Hi,

I reread all of the posts about memory size checking for the ae ethernet
boards, and would like to know if this issue is resolved. It seems that
the last word was that reversing the memory test order would resolve most
issues; how many people have tested this so far, and has anyone found any
cds which cause problems?

Perhaps, if noone has any problems to report, it's time for the code to go
into the source tree proper.

I am currently testing the code with a Quadra 950 with five ae type
ethernet cards. Yes, this machine is a router and needs six ethernet
interfaces!

ae0 at nubus0 slot a: MacCon Ethernet, 32KB memory
ae0: Ethernet address 00:00:94:21:5e:a7
ae1 at nubus0 slot b: Ethernet A-Series, 32KB memory
ae1: Ethernet address 00:40:10:50:1f:5c
ae2 at nubus0 slot c: MacCon NuBus-A , 32KB memory
ae2: Ethernet address 00:00:94:60:a2:e5
ae3 at nubus0 slot d: MacCon Ethernet, 32KB memory
ae3: Ethernet address 00:00:94:61:53:ce
ae4 at nubus0 slot e: MacCon Ethernet, 32KB memory
ae4: Ethernet address 00:00:94:61:65:a0

And here is the list after the patch:

ae0 at nubus0 slot a: MacCon Ethernet, 64KB memory
ae0: Ethernet address 00:00:94:21:5e:a7
ae1 at nubus0 slot b: Ethernet A-Series, 64KB memory
ae1: Ethernet address 00:40:10:50:1f:5c
ae2 at nubus0 slot c: MacCon NuBus-A , 32KB memory
ae2: Ethernet address 00:00:94:60:a2:e5
ae3 at nubus0 slot d: MacCon Ethernet, 64KB memory
ae3: Ethernet address 00:00:94:61:53:ce
ae4 at nubus0 slot e: MacCon Ethernet, 64KB memory
ae4: Ethernet address 00:00:94:61:65:a0
aed0 at adb0 addr 0: ADB Event device

And a Quadra 605, after:
ae0 at nubus0 slot e: Farallon EtherMac LC-TP, 64KB memory
ae0: Ethernet address 00:00:c5:43:3b:48

Does anyone else have experiences, good or bad, with the patches?

John Klos
Sixgirls Computing Labs