Re: 2.6.8.1-mm3

[Nick Piggin <nickpiggin@yahoo.com.au>]

Jesse Barnes wrote:
> On Saturday, August 21, 2004 1:05 pm, Jesse Barnes wrote:
> 
>>On Friday, August 20, 2004 9:26 pm, Nick Piggin wrote:
>>
>>>Jesse Barnes wrote:
>>>
>>>>On Friday, August 20, 2004 6:19 am, Andrew Morton wrote:
>>>>
>>>>>- This is (very) lightly tested.  Mainly a resync with various parties.
>>>>
>>>>Woo-hoo!  This boots *without changes* on a 512p Altix!  Now to re-run
>>>>the profiles and try wli's new per-cpu profiling buffers.
>>>
>>>What changes were needed to achieve this previously?
>>
>>The arch specific SD_NODE_INIT was the missing piece from previous -mm
>>releases.  Now that it's there, things seem to work.  I still have to poke
>>around to see if there are places where we're trying to scan across all
>>CPUs looking for busy ones and *then* checking to see if they're in a
>>domain. John said he'd take a closer look.
> 
> 
> It looks like the group span code will still try to look at everything in the 
> system.  That'll also have to be fixed to use the span from the associated 
> node's domain, since it looks like there are still places where we'll walk 
> all the CPUs in the system if it's not.  I'll take a closer look and see if I 
> can come up with something useful.
> 

I don't see it - what function are you talking about?