Re: How to reserve pci bus numbers for hotplug?

From: Greg KH
Date: Thu Dec 19 2013 - 19:55:27 EST


On Sat, Dec 14, 2013 at 11:15:39PM +0100, Andreas Noever wrote:
> On Sat, Dec 14, 2013 at 11:01 PM, Greg KH <gregkh@xxxxxxxxxxxxxxxxxxx> wrote:
> > On Thu, Dec 12, 2013 at 12:37:50AM +0100, Andreas Noever wrote:
> >>
> >> If I could get Linux to assign enough resources (bus numbers for now)
> >> then I could drop the acpi_osi parameter and make thunderbolt work
> >> after suspend... So, is there an easy way to fix this? (Quirks,
> >> reconfiguring bus number assignments from a platform driver, ...?)
> >
> > I think you are going to have to do a lot more here, what is needed is a
> > whole connection manager for Thunderbolt, emulating what the BIOS does,
> > in order to get all of this working properly.
> >
> > But don't let that stop you from trying, however the work involved is
> > really not trivial at all.
> >
> > good luck,
> >
> > greg k-h
>
> Actually the thunderbolt part is already working (for non chaining
> devices, see https://lkml.org/lkml/2013/11/28/492 if you are
> interested). I can completely bootstrap the tb side of things. The
> only problem is that if power has been cut by the firmware (during
> boot without osi_acpi=Darwin or after a suspend/resume cycle) then
> Linux does not allocate enough PCI resources...

Ah, that's nice to know, I didn't realize that.

There should be some Intel people around on the list that has access to
a Thunderbolt monitor for testing. Or, if needed, I can see if I can
expense one, as it would be good to get this working :)

thanks,

greg k-h
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/