From: Joe Damato <jdamato@fastly•com>
To: Edward Cree <ecree.xilinx@gmail•com>
Cc: Jakub Kicinski <kuba@kernel•org>,
netdev@vger•kernel.org, saeedm@nvidia•com, tariqt@nvidia•com,
ecree@solarflare•com, andrew@lunn•ch, davem@davemloft•net,
leon@kernel•org, pabeni@redhat•com, bhutchings@solarflare•com,
arnd@arndb•de, linux-kernel@vger•kernel.org
Subject: Re: [net 0/2] rxfh with custom RSS fixes
Date: Tue, 25 Jul 2023 13:47:19 -0700 [thread overview]
Message-ID: <20230725204719.GA1789@fastly.com> (raw)
In-Reply-To: <f565a8d6-e3b8-96d1-a7ac-212c64c60b1c@gmail.com>
On Tue, Jul 25, 2023 at 09:40:24AM +0100, Edward Cree wrote:
> On 24/07/2023 23:08, Jakub Kicinski wrote:
> > It'd be great to push the uAPI extensions back and make them
> > netlink-only, but we can't make Joe wait if it takes a long time
> > to finish up the basic conversion :(
>
> Yeah as I said upthread I don't think we should make Joe wait, if
> he's got a use case that actually needs it (have you, Joe? Or
> is it only GRXFH you need and the investigation just led you to
> notice SRXFH was broken?)
In short, yes: I'd like to be able to get and set the flow hash keys for
custom RSS contexts on mlx5 which is why I included the patch to mlx5 in
this series... but to be fair I am just one user :) I think it's really
up to you all on the direction you want to go.
Longer story: I am working on building a system which relies on custom RSS
contexts, flow rules to associate flows with RSS contexts (and thus
specific sets of queues), and epoll based busy poll. It's a long story ;)
I had considered changing the flow hash key to see if I could alter the
behavior of the system I am working on, but I ran into both issues
immediately (GRXFH and FLOW_RSS was not supported by mlx5, and SRXFH
was broken) which led me down the path of attempting to fix both so that I
could get and set the flow hash keys. I thought the code might be useful,
so I submit it upstream -- I was not aware of the netlink work (but it
looks really useful!).
It seems that the Mellanox folks are OK with the proposed driver change,
so I am going to send a v2 rebased on net-next with their requested changes.
next prev parent reply other threads:[~2023-07-25 20:47 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-07-23 15:06 [net 0/2] rxfh with custom RSS fixes Joe Damato
2023-07-23 15:06 ` [net 1/2] net: ethtool: Unify ETHTOOL_{G,S}RXFH rxnfc copy Joe Damato
2023-07-24 19:27 ` Edward Cree
2023-07-23 15:06 ` [net 2/2] net/mlx5: Fix flowhash key set/get for custom RSS Joe Damato
2023-07-25 9:59 ` Tariq Toukan
2023-07-26 8:28 ` Joe Damato
2023-07-24 19:27 ` [net 0/2] rxfh with custom RSS fixes Edward Cree
2023-07-24 21:36 ` Joe Damato
2023-07-24 22:08 ` Jakub Kicinski
2023-07-25 8:40 ` Edward Cree
2023-07-25 20:47 ` Joe Damato [this message]
2023-07-27 15:49 ` 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=20230725204719.GA1789@fastly.com \
--to=jdamato@fastly$(echo .)com \
--cc=andrew@lunn$(echo .)ch \
--cc=arnd@arndb$(echo .)de \
--cc=bhutchings@solarflare$(echo .)com \
--cc=davem@davemloft$(echo .)net \
--cc=ecree.xilinx@gmail$(echo .)com \
--cc=ecree@solarflare$(echo .)com \
--cc=kuba@kernel$(echo .)org \
--cc=leon@kernel$(echo .)org \
--cc=linux-kernel@vger$(echo .)kernel.org \
--cc=netdev@vger$(echo .)kernel.org \
--cc=pabeni@redhat$(echo .)com \
--cc=saeedm@nvidia$(echo .)com \
--cc=tariqt@nvidia$(echo .)com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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