From: Jakub Kicinski <jakub.kicinski@netronome.com>
To: Tariq Toukan <tariqt@mellanox.com>
Cc: Saeed Mahameed <saeedm@mellanox.com>,
"David S. Miller" <davem@davemloft.net>,
"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
Eran Ben Elisha <eranbe@mellanox.com>,
Boris Pismenny <borisp@mellanox.com>
Subject: Re: [net-next 14/14] net/mlx5e: Add kTLS TX HW offload support
Date: Fri, 5 Jul 2019 12:43:51 -0700 [thread overview]
Message-ID: <20190705124351.4abfc7f3@cakuba.netronome.com> (raw)
In-Reply-To: <079d4170-d591-18c6-572e-dbec428f169e@mellanox.com>
On Fri, 5 Jul 2019 14:31:29 +0000, Tariq Toukan wrote:
> On 7/4/2019 11:12 PM, Jakub Kicinski wrote:
> > On Thu, 4 Jul 2019 18:16:15 +0000, Saeed Mahameed wrote:
> >> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
> >> index 483d321d2151..6854f132d505 100644
> >> --- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
> >> +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
> >> @@ -50,6 +50,15 @@ static const struct counter_desc sw_stats_desc[] = {
> >> #ifdef CONFIG_MLX5_EN_TLS
> >> { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_tls_ooo) },
> >> { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_tls_resync_bytes) },
> >> +
> >> + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_ktls_ooo) },
> >
> > Why do you call this stat tx_ktls_ooo, and not tx_tls_ooo (extra 'k')?
> >
> > For nfp I used the stats' names from mlx5 FPGA to make sure we are all
> > consistent. I've added them to the tls-offload.rst doc and Boris has
> > reviewed it.
> >
> > * ``rx_tls_decrypted`` - number of successfully decrypted TLS segments
> > * ``tx_tls_encrypted`` - number of in-order TLS segments passed to device
> > for 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_drop_no_sync_data`` - number of TX packets dropped because
> > they arrived out of order and associated record could not be found
> >
> > Why can't you use the same names for the stats as you used for your mlx5
> > FPGA?
> >
>
> Agree. Fixing.
>
> What about having stats both for packets and bytes?
> tx_tls_encrypted_packets
> tx_tls_encrypted_bytes
Makes sense, I wasn't sure we want too many counters on the fastpath,
therefore I had no bytes counter. Renaming rx_tls_decrypted and
tx_tls_encrypted sounds like a good idea, though! We only have them in
the nfp, and there wasn't any kernel released with nfp TLS offload,
yet, so we can adjust.
Perhaps we should also make it clear in the doc that those counters
count "packets"/bytes on the wire, while the other counters count the
number of skbs, which may be TSO?
Would you be able to make those adjustments if you agree?
next prev parent reply other threads:[~2019-07-05 19:43 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-07-04 18:15 [pull request][net-next 00/14] Mellanox, mlx5 updates 2019-07-04 Saeed Mahameed
2019-07-04 18:15 ` [net-next 01/14] net/mlx5: Added fw version query command Saeed Mahameed
2019-07-04 18:15 ` [net-next 02/14] net/mlx5: Added devlink info callback Saeed Mahameed
2019-07-04 18:15 ` [net-next 03/14] net/mlx5: Accel, Expose accel wrapper for IPsec FPGA function Saeed Mahameed
2019-07-04 18:15 ` [net-next 04/14] net/mlx5: Kconfig, Better organize compilation flags Saeed Mahameed
2019-07-04 18:15 ` [net-next 05/14] net/mlx5: Add crypto library to support create/destroy encryption key Saeed Mahameed
2019-07-04 20:50 ` Jakub Kicinski
2019-07-04 18:15 ` [net-next 06/14] net/mlx5: Accel, Add core TLS support for the Connect-X family Saeed Mahameed
2019-07-04 18:15 ` [net-next 07/14] net/mlx5e: Move helper functions to a new txrx datapath header Saeed Mahameed
2019-07-04 18:16 ` [net-next 08/14] net/mlx5e: Tx, Enforce L4 inline copy when needed Saeed Mahameed
2019-07-04 18:16 ` [net-next 09/14] net/mlx5e: Tx, Make SQ WQE fetch function type generic Saeed Mahameed
2019-07-04 18:16 ` [net-next 10/14] net/mlx5e: Tx, Don't implicitly assume SKB-less wqe has one WQEBB Saeed Mahameed
2019-07-04 18:16 ` [net-next 11/14] net/mlx5e: Tx, Unconstify SQ stop room Saeed Mahameed
2019-07-04 18:16 ` [net-next 12/14] net/mlx5e: Re-work TIS creation functions Saeed Mahameed
2019-07-04 18:16 ` [net-next 13/14] net/mlx5e: Introduce a fenced NOP WQE posting function Saeed Mahameed
2019-07-04 18:16 ` [net-next 14/14] net/mlx5e: Add kTLS TX HW offload support Saeed Mahameed
2019-07-04 20:12 ` Jakub Kicinski
2019-07-04 20:30 ` Saeed Mahameed
2019-07-04 20:45 ` Jakub Kicinski
2019-07-05 14:31 ` Tariq Toukan
2019-07-05 19:43 ` Jakub Kicinski [this message]
2019-07-04 21:35 ` Jakub Kicinski
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190705124351.4abfc7f3@cakuba.netronome.com \
--to=jakub.kicinski@netronome.com \
--cc=borisp@mellanox.com \
--cc=davem@davemloft.net \
--cc=eranbe@mellanox.com \
--cc=netdev@vger.kernel.org \
--cc=saeedm@mellanox.com \
--cc=tariqt@mellanox.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).