Experimental Physics and
Industrial Control System

Andrew Johnson <[email protected]> · Fri, 16 Nov 2001 09:50:57 -0600

john sinclair wrote:
> 
> What is the sysBatDesc[] table and does it exist in the x86 architecture?

It's a PowerPC thing; I don't know if anything equivalent exists in that
architecture.  The 68K equivalent is the Transparent Translation registers
(TTRegs) if that helps.

> Mapping memory for PCI devices via sysPhysMemDesc is similarly
> problematic.

I suspect that some of the delay may be related to the autoconfiguration
of the PCI bus - our MVME2100 boards that do autoconfig take something
like 10+ seconds longer to startup than our MVME2700 boards which don't. 
If you don't autoconfigure, it's obviously possible to create suitable
sysPhysMemDesc[] entries for the hardware which you have to know about in
advance.

However having had a private email exchange with Nick last night I now
think that Mark's suggestion of the network autonegotiation is likely to
be the main culprit in his case.

- Andrew
-- 
Perfection is reached, not when there is no longer anything to add,
but when there is no longer anything to take away.
- Antoine de Saint-Exupery

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System