This is just a final note about this problem.  I won't say it is
"solved" but at least we have a workaround.

This is from 4.10 release, src/dev/fxp/if_fxp.c, lines 509 et ff.
with the 'offending' code ifdef'd out by me:

        /*
         * Determine whether we must use the 503 serial interface.
         */
        fxp_read_eeprom(sc, &data, 6, 1);
#ifndef BYPASS_503_SERIAL_CHECK
        if ((data & FXP_PHY_DEVICE_MASK) != 0 &&
            (data & FXP_PHY_SERIAL_ONLY))
                sc->flags |= FXP_FLAG_SERIAL_MEDIA;
#else
                sc->flags &= ~FXP_FLAG_SERIAL_MEDIA;
#endif

        /*
         * Create the sysctl tree
         */

When compiled and ran on an adlink motherboard (EBC-2000),
it now correctly IDs and configures the three fxp devices
on the motherboard:

fxp0: <Intel 82559 Pro/100 Ethernet> port 0xd400-0xd43f mem 
0xe7200000-0xe72fffff,0xe7302000-0xe7302fff irq 11 at device 8.0 on pci0
fxp0: Ethernet address 00:30:64:01:86:ff
inphy0: <i82555 10/100 media interface> on miibus0
inphy0:  10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
fxp1: <Intel 82559 Pro/100 Ethernet> port 0xd800-0xd83f mem 
0xe7000000-0xe70fffff,0xe7301000-0xe7301fff irq 5 at device 9.0 on pci0
fxp1: Ethernet address 00:30:64:01:94:00
inphy1: <i82555 10/100 media interface> on miibus1
inphy1:  10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
fxp2: <Intel 82559 Pro/100 Ethernet> port 0xdc00-0xdc3f mem 
0xe7100000-0xe71fffff,0xe7300000-0xe7300fff irq 12 at device 10.0 on pci0
fxp2: Ethernet address 00:30:64:01:94:01
inphy2: <i82555 10/100 media interface> on miibus2
inphy2:  10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto

and I can 'ifconfig fxpN media xyz' between the supported
media types. 

Once this project is shipped, I'll try and track down the
problem's root cause.  This might still be a hardware issue
(the eeprom is being read wrong, or is not reporting capabilities
correctly, or ???).  

Thanks to Simon Barner and others for helping out with the 
OP.

rip

At 16:21 28/06/2004. Richard P. Williamson had this to say:
>At 16:52 25/06/2004. Simon Barner had this to say:
>>Richard P. Williamson wrote:
>>
>>[...]
>>
>>Here another thing you could try: I once had a problem with a 3Com NIC
>>not being detected properly when the driver was compiled statically into
>>the kernel.
>>
>>Removing it and using the module instead made it work...
>
>Nope, still no joy.  However my boot is telling me userland is out of
>sync with the kernel now, so I'm not prepared to say "it didn't work"
>yet.
>
>Per the numbers returned by pciconf, this is a 82559 rev 0x8, which
>is in the table of supported devices.  According to the code, however,
>if the device claims to be 10Mb only, then it isn't even handed to the
>miibus code to configure.  I'm still working my way through the code to
>determine if the 10Mb-ness is being reported by the devices directly,
>or if this is the code's interpretation based on other data requested.
>
>The two additional 82559 rev 0x8s (on the plug-in PCI card)
>are correctly being passed to the miibus, and this is annoying.
>
>Thanks for your time in this, anyway.
>
>MfG,
>rip
>
>_______________________________________________
>[EMAIL PROTECTED] mailing list
>http://lists.freebsd.org/mailman/listinfo/freebsd-questions
>To unsubscribe, send any mail to "[EMAIL PROTECTED]"

_______________________________________________
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Reply via email to