netdev - Re: [PATCH] stmmac: export RX mitigation control to device tree

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

Message-Id: <20191224.160459.513700743670795651.davem@davemloft.net>
Date:   Tue, 24 Dec 2019 16:04:59 -0800 (PST)
From:   David Miller <davem@...emloft.net>
To:     timofey.babitskiy@...dulum-instruments.com
Cc:     netdev@...r.kernel.org, peppe.cavallaro@...com,
        alexandre.torgue@...com, joabreu@...opsys.com
Subject: Re: [PATCH] stmmac: export RX mitigation control to device tree

From: Timofey Babitskiy <timofey.babitskiy@...dulum-instruments.com>
Date: Fri, 20 Dec 2019 12:56:50 +0000

> Having Rx mitigation via HW watchdog timer on is not suitable for time 
> servers distributing time via NTP and PTP protocol and relying on SW 
> timestamping, because Rx mitigation adds latency on receive and hence adds 
> path delay assymetry, which leads to time offset on timing clients. Turning 
> Rx mitigation off via platform config is not always a good option, because 
> some systems use default platform configs and only tune the device tree.
> 
> Signed-off-by: Timofey Babitskiy <timofey.babitskiy@...dulum-instruments.com>

The device tree is not an appropriate place to control what is already settable
via ethtool.

Make the appropriate ethtool setting on your NTP servers.