Re: pci bridge suspend/resume (Re: PCI not restored correctly after suspend to ram)

[Grant Grundler <grundler-6jwH94ZQLHl74goWV3ctuw@public.gmane.org>]

On Wed, Mar 31, 2004 at 08:28:34AM +0100, Russell King wrote:
> Also, if what you say is true, these device drivers can only ever be
> inserted and never removed and re-inserted - which would be a driver
> bug.

We could call it a driver bug if we like. But reality is we don't
know what magic BIOS is doing and can't fix the problem. Trying
to get gfx cards working on parisc-linux is an example of that.

Can someone provide a cleaner example than PCI-PCI Bridge of generic 
class driver?
Maybe from serial or LAN cards? (eg 16550 or tulip chip support)
Talking about a concrete example(s) might define the problem(s) better.

PCI-PCI Bridge class support is integrated in the generic PCI bus
enumeration in order to walk PCI config space and ultimately configure
devices behind the PCI-PCI bridge. A seperate PCI-PCI Bridge "class"
driver could do some other house-keeping stuff to support hot plug.

thanks,
grant


-------------------------------------------------------
This SF.Net email is sponsored by: IBM Linux Tutorials
Free Linux tutorial presented by Daniel Robbins, President and CEO of
GenToo technologies. Learn everything from fundamentals to system
administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click