On Sunday 01 July 2007 17:00:06 Lennert Buytenhek wrote:

On Sun, Jul 01, 2007 at 12:23:16PM +0200, Michael Buesch wrote:

quoted

quoted

More or less.  You can't add the resistances like that, since the
bus isolation chip buffers the IDSEL signal, but it is correct that
if the host's IDSEL resistor is larger than a certain value, the
combination of the resistive coupling of IDSEL plus the extra buffer
in the isolator might be causing the IDSEL input on the 'guest' PCI
board to assert too late (or not assert at all), causing config
accesses to fail.

(This also depends on the specific 'guest' PCI board used, as you
noted, due to differing IDSEL trace lengths/capacitances and input
pin capacitances on different PCI boards.  Also, it might work at
33 MHz but not work at 66 MHz, etc.)

It doesn't work on any of my boards :(

What extender board is this?  Do you have docs/schematics?

catalyst pcibx32
http://bu3sch.de/pcibx.php
Docs yes, schematics no.

And what motherboard brand/type?

ABit AI7
The other was some MSI and some very old random board. dunno.
It works perfectly fine with other cards, like a linksys
wlan card with a broadcom 4318 chip. It's just the b44
that doesn't work in the extender.

Actually, the IDSEL resistor would be on the computer's
motherboard, not on the PCI board.  And to which address line

Yeah, I know.

the IDSEL line is connected depends on which PCI slot on the
motherboard you're looking at.

Sure.

A multimeter should do the trick, but I would advise against this
if you're not totally comfortable with hacking hardware.

Well, you mean to measure the idsel against each possible AD line?
It's difficult, because the motherboard is inside of a standard
computer case and a watercooling system is mounted. So I would
have to disassemble all that stuff. :/
Probably I can measure it with very thin probes on the slots
without unmounting the board, hm...

-- 
Greetings Michael.