Problems with PCI-E devices not being detected with switch

[Stefan Roese]

On Thursday 16 October 2008, Benjamin Herrenschmidt wrote:
> On Thu, 2008-10-16 at 10:03 +0200, Stefan Roese wrote:
> > Doing this unconditionally is not a good idea since we could have an old
> > (buggy) firmware which didn't configure the PCIe controller correctly.
> > But I really like your idea with the device-tree property to optionally
> > skip this re-configuration. Now we only need to find some "volunteer" to
> > do this job... ;)
>
> I don't have a problem adding support for testing that property and
> skipping most of the initial HW setup, basically treating the endpoint
> as pre-configured.
>
> What about using a value for "status" ?

No. "status" is already used to disable/skip the PCIe slot completely. For 
example on Canyonlands where PCIe#0 is multiplexed with the SATA port.

> Or an empty "configured" 
> property ? Ideally, it should have been the other way around, ie
> "unconfigured" for old/buggy stuff but I'm worried there may be existing
> out-of-tree device-trees without it :-)

Yeah. I could add this "configured" property to the current U-Boot version. 
Perhaps we should add some version information to it so that Linux could 
eventually decide to re-configure when the "configured" version is known to 
be buggy. What do you think?

Best regards,
Stefan