cvs commit: src/sys/dev/pci pci.c

[M. Warner Losh]

In message: <20040702120442.A8994 at grasshopper.cs.duke.edu>
            Andrew Gallatin <gallatin at cs.duke.edu> writes:
: Søren Schmidt [sos at deepcore.dk] wrote:
: > Dag-Erling Smørgrav wrote:
: > > Warner Losh <imp at FreeBSD.org> writes:
: > > 
: > >>  Log:
: > >>  Disable native ata support for now, too much breaks
: > > 
: > > 
: > > Could this be controlled by a hint or a tunable?
: > 
: > We'll soon need something better than that, since systems that needs 
: > this to boot at all will appear in the not too distant future.
: > Question is how to deal with old systems where the config is borked in 
: > some sense...
: 
: Is there any commonality between the broken systems?
: Mine is a ServerWorks based system.  Details at
: http://people.freebsd.org/~gallatin/ata_irq/
: 
: At the very least, we could have a quirk table..

ServerWorks seems to be a problem child for the new ata code.  It has
busted a few times now as changes have gone in.  I believe that if the
ServerWorks ata is broken, then I have a laptop that is also broken...

Warner