determine whether there was an error (if so, we unmap the slot space
for this device). Use this functionality to tell the attach function
that we couldn't clear the card's buffer.
Also in aesetup(), eliminate the bogus repetition of the error message
we print when the buffer clear actually fails. Noticed by Hauke Fath.
and allow the attach function to override the default watchdog. Also,
do some minor cosmetic surgery (rename bus space tags/handles and some
KNFing I missed the first time around).
second argument. The NuBus autoconfig code had to be reorganized as a
result of this, and looks much more like a directly-attached bus now.
These changes eliminate __BROKEN_INDIRECT_CONFIG.
device and a printable "external name" (name + unit number), thus eliminating
if_name and if_unit. Updated interface to (*if_watchdog)() and (*if_start)()
to take a struct ifnet *, rather than a unit number.
Also change the device probing scheme to use something a bit more rational.
A current side-effect is that nubus cards are double-mapped. I expect
to fix that shortly.
Also change splclock() to block everything but serial hardware interrupts.
- split softc size and match/attach out from cfdriver into
a new struct cfattach.
- new "attach" directive for files.*. May specify the name of
the cfattach structure, so that devices may be easily attached
to parents with different autoconfiguration semantics.
Also, switch back from using struct ed_ring to using struct ae_ring
because BYTE_ORDER == BIG_ENDIAN, but the card is in x86 byte order
mode because NuBUS does the swap "for us" on 16- and 32-bit transfers.
We should probably figure some way to use the std ed_ring if/when
all the 8390 drivers are all fully merged.
caused by the VIA missing the ethernet slot interrupt. If we get a
watchdog, try calling the via's nubus interrupt function. If that
causes an interrupt, don't reset anything. If it doesn't, use old
behavior (log timeout and reset).
Yuck.
and bcopy and bzero were being used. The latter use longs to copy/zero,
and some ethernet cards can't handle longs. I've added bbcopy and bbzero
as a temporary hack, but these should later be added to locore and maybe
changed to use shorts. Still doesn't work: I can now get an interrupt on
the first transmit, but further transmits don't generate more interrupts.
