Perry Metzger wrote:
> Carl Shapiro writes:
> >    The second problem is that the boot process hangs at the ep1 device
> >    "stage".  I have re-installed my 3com 3c589d ethernet card and it
> >    seems to have been successfully probed, but the boot hangs here and
> >    goes no further, no panic nothing.  The message is:
> > 
> >    ep1 at pcmcia0 function 0 port 0x330-0x33f: 3Com 3C589 Ethernet
> >    ep1: MAC address 00:60:08:8a:68:d3
> > 
> > I have noticed the exact same thing on an IBM Thinkpad 560, as well
> > as a Compaq Contura.  In both cases all devices that are not in use
> > are disabled in the firmware.
> 
> Merely disabling things in firmware may not in all cases be
> sufficient, though. Sigh.
> 
> Probably the best thing to do is to try getting a fully custom kernel
> built and see if it works for you.

Another data point:

I'm using a custom config file, which has

ep0 at pcmcia0 function 0
ep0 at pcmcia? function ?

The latest working kernel i have which works on my laptop (ThinkPad
760EL) is dated December 8 (built from sources dated approximately the
same), 1.3_BETA.

It works great:

ep0 at pcmcia0 function 0 port 0x330-0x33f: 3Com 3C589 Ethernet
ep0: MAC address 00:60:97:8c:fd:37
ep0: 8KB byte-wide FIFO, 5:3 Rx:Tx split, utp/aui/bnc (default utp)


1.3 kernels hang after:

ep0 at pcmcia0 function 0 port 0x3c0-0x3cf: 3Com 3C589 Ethernet

I.e. they don't even print the MAC address.  Note that the address
that the board was configured at was changed.


I poked around the source tree (sys/dev and sys/arch/i386, in
particular), looking for changes between 1.3_BETA and 1.3, but
couldn't find _any_ changes that looked suspicious.



chris