On Wed, Jan 23, 2019 at 08:15:27PM +0100, Lukas Wunner wrote:
> On Wed, Jan 23, 2019 at 12:07:38PM -0700, Keith Busch wrote:
> > Right, so in Alex's case, it looks like we are observing
> > pcie_wait_for_link() returning true before the PDC event.
> >
> > I'm wondering about PDS because if the link is up but Presence reports an
> > empty slot, does that matter for any implementations? Or is it perfectly
> > fine to enumerate an active link on an empty slot? An empty slot and
> > active link doesn't make a lot of sense, but that observation appears to
> > be what is reported here.
>
> We allow enumeration with an active link of an allegedly empty slot
> because some hardware hardwires PDS to zero, see commit 80696f991424
> ("PCI: pciehp: Tolerate Presence Detect hardwired to zero").
>
> The brokenness of PCIe hotplug implementations found in the wild is
> astonishing.
Heh, sounds like Alex's proposal may work until we find a slot
implementation that doesn't do DLLSC as expected. :)
