2016 at 6:36 PM, John Fastabend
<john.fastabend@...il.com> wrote:
> I started to dig through these patches and the last series here,
>
> Re: [PATCH net-next 00/13] Mellanox 100G SRIOV offloads tunnel_key
> set/release
> Can you explain how these two are related? I'm guessing in that first
> series the actual redirect action to a tunnel device was being ignore?
> Does this series clean up that bit of software/hardware alignment.
If only the 1st series is applied, the kernel pieces are functional
but with user-space
manually putting the decap (tunnel key release) TC rule on the uplink device
(tc tool -> UAPI --> tc core --> flower --> mlx5), but under real life
use-case, the rule
is put on shared-tunnel device, so this series adds on the prev one.
Or.
Powered by blists - more mailing lists