[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <d35e1bc263bdbe89016de92ef6488326d666c641.camel@mellanox.com>
Date: Tue, 22 Oct 2019 18:52:34 +0000
From: Saeed Mahameed <saeedm@...lanox.com>
To: "jakub.kicinski@...ronome.com" <jakub.kicinski@...ronome.com>
CC: "davem@...emloft.net" <davem@...emloft.net>,
"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
Tariq Toukan <tariqt@...lanox.com>
Subject: Re: [net 12/15] net/mlx5e: kTLS, Enhance TX resync flow
On Tue, 2019-10-22 at 11:42 -0700, Jakub Kicinski wrote:
> On Tue, 22 Oct 2019 18:10:11 +0000, Saeed Mahameed wrote:
> > On Fri, 2019-10-18 at 18:51 -0700, Jakub Kicinski wrote:
> > > On Fri, 18 Oct 2019 19:38:24 +0000, Saeed Mahameed wrote:
> > > > From: Tariq Toukan <tariqt@...lanox.com>
> > > >
> > > > Once the kTLS TX resync function is called, it used to return
> > > > a binary value, for success or failure.
> > > >
> > > > However, in case the TLS SKB is a retransmission of the
> > > > connection
> > > > handshake, it initiates the resync flow (as the tcp seq check
> > > > holds),
> > > > while regular packet handle is expected.
> > > >
> > > > In this patch, we identify this case and skip the resync
> > > > operation
> > > > accordingly.
> > > >
> > > > Counters:
> > > > - Add a counter (tls_skip_no_sync_data) to monitor this.
> > > > - Bump the dump counters up as they are used more frequently.
> > > > - Add a missing counter descriptor declaration for
> > > > tls_resync_bytes
> > > > in sq_stats_desc.
> > > >
> > > > Fixes: d2ead1f360e8 ("net/mlx5e: Add kTLS TX HW offload
> > > > support")
> > > > Signed-off-by: Tariq Toukan <tariqt@...lanox.com>
> > > > Signed-off-by: Saeed Mahameed <saeedm@...lanox.com>
> > >
> > > Could you document the new counter in tls-offload.rst?
> >
> > Tariq already prepared the patch, do we want this in net or net-
> > next ?
> > most of our kTLS users are going to use 5.3/5.4 stable kernels, but
> > i
> > am not sure this should go to net..
>
> Thanks, I asked Dave recently about docs and he said we can apply to
> net
> (since doc changes can't really break any code). Since the code is in
> net, I think net would be appropriate for the doc as well.
Alright! will send it in my next batch of fixes to net.
Thanks,
Saeed.
Powered by blists - more mailing lists