From: Jason Gunthorpe <jgg@nvidia.com>
To: David Ahern <dsahern@kernel.org>
Cc: Jiri Pirko <jiri@resnulli.us>, Jakub Kicinski <kuba@kernel.org>,
Dan Williams <dan.j.williams@intel.com>,
Jonathan Corbet <corbet@lwn.net>,
Itay Avraham <itayavr@nvidia.com>,
Leon Romanovsky <leon@kernel.org>,
linux-doc@vger.kernel.org, linux-rdma@vger.kernel.org,
netdev@vger.kernel.org, Paolo Abeni <pabeni@redhat.com>,
Saeed Mahameed <saeedm@nvidia.com>,
Tariq Toukan <tariqt@nvidia.com>,
Andy Gospodarek <andrew.gospodarek@broadcom.com>,
Aron Silverton <aron.silverton@oracle.com>,
Christoph Hellwig <hch@infradead.org>,
Jiri Pirko <jiri@nvidia.com>, Leonid Bloch <lbloch@nvidia.com>,
Leon Romanovsky <leonro@nvidia.com>,
linux-cxl@vger.kernel.org, patches@lists.linux.dev
Subject: Re: [PATCH 0/8] Introduce fwctl subystem
Date: Fri, 7 Jun 2024 12:14:51 -0300 [thread overview]
Message-ID: <20240607151451.GL19897@nvidia.com> (raw)
In-Reply-To: <887d1cb7-e9e9-4b12-aebb-651addc6b01c@kernel.org>
On Fri, Jun 07, 2024 at 08:50:17AM -0600, David Ahern wrote:
> Mellanox offers both with the Spectrum line and should have a pretty
> good understanding of how many customers deploy with the SDK vs
> switchdev. Why is that?
We offer lots of options with mlx5 switching too, and switchdev is not
being selected by customers principally for performance reasons, in my
view.
The OVS space wants to operate the switch much like a firewall and
this creates a high rate of database updates and exception
packets. DPDK can operate all the same offload HW from userspace and
avoid all the system call and other kernel overhead. It is much more
purpose built to what OVS wants to do. In the >50Gbps space this
matters a lot and overall DPDK performance notably wins over switchdev
for many OVS workloads - even though the high speed path is
near-identical.
In this role DPDK is effectively a switch SDK, an open source one at
least.
Sadly I'm seeing signs that proprietary OVS focused SDKs (think
various P4 offerings and others) are out competing open DPDK on
merit :(
For whatever reason the market for switching is not strongly motivated
toward open SDKs, and the available open solutions are struggling a
bit to compete.
But to repeat again, fwctl is not for dataplane, it is not for
implementing a switch SDK (go use RDMA if you want to do that). I will
write here a commitment to accept patches blocking such usages if
drivers try to abuse the purpose of the subsystem.
Jason
next prev parent reply other threads:[~2024-06-07 15:14 UTC|newest]
Thread overview: 73+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-06-03 15:53 [PATCH 0/8] Introduce fwctl subystem Jason Gunthorpe
2024-06-03 15:53 ` [PATCH 1/8] fwctl: Add basic structure for a class subsystem with a cdev Jason Gunthorpe
2024-06-04 9:32 ` Leon Romanovsky
2024-06-04 15:50 ` Jason Gunthorpe
2024-06-04 17:05 ` Jonathan Cameron
2024-06-04 18:52 ` Jason Gunthorpe
2024-06-05 11:08 ` Jonathan Cameron
2024-06-04 16:42 ` Randy Dunlap
2024-06-04 16:44 ` Jason Gunthorpe
2024-06-03 15:53 ` [PATCH 2/8] fwctl: Basic ioctl dispatch for the character device Jason Gunthorpe
2024-06-04 12:16 ` Zhu Yanjun
