Re: Re: 82557 driver can drive 82559..?

On 2 Sep 2002, atul srivastava wrote:

> >firmware change in a later chip caused them to take an 
> >unpredictable (or
> >just undocumented) amount of time.
> >
> > 	outl(0, ioaddr + SCBPointer);
> > 	inl(ioaddr + SCBPointer);				/* Flush to PCI. */
> > 	udelay(10);					/* Bogus, but it avoids the bug. */
...
> my query is that does above behaviour is indicated by 
> 82559/82559er cards everwhere or it also depends on
> which board these cards are inserted in..?

I haven't traced the PCI transactions associated with these lines of
code.

It's very possible that the bug here depends on both the firmware (chip
revision) and the specific PCI bus implementation.

-- 
Donald Becker				[email protected]
Scyld Computing Corporation		http://www.scyld.com
410 Severn Ave. Suite 210		Second Generation Beowulf Clusters
Annapolis MD 21403			410-990-9993