Re: [PATCH v2 2/2] pci: Support "removable" attribute for PCI devices

[Rafael J. Wysocki]

On Wed, Apr 28, 2021 at 8:57 AM Oliver Neukum <oneukum@xxxxxxxx> wrote:
>
> Am Dienstag, den 27.04.2021, 12:59 +0000 schrieb David Laight:
> > From: Oliver Neukum
> > > Sent: 27 April 2021 13:00
>
> > > that is true for those options, but not for the style
> > > of PCI hotplug which requires you to push a button and wait
> > > for the blinking light.
> >
> > True, I remember some of those PCI hotplug chassis from 25 years ago.
> > ISTR we did get the removal events working (SVR4/Unixware) but I
> > don't remember the relevant chassis ever being sold.
> > In spite of the marketing hype I suspect it was only ever possible
> > to remove a completely working board and replace it with an
> > exactly equivalent one.
> >
> > In any case those chassis are not 'surprise removal'.
> >
> > More modern drivers are less likely to crash (and burn?) when
> > a PCI read returns ~0u.
> > But I suspect an awful lot really don't handle surprise removal
> > very well at all.
>
> So you are saying that these systems are so rare that it should be
> handled  as special cases if at all?

In principle, in the wake of Thunderbolt every PCI driver handling
PCIe devices needs to be able to deal with a device that's gone away
without notice, because in principle any PCIe device can be included
into a Thunderbolt docking station which may go away as a whole
without notice.