linux-kernel - Re: [PATCH v2] net: phy: Prevent reporting advertised modes when autoneg is off

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: <20201016210055.GK139700@lunn.ch>
Date:   Fri, 16 Oct 2020 23:00:55 +0200
From:   Andrew Lunn <andrew@...n.ch>
To:     Lukasz Stelmach <l.stelmach@...sung.com>
Cc:     Heiner Kallweit <hkallweit1@...il.com>,
        Russell King <linux@...linux.org.uk>,
        "David S. Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>, netdev@...r.kernel.org,
        linux-kernel@...r.kernel.org,
        Bartłomiej Żolnierkiewicz 
        <b.zolnierkie@...sung.com>,
        Marek Szyprowski <m.szyprowski@...sung.com>
Subject: Re: [PATCH v2] net: phy: Prevent reporting advertised modes when
 autoneg is off

On Fri, Oct 16, 2020 at 09:37:22PM +0200, Lukasz Stelmach wrote:
> It was <2020-10-16 pią 20:09>, when Andrew Lunn wrote:
> > On Thu, Oct 15, 2020 at 10:44:35AM +0200, Łukasz Stelmach wrote:
> >> Do not report advertised link modes (local and remote) when
> >> autonegotiation is turned off. mii_ethtool_get_link_ksettings() exhibits
> >> the same behaviour and this patch aims at unifying the behavior of both
> >> functions.
> >
> > Does ethtool allow you to configure advertised modes with autoneg off?
> > If it can, it would be useful to see what is being configured, before
> > it is actually turned on.
> >
> > ethtool -s eth42 autoneg off advertise 0xf
> >
> > does not give an error on an interface i have.
> 
> Yes, this is a good point. Do you think I should change the if()[1] in 
> mii_ethtool_get_link_ksettings() instead? I really think these two
> function should report the same.

Yes, i would change mii. Ideally we want all drivers to use
phylib/phylink, not mii. So i would modify mii to match
phylib/phylink, not the other way around.

And then there will be drivers which do their own PHY handling, hidden
away in firmware, and not using either of mii or phylib/phylink. You
can expect them to be inconsistent.

    Andrew