Hello again

>Does Angel go off and probe for PCI devices?  If so a master abort is to be
>expected.  Otherwise yes, I would agree it's odd.
>
Ah yes, I see this is the case. I guess I assumed that this condition would
get cleared and thus if it persisted it means it is recurring (which is a
problem?). Maybe its no big deal so nobody bothers to clear it?

The PCI registers of 21285 immediately prior to returning from
"footbridge_attach" in NetBSD (I.O.W. the PCI bus has been probed and
devices
attached) are below. The Angel values given in brackets, are what Angel for
Ethernet sets up .

Vendor: 0x1011
Device: 0x1065
Command: 0x0017 (Angel : 0x0007)
Revision: 0x4
Class Code: 0x0b4001
Cache Line Sz: 0x00
Lat Timer:0x00       (Angel : 0xff)
Header Type:0x00
BIST:0x00
BAR 0x10 0x40000000 (Angel : 0x00100080)
BAR 0x14 0x0000f001   (Angel :0x00000481)
BAR 0x18 0x00000008  (Angel :0x00200008)
BAR's  0x1c TO 0x24 ALL : 0x00000000
Cardbus CIS 0x00000000
Subsystem ID:0x0000
Subsystem Vendor ID: 0x0000
Expansion ROM Base: 0x00000000 (Angel : 0xfff00001)
CapPtr:0x70
Int Line:0x01
Int Pin:0x00
Min Gnt:0x80 (Angel 0x00 weird cos these are ReadOnly??)
Max Lat:0x00 (Angel 0x00 weird cos these are ReadOnly??)

I see the NetBSD values (presumeably setup by the bootloader) appear to be
what the EBSA manual describes the 21285 configuration (page 4-7 of EBSA 285
reference manual).

The Angel values are significantly different. I haven't yet considered the
implications of this. The Latency Timer value might be problematic (0x00 vs.
0xff).

Also I see the BAR values are quite different?


What should NetBSD's 21285 values be? Comments anyone please. I'd really
appreciate it.

Thanks a 1000000.

Cheers
D