linuxppc-dev.lists.ozlabs.org archive mirror
 help / color / mirror / Atom feed
* why does ppc64 require boards to call pci_devs_phb_init()?
@ 2009-04-30 15:37 Kumar Gala
  2009-04-30 18:04 ` Kumar Gala
  2009-04-30 21:08 ` Benjamin Herrenschmidt
  0 siblings, 2 replies; 4+ messages in thread
From: Kumar Gala @ 2009-04-30 15:37 UTC (permalink / raw)
  To: Benjamin Herrenschmidt; +Cc: Linuxppc-dev Development

looking at merging this in w/ppc32 and wondering why board code has to  
call pci_devs_phb_init() and not do it via pcibios_init()

- k

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: why does ppc64 require boards to call pci_devs_phb_init()?
  2009-04-30 15:37 why does ppc64 require boards to call pci_devs_phb_init()? Kumar Gala
@ 2009-04-30 18:04 ` Kumar Gala
  2009-04-30 21:09   ` Benjamin Herrenschmidt
  2009-04-30 21:08 ` Benjamin Herrenschmidt
  1 sibling, 1 reply; 4+ messages in thread
From: Kumar Gala @ 2009-04-30 18:04 UTC (permalink / raw)
  To: Kumar Gala; +Cc: Linuxppc-dev Development


On Apr 30, 2009, at 10:37 AM, Kumar Gala wrote:

> looking at merging this in w/ppc32 and wondering why board code has  
> to call pci_devs_phb_init() and not do it via pcibios_init()
>
> - k

As I look at this a bit more, I'm thinking why don't we do:

pci_devs_phb_init_dynamic() in pcibios_alloc_controller

An add some code into pci_devs_phb_init_dynamic so we don't re-do  
update_dn_pci_info if dn->data is already set.

I am not sure what impact this has ppc64 systems that call  
pci_devs_phb_init()

- k

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: why does ppc64 require boards to call pci_devs_phb_init()?
  2009-04-30 15:37 why does ppc64 require boards to call pci_devs_phb_init()? Kumar Gala
  2009-04-30 18:04 ` Kumar Gala
@ 2009-04-30 21:08 ` Benjamin Herrenschmidt
  1 sibling, 0 replies; 4+ messages in thread
From: Benjamin Herrenschmidt @ 2009-04-30 21:08 UTC (permalink / raw)
  To: Kumar Gala; +Cc: Linuxppc-dev Development

On Thu, 2009-04-30 at 10:37 -0500, Kumar Gala wrote:
> looking at merging this in w/ppc32 and wondering why board code has to  
> call pci_devs_phb_init() and not do it via pcibios_init()

As I said on IRC, not too sure at the moment, it could be due to some
ordering issues where pSeries needs the pci_dn's to be there before
pcibios_init(). We'll have to dig a bit on that one.

Cheers,
Ben.

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: why does ppc64 require boards to call pci_devs_phb_init()?
  2009-04-30 18:04 ` Kumar Gala
@ 2009-04-30 21:09   ` Benjamin Herrenschmidt
  0 siblings, 0 replies; 4+ messages in thread
From: Benjamin Herrenschmidt @ 2009-04-30 21:09 UTC (permalink / raw)
  To: Kumar Gala; +Cc: Linuxppc-dev Development

On Thu, 2009-04-30 at 13:04 -0500, Kumar Gala wrote:
> On Apr 30, 2009, at 10:37 AM, Kumar Gala wrote:
> 
> > looking at merging this in w/ppc32 and wondering why board code has  
> > to call pci_devs_phb_init() and not do it via pcibios_init()
> >
> > - k
> 
> As I look at this a bit more, I'm thinking why don't we do:
> 
> pci_devs_phb_init_dynamic() in pcibios_alloc_controller
> 
> An add some code into pci_devs_phb_init_dynamic so we don't re-do  
> update_dn_pci_info if dn->data is already set.
> 
> I am not sure what impact this has ppc64 systems that call  
> pci_devs_phb_init()

That would probably work, remains the reconfig notifier thing but that
can probably be moved to pseries platform.

Cheers,
Ben.

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2009-04-30 21:09 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-04-30 15:37 why does ppc64 require boards to call pci_devs_phb_init()? Kumar Gala
2009-04-30 18:04 ` Kumar Gala
2009-04-30 21:09   ` Benjamin Herrenschmidt
2009-04-30 21:08 ` Benjamin Herrenschmidt

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).