netdev - Re: [PATCH] net: phy: leds: Trigger leds only if PHY speed is known

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: <20210719152942.GQ22278@shell.armlinux.org.uk>
Date:   Mon, 19 Jul 2021 16:29:42 +0100
From:   "Russell King (Oracle)" <linux@...linux.org.uk>
To:     "Ivan T. Ivanov" <iivanov@...e.de>
Cc:     Andrew Lunn <andrew@...n.ch>,
        Heiner Kallweit <hkallweit1@...il.com>,
        "David S . Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>, netdev@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] net: phy: leds: Trigger leds only if PHY speed is known

On Fri, Jul 16, 2021 at 09:32:00PM +0300, Ivan T. Ivanov wrote:
> Hi,
> 
> Quoting Andrew Lunn (2021-07-16 18:19:58)
> > On Fri, Jul 16, 2021 at 05:11:42PM +0300, Ivan T. Ivanov wrote:
> > > This prevents "No phy led trigger registered for speed(-1)"
> > > alert message which is coused by phy_led_trigger_chage_speed()
> > > being called during attaching phy to net_device where phy device
> > > speed could be still unknown.
> > 
> > Hi Ivan
> > 
> > It seems odd that when attaching the PHY we have link, but not the
> > speed. What PHY is this?
> 
> This is lan78xx on RPi3B+
> 
> > 
> > > -     if (phy->speed == 0)
> > > +     if (phy->speed == 0 || phy->speed == SPEED_UNKNOWN)
> > >               return;
> > 
> > This change makes sense. But i'm wondering if the original logic is
> > sound. We have link, but no speed information.
> 
> Well, probably my interpretation was not correct. The most probable
> call to phy_led_trigger_change_speed() which couses this alert is
> phy_attach_direct() -> phy_led_triggers_register(), I think. I am
> not sure that we have link at this stage or not.

This does sound weird.

When a phy_device is allocated, it's explicitly initialised with:

	dev->speed = SPEED_UNKNOWN;
	dev->duplex = DUPLEX_UNKNOWN;
	dev->link = 0;
	dev->state = PHY_DOWN;

so, unless something is causing state to be read before we've attached
the phy to a network device, this is how this state should remain. I
wonder why you are seeing dev->link be non-zero.

-- 
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!