netdev - RE: [PATCH RFC 2/6] dpaa2-eth: add support for new link state APIs

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

Message-ID: <VI1PR0402MB2800CB7ABB8DE4472A80D586E0EE0@VI1PR0402MB2800.eurprd04.prod.outlook.com>
Date:   Fri, 14 Jun 2019 14:03:59 +0000
From:   Ioana Ciornei <ioana.ciornei@....com>
To:     Andrew Lunn <andrew@...n.ch>
CC:     "linux@...linux.org.uk" <linux@...linux.org.uk>,
        "hkallweit1@...il.com" <hkallweit1@...il.com>,
        "f.fainelli@...il.com" <f.fainelli@...il.com>,
        "davem@...emloft.net" <davem@...emloft.net>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        Alexandru Marginean <alexandru.marginean@....com>,
        Ioana Ciocoi Radulescu <ruxandra.radulescu@....com>,
        Valentin-catalin Neacsu <valentin-catalin.neacsu@....com>
Subject: RE: [PATCH RFC 2/6] dpaa2-eth: add support for new link state APIs

> Subject: Re: [PATCH RFC 2/6] dpaa2-eth: add support for new link state APIs
> 
> >  /**
> > + * Advertised link speeds
> > + */
> > +#define DPNI_ADVERTISED_10BASET_FULL           BIT_ULL(0)
> > +#define DPNI_ADVERTISED_100BASET_FULL          BIT_ULL(1)
> > +#define DPNI_ADVERTISED_1000BASET_FULL         BIT_ULL(2)
> > +#define DPNI_ADVERTISED_10000BASET_FULL        BIT_ULL(4)
> 
> So 10 Half and 100Half are not supported by the PHYs you use?  What happens if
> somebody does connect a PHY which supports these speeds? Do you need to
> change the firmware? I suppose you do anyway, since it is the firmware which is
> driving the PHY.

First of all, if the firmware had access to the open-source PHY driver code, the design wouldn't have been like this. But as it is, the DPMAC object/driver is used squarely as a way to gather this information from the PHY library.

Half duplex modes are not supported in our MAC. This is why the firmware does not export any advertisement bits for these modes.
If somebody connects a PHY which supports 10 Half and 100Half modes, the intersection of the MAC capabilities and the PHY one's will be only the Full duplex modes.

> 
> >  struct dpni_link_state {
> >  	u32	rate;
> >  	u64	options;
> > +	u64	supported;
> > +	u64	advertising;
> >  	int	up;
> > +	int	state_valid;
> >  };
> 
> Does the firmware report Pause? Asym Pause? EEE? Is this part of options? Can
> you control the advertisement of these options?

The firmware knows about conveying the pause and asym pause configuration from/to the mac driver.

EEE is not a feature supported by our MAC so there is no configuration knob for this.

--
Ioana

> 
>      Andrew