netdev - NETIF_N_NTUPLE - should it control flow classification?

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>] [day] [month] [year] [list]

Message-ID: <BLUPR0701MB20049CF05B3D3FA8BBC3599F8DE70@BLUPR0701MB2004.namprd07.prod.outlook.com>
Date:   Wed, 17 May 2017 10:25:54 +0000
From:   "Mintz, Yuval" <Yuval.Mintz@...ium.com>
To:     netdev <netdev@...r.kernel.org>
Subject: NETIF_N_NTUPLE - should it control flow classification?

Code-reading a bit, and I'm not sure of the exact definition of said bit;
Couldn't really deduce from the comment "N-tuple filters supported"
what falls under its jurisdiction and failed to find any other documentation.

I understand RFS acceleration is going to be disabled for an interface
when this is unset, and some drivers also pro-actively take action when
it's changed via set_features().

But there's seems to be some disagreement about the effect it has over
ethtool's flow classification [ETHTOOL_SRXCLSRLINS] -
Most drivers don't seem to care, but at least the intel drivers are
blocking received configurations in case user removed the feature.