This is the mail archive of the ecos-discuss@sources.redhat.com mailing list for the eCos project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

RE: FEC driver problem


The connections to the PHY are a little platform specific.  In particular,
the driver in our CVS tree is rather oriented to the A&M Viper.

You may need to refer to your hardware documentation to see how your PHY
is connected to the FEC - in particular any control lines on the PHY like
reset/enable, etc.  That's what this piece of code in if_fec.c is all
about (which I intend to make more general someday):


    // Shut down ethernet, in case it is already running
    fec->eControl = eControl_RESET;
    eppc->pip_pbdat &= ~0x00004000;  // Reset PHY chip
    CYGACC_CALL_IF_DELAY_US(1000);   // 1ms
    eppc->pip_pbdat |= 0x00004000;   // Enable PHY chip

On 28-Jun-2001 Christoph Csebits wrote:
> hi
> 
> i have MPC860T based custom board
> and try to activate the FEC
> driver ( viper based )
> Working with a recent cvs version.
> 
> it seems that the MII communication
> is not working properly.
> 
> after enabling the FEC i am trying to
> get the id register of the PHY (DP83843 PHYTER)
> ( containing an organizational unique id ) but
> i always get 0x60E7ffff. (0xffff = data)
> (reading every other register results in the
> same behaviour)
> 
> if_fec.c:
> 
>   fec->eControl = eControl_EN | eControl_MUX;
>   :
>   .
>   os_printf("%08x\n", phy_read(0x02, PHY_ADDRESS );
> 
> any idea?
> 
> best regards, christoph
>     
> -- 
> cd /pub
> more beer


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]