[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <c1a48a1a470e10acefb52ec13ab799514c1cd31d.camel@mellanox.com>
Date: Tue, 29 Oct 2019 23:09:03 +0000
From: Saeed Mahameed <saeedm@...lanox.com>
To: Tariq Toukan <tariqt@...lanox.com>,
"jakub.kicinski@...ronome.com" <jakub.kicinski@...ronome.com>
CC: "davem@...emloft.net" <davem@...emloft.net>,
"netdev@...r.kernel.org" <netdev@...r.kernel.org>
Subject: Re: [net 11/11] Documentation: TLS: Add missing counter description
On Tue, 2019-10-29 at 14:48 +0000, Tariq Toukan wrote:
>
> On 10/25/2019 12:36 AM, Jakub Kicinski wrote:
> > On Thu, 24 Oct 2019 19:39:02 +0000, Saeed Mahameed wrote:
> > > From: Tariq Toukan <tariqt@...lanox.com>
> > >
> > > Add TLS TX counter description for the packets that skip the
> > > resync
> > > procedure and handled in the regular transmit flow, as they
> > > contain
> > > no data.
> > >
> > > Fixes: 46a3ea98074e ("net/mlx5e: kTLS, Enhance TX resync flow")
> > > Signed-off-by: Tariq Toukan <tariqt@...lanox.com>
> > > Signed-off-by: Saeed Mahameed <saeedm@...lanox.com>
> > > ---
> > > Documentation/networking/tls-offload.rst | 3 +++
> > > 1 file changed, 3 insertions(+)
> > >
> > > diff --git a/Documentation/networking/tls-offload.rst
> > > b/Documentation/networking/tls-offload.rst
> > > index 0dd3f748239f..87db87099607 100644
> > > --- a/Documentation/networking/tls-offload.rst
> > > +++ b/Documentation/networking/tls-offload.rst
> > > @@ -436,6 +436,9 @@ by the driver:
> > > encryption.
> > > * ``tx_tls_ooo`` - number of TX packets which were part of a
> > > TLS stream
> > > but did not arrive in the expected order.
> > > + * ``tx_tls_skip_no_sync_data`` - number of TX packets which
> > > were part of
> > > + a TLS stream and arrived out-of-order, but skipped the HW
> > > offload routine
> > > + and went to the regular transmit flow as they contained no
> > > data.
> >
> > That doesn't sound right. It sounds like you're talking about pure
> > Acks
> > and other segments with no data. For mlx5 those are skipped
> > directly in
> > mlx5e_ktls_handle_tx_skb().
> >
> > This counter is for segments which are retransmission of data
> > queued to
> > the socket before kernel got the keys installed.
> >
> > You explained it reasonably well in 46a3ea98074e ("net/mlx5e: kTLS,
> > Enhance TX resync flow"):
> >
> > 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.
> >
>
> Right. I'll rephrase.
>
Thanks Tariq, I will drop this patch from my series, please send it to
net branch as stand alone when ready.
Thanks,
Saeed.
Powered by blists - more mailing lists