Re: [ACPI] Re: [Proposal]Another way to save/restore PCI config space for suspend/resume

[Benjamin Herrenschmidt <benh@kernel.crashing.org>]

On Wed, 2004-10-27 at 02:30 -0400, Len Brown wrote:
> Li Shaohua wrote:
> > On Tue, 2004-10-26 at 17:21, Pavel Machek wrote:
> 
> >>>Here is a another idea: 
> >>>Record all PCI writes in Linux kernel...
> >>
> >>That looks extremely ugly to me. If you want to do something special
> >>in resume function, just do it there. It will probably share a lot of
> >>code with your init function, anyway.
> > 
> > How can you handle devices without driver? And how to save/restore
> > config space for special devices, such as LPC bridge and host bridge?
> 
> Say that writing the missing drivers is the only workable solution.
> Does anybody have an estimate of how many there are and how big
> a task that would be?

This has been discussed a bit on linux-pm (on osdl lists, it's a new
list to discuss PM specific matters). I tend to think the core should
know at least a few "standard" things like P2P bridges, by simply
saving/restoring a bigger chunk of config space. We also need to fix the
current restore code I suppose, so that it disables IO & MEM, then
restore all registers, then re-enable those 2 ones...

I don't want to cross-post between lists, but people are welcome to join
linux-pm to talk about implementation issues related to power
management, including the changes we are trying to define to the various
driver callbacks. 

Ben.