netdev - Re: [PATCH net-next v2] ethtool: add netlink based get rxfh support

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20221110161257.35d37983@kernel.org>
Date:   Thu, 10 Nov 2022 16:12:57 -0800
From:   Jakub Kicinski <kuba@...nel.org>
To:     "Samudrala, Sridhar" <sridhar.samudrala@...el.com>
Cc:     "Mogilappagari, Sudheer" <sudheer.mogilappagari@...el.com>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        "mkubecek@...e.cz" <mkubecek@...e.cz>,
        "andrew@...n.ch" <andrew@...n.ch>,
        "corbet@....net" <corbet@....net>,
        "Nguyen, Anthony L" <anthony.l.nguyen@...el.com>
Subject: Re: [PATCH net-next v2] ethtool: add netlink based get rxfh support

On Thu, 10 Nov 2022 17:24:15 -0600 Samudrala, Sridhar wrote:
> > The RSS context thing is a pretty shallow abstraction, I don't think we
> > should be extending it into "queue groups" or whatnot. We'll probably
> > need some devlink objects at some point (rate configuration?) and
> > locking order is devlink > rtnl, so spawning things from within ethtool
> > will be a pain :S  
> 
> We are going this path of extending ethtool rss context interface to support
> per queue-group parameters based on this feedback.
>    https://lore.kernel.org/netdev/20220314131114.635d5acb@kicinski-fedora-pc1c0hjn.dhcp.thefacebook.com/
> 
> Per queue-group rate can already be configured when creating queue groups via
> tc-mpqrio interface using bw_rlimit parameters.

Right, but that's still just flow-director-y/hash-y thing?
Does the name RSS imply purely hash based distribution?

My worry is that if we go with a more broad name like
"queue group" someone may be mislead to adding controls 
unrelated to flow <> queue assignment here.