Alan Cox wrote:
> I pointed out before the null check was flawed. And all I see is the
> same identical patch churned out again. Regardless of whether that
> paticular stupid error was in the old code, not fixing it in the new
> code when its pointed out is a bit of a mess.
The NULL check has been removed.  The current version is attatched.  It now 
pulls out 75 lines of code.  I must have missed your complaints about the null 
check in previous emails...  The check was done in the old pci_conf1 funcs, and 
  not in the pci_conf2 ones.. I thought this was strange as well, but I just 
left them in...

> I'm not sure its a simplification either. More function pointers don't
> always make for neater - but thats a side issue. If the NULL check goes
> I'm not too worried about the other stuff.
I think that it is definitely a simplification, although I am a bit biased ;) 
It makes it easier for other configuration types to hook into the system as 
well (I'm partial to NUMA-Q as well ;).  All they have to do is hijack the 
pci_config_(read|write) function pointers.

Also, has anyone had a chance to look at it with pci_conf2?  I have no 
particular reason to believe it doesn't work, as the code paths are almost 
identical, I'd just be nice to have a confirmation.

Cheers!

-Matt