netdev - Re: [PATCH ethtool] ethtool: add support show/set-hwtstamp

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-next>] [day] [month] [year] [list]

Date:   Wed, 2 Sep 2020 09:03:59 +0200
From:   Michal Kubecek <mkubecek@...e.cz>
To:     Kevin Yang <yyd@...gle.com>
Cc:     Andrew Lunn <andrew@...n.ch>, Networking <netdev@...r.kernel.org>,
        Neal Cardwell <ncardwell@...gle.com>,
        Eric Dumazet <edumazet@...gle.com>
Subject: Re: [PATCH ethtool] ethtool: add support show/set-hwtstamp

On Tue, Sep 01, 2020 at 08:36:08PM -0400, Kevin Yang wrote:
> On Tue, Sep 1, 2020 at 6:02 PM Andrew Lunn <andrew@...n.ch> wrote:
> > On Tue, Sep 01, 2020 at 05:20:09PM -0400, Kevin(Yudong) Yang wrote:
> > > Before this patch, ethtool has -T/--show-time-stamping that only
> > > shows the device's time stamping capabilities but not the time
> > > stamping policy that is used by the device.
> >
> > How does this differ from hwstamp_ctl(1)?
> 
> They are pretty much the same, both use ioctl(SIOCSHWTSTAMP).
> 
> Adding this to ethtool is because:
> - This time stamping policy is a hardware setting aligned with ethtool's
> purpose "query and control network device driver and hardware settings"
> - ethtool is widely used, system administrators don't need to install
> another binary to control this feature.

Adding this feature to ethtool IMHO makes good sense, I'm just not sure
if it's necessary to add new subcommands, perhaps we could add the
"show" part to -T / --show-time-stamping and add --set-time-stamping.

However, I don't like the idea of adding a new ioctl based interface to
ethtool while we are working on replacing and deprecating existing one.
I would much rather like adding this to the netlink interface (which
would, of course, require also kernel side implementation).

Michal