If anyone has any further information on the software fix/patch for
this issue I would be very interested in hearing about it (and
backporting into the kernel from late last year I am using) - or even
the best way to search for patches to particular files.

Regards,
Andrew.

On 12 March 2014 11:41, Loren Amelang <[email protected]> wrote:
> Got it! (Chrome print to PDF, copy from the print...)
>
> -----
> 3.7.1 PHYAD[2:0]: PHY Address Configuration
>
> The PHYAD[2:0] configuration straps are driven high or low to give each PHY
> a unique address. This address is latched into an internal register at the
> end of a hardware reset (default = 000b). In a multitransceiver application
> (such as a repeater), the controller is able to manage each transceiver via
> the unique address. Each transceiver checks each management data frame for a
> matching address in the relevant bits. When a match is recognized, the
> transceiver responds to that particular frame. The PHY address is also used
> to seed the scrambler. In a multi-transceiver application, this ensures that
> the scramblers are out of synchronization and disperses the electromagnetic
> radiation across the frequency spectrum.
>
> The device's SMI address may be configured using hardware configuration to
> any value between 0 and 7. The user can configure the PHY address using
> Software Configuration if an address greater than 7 is required. The PHY
> address can be written (after SMI communication at some address is
> established) using the PHYAD bits of the Special Modes Register. The
> PHYAD[2:0] configuration straps are multiplexed with other signals as shown
> in Table 3.5.
> -----
>
>
> From my A5C Schematic, showing processor names = LAN8710A names = Microchip
> names:
>
> GMII1_RXERR/RMII1_RXERR/SPI1_D1/I2C1_SCL/MCASP1_FSX/UART5_RTSN/UART2_TXD/GPIO3_2
> = RXER/PHYAD0 = RXER/RXD4/PHYAD0
>
> GMII1_RXCLK/UART2_TXD/RGMII1_RCLK/MMC0_DAT6/MMC1_DAT1/UART1_DSRN/MCASP0_FSX/GPIO3_10
> = REFCLKO = RXCLK/PHYAD1
>
> GMII1_RXCLK/UART2_TXD/RGMII1_RCLK/MMC0_DAT6/MMC1_DAT1/UART1_DSRN/MCASP0_FSX/GPIO3_10
> = RXD3/PHYAD2 = RXD3/PHYAD2
>
>
> Obviously a lot of multiplexed uses there! Microchip says, "This address is
> latched into an internal register at the end of a hardware reset (default =
> 000b)." But their reset is "SYS_RESETn", right? The same reset caused by the
> user reset button? How is the processor supposed to control those three
> signals while it is reset?
>
> So how do three bits default to "000b"? I guess they are counting the
> address bits you can only program in software? But if those default to 000b,
> how does my chip end up as phy[0]?
>
> Maybe that's where the mask comes in? Mine works with mask fffffffe:
> [    2.482812] davinci_mdio 4a101000.mdio: detected phy mask fffffffe
> [    2.490309] libphy: 4a101000.mdio: probed
> [    2.494577] davinci_mdio 4a101000.mdio: phy[0]: device 4a101000.mdio:00,
> driver SMSC LAN8710/LAN8720
> [    2.504409] Detected MACID = 90:59:af:4d:71:eb
> ...
> [    5.854282] libphy: PHY 4a101000.mdio:01 not found
> [    5.859335] net eth0: phy 4a101000.mdio:01 not found on slave 1
>
> While I see reports above in this thread that
> On Tuesday, November 26, 2013 2:22:42 PM UTC-8, AndrewTaneGlen wrote:
> On a good phy boot I see the following:
> [    2.810749] davinci_mdio 4a101000.mdio: davinci mdio revision 1.6
> [    2.817206] davinci_mdio 4a101000.mdio: detected phy mask fffffffe
> ...
> On a 'bad phy' boot I see the following (differences highlighted):
> [    2.806763] davinci_mdio 4a101000.mdio: davinci mdio revision 1.6
> [    2.813213] davinci_mdio 4a101000.mdio: detected phy mask fffffffb
>
>
> I'm betting the "fe" mask eliminates the default "000b" bits that are not
> set by hardware, allowing the good boot.
>
> Of the hardware set bits, it seems my chip must get address zero, so all
> three signals must be zero when reset ends. With mask "fe", only the
> hardware lines are checked.
>
> But... Where does the mask come from? I'm not finding it in the boot
> environment settings. It looks like it is read from the chip:
> [    2.817206] davinci_mdio 4a101000.mdio: detected phy mask fffffffe
> But it is not mentioned that I see in the Microchip doc.
>
> Always another mystery...
>
>
>
>
> On Tuesday, March 11, 2014 12:56:43 PM UTC-7, Gerald wrote:
>>
>> Look in the LAN 8710A data sheet from SMSC. I would cut an paste it, but
>> Microchip has cut and paste blocked.
>>
>> http://www.microchip.com/wwwproducts/Devices.aspx?product=LAN8710A Section
>> 3.7.1
>>
>>
>> Gerald
>>
> --
> For more options, visit http://beagleboard.org/discuss
> ---
> You received this message because you are subscribed to a topic in the
> Google Groups "BeagleBoard" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/beagleboard/9mctrG26Mc8/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to
> [email protected].
> For more options, visit https://groups.google.com/d/optout.

-- 
For more options, visit http://beagleboard.org/discuss
--- 
You received this message because you are subscribed to the Google Groups 
"BeagleBoard" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to