On Sat, 24 Oct 1998, Miles Lane wrote:

> To: Linux Networking <[EMAIL PROTECTED]>

This is probably best handled on the [EMAIL PROTECTED] list.

> I have downloaded the 2.0.35-2 sources/headers and the pcmcia-cs-3.0.5
> source.
> I have tweaked the 3c575_cb.c file to include the card id info for the
> 3CCFE575BT
> card (a later varient of the 3c575).

The 3ccfe575 is actually a completely new design, based on the 3Com Cyclone
core instead of the older Boomerang core.  It's *not at all* the same card!

>  I added the adapter info in the
> /etc/pcmcia/config
> per the instructions on one of the web pages on the pcmcia driver web
> pages.
> Finally, the pcmcia drivers load and work, the 3c575_cb driver loads and
> the system log
> shows that the card is getting recognized and configured, eth0 is being
> created
> and the network is getting started.

But I'll bet you are not getting more than one interrupt (as reported by
/proc/interrupts)..

What changed?  The 3ccfe575 implementes new brain-damage from the
PCMCIA committee.  Despite the fact that PCI handles shared interrupt just
fine, the committee felt that an new register was needed to clear the
card's latched interrupt, and this new register is in an additional (!)
address space.

So after the interrupt is handled normally, using code common to the PCI
version, the v0.99G driver clears this additional register.  (And yes, there
is a race condition here..)

> The only clue I have is that the indicator on the ethernet card shows
> that the card thinks it is
> accessing a 100BaseTX connection.  When I boot Win98 on the machine, the
> indicator shows
> a 10BaseT connection.

This is unrelated: the LED state is inverted by default (?!).
The driver can set bit 4 (0x0010) in the 'ResetOptions' to reverse the
state.  This is clearly a hack, since the most of the ResetOptions bits are
read-only, and the manual still lists bits 4-7 as "reserved".

Look for this change in v0.99H.

Donald Becker                                     [EMAIL PROTECTED]
USRA-CESDIS, Center of Excellence in Space Data and Information Sciences.
Code 930.5, Goddard Space Flight Center,  Greenbelt, MD.  20771
301-286-0882         http://cesdis.gsfc.nasa.gov/people/becker/whoiam.html

-
To unsubscribe from this list: send the line "unsubscribe linux-net" in
the body of a message to [EMAIL PROTECTED]

Reply via email to