Philip Blundell writes:
>
> >I suspect that this could be related to the Outbound write flush problem
> >(ie, issuing an outbound write flush to the DC21285 causes deadlock).
>
> Wasn't the problem with outbound flush just that that functionality didn't
> actually get implemented?
>From what I remember reading, and from what Dave Gilberts' description
of what happened when he tried it with the tulip driver, I think it
does deadlock if attempted on rev. 01 21285s...
> It sounds like the 21285 is trying to perform some transaction on the bus
> either to satisfy a read or to unload its write FIFO (and it's stalled the
> SA110 while it does so) but it keeps getting a retry response. Can you
> examine the PCI bus and see what's going on in terms of the address that the
> Footbridge asserts and the response it gets from the target?
I'll try, but all my slots are full, and I can only get to the 'B' side
of the PCI connectors safely. This means that I can get to all but
CBE0, STOP, SDONE and TRDY signals. I'll also reduce the bus speed
to make it easier on the poor 60MHz scope...
_____
|_____| ------------------------------------------------- ---+---+-
| | Russell King [EMAIL PROTECTED] --- ---
| | | | http://www.arm.linux.org.uk/~rmk/armlinux.html / / |
| +-+-+ --- -+-
/ | THE developer of ARM Linux |+| /|\
/ | | | --- |
+-+-+ ------------------------------------------------- /\\\ |
unsubscribe: body of `unsubscribe linux-arm' to [EMAIL PROTECTED]