2024-06-04 12:22 ` Leon Romanovsky
2024-06-04 16:50 ` Jonathan Cameron
2024-06-04 16:58 ` Jason Gunthorpe
2024-06-05 11:07 ` Jonathan Cameron
2024-06-05 18:27 ` Jason Gunthorpe
2024-06-06 13:34 ` Jonathan Cameron
2024-06-06 15:37 ` Randy Dunlap
2024-06-05 15:42 ` Przemek Kitszel
2024-06-05 15:49 ` Jason Gunthorpe
2024-06-03 15:53 ` [PATCH 3/8] fwctl: FWCTL_INFO to return basic information about the device Jason Gunthorpe
2024-06-13 23:32 ` Dave Jiang
2024-06-13 23:40 ` Jason Gunthorpe
2024-06-14 16:37 ` Dave Jiang
2024-06-03 15:53 ` [PATCH 4/8] taint: Add TAINT_FWCTL Jason Gunthorpe
2024-06-03 15:53 ` [PATCH 5/8] fwctl: FWCTL_RPC to execute a Remote Procedure Call to device firmware Jason Gunthorpe
2024-06-03 15:53 ` [PATCH 6/8] fwctl: Add documentation Jason Gunthorpe
2024-06-05 2:31 ` Randy Dunlap
2024-06-05 16:03 ` Jason Gunthorpe
2024-06-05 20:14 ` Randy Dunlap
2024-06-03 15:53 ` [PATCH 7/8] fwctl/mlx5: Support for communicating with mlx5 fw Jason Gunthorpe
2024-06-03 15:53 ` [PATCH 8/8] mlx5: Create an auxiliary device for fwctl_mlx5 Jason Gunthorpe
2024-06-03 18:42 ` [PATCH 0/8] Introduce fwctl subystem Jakub Kicinski
2024-06-04 3:01 ` David Ahern
2024-06-04 14:04 ` Jakub Kicinski
2024-06-04 21:28 ` Saeed Mahameed
2024-06-04 22:32 ` Jakub Kicinski
2024-06-05 14:50 ` Jason Gunthorpe
2024-06-05 15:41 ` Jakub Kicinski
2024-06-04 23:56 ` Dan Williams
2024-06-05 3:05 ` Jakub Kicinski
2024-06-05 11:19 ` Jonathan Cameron
2024-06-05 13:59 ` Jason Gunthorpe
2024-06-06 2:35 ` David Ahern
2024-06-06 14:18 ` Jakub Kicinski
2024-06-06 14:48 ` Jason Gunthorpe
2024-06-06 15:05 ` Jakub Kicinski
2024-06-06 17:47 ` David Ahern
2024-06-07 6:48 ` Jiri Pirko
2024-06-07 14:50 ` David Ahern
2024-06-07 15:14 ` Jason Gunthorpe [this message]
2024-06-07 15:50 ` Jiri Pirko
2024-06-07 17:24 ` Jason Gunthorpe
2024-06-07 7:34 ` Jiri Pirko
2024-06-07 12:49 ` Andrew Lunn
2024-06-07 13:34 ` Jiri Pirko
2024-06-08 1:43 ` Jakub Kicinski
2024-06-06 4:56 ` Dan Williams
2024-06-06 8:50 ` Leon Romanovsky
2024-06-06 22:11 ` Dan Williams
2024-06-07 0:02 ` Jason Gunthorpe
2024-06-07 13:12 ` Leon Romanovsky
2024-06-06 14:41 ` Jason Gunthorpe
2024-06-06 14:58 ` Jakub Kicinski
2024-06-06 17:24 ` Dan Williams
2024-06-07 0:25 ` Jason Gunthorpe
2024-06-07 10:47 ` Przemek Kitszel
2024-06-11 15:36 ` Daniel Vetter
2024-06-11 16:17 ` Jason Gunthorpe
2024-06-11 16:54 ` Daniel Vetter
2024-06-06 1:58 ` David Ahern
2024-06-05 3:11 ` Jakub Kicinski
2024-06-05 12:06 ` Jason Gunthorpe
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=20240607151451.GL19897@nvidia.com \
--to=jgg@nvidia.com \
--cc=andrew.gospodarek@broadcom.com \
--cc=aron.silverton@oracle.com \
--cc=corbet@lwn.net \
--cc=dan.j.williams@intel.com \
--cc=dsahern@kernel.org \
--cc=hch@infradead.org \
--cc=itayavr@nvidia.com \
--cc=jiri@nvidia.com \
--cc=jiri@resnulli.us \
--cc=kuba@kernel.org \
--cc=lbloch@nvidia.com \
--cc=leon@kernel.org \
--cc=leonro@nvidia.com \
--cc=linux-cxl@vger.kernel.org \
--cc=linux-doc@vger.kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=patches@lists.linux.dev \
--cc=saeedm@nvidia.com \
--cc=tariqt@nvidia.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).