Benjamin Herrenschmidt <[EMAIL PROTECTED]> wrote: > On Fri, 2008-04-04 at 15:42 +0900, Ishizaki Kou wrote: > > > > As you pointed, spider I/O functions in Cell blades need 2 step > > indirections by our patch. Shall I make another one for Cell blades > > whose spider I/O functions need one step indirection? > > (But you will need 2 step indirections when you use PCI-ex.) > > I think the blades will need the same stuff as celleb since it's > possible to use the PCI-Express on them too. > > Maybe an option is to do a if () / else statement rather than a function > pointer in there, it would at least be cheaper in term of CPU cycle > don't you think ? Anyway, do as you prefer.
I want to keep my patch as it is, because I think it is easier to add function pointers than if () / else statement when someone adds new I/O functions to the io-workaround. For 2.6.27 or later, I think about adding null I/O functions for Celleb internal pci bus to reduce overhead for searching bus. Best regards, Kou Ishizaki _______________________________________________ Linuxppc-dev mailing list Linuxppc-dev@ozlabs.org https://ozlabs.org/mailman/listinfo/linuxppc-dev
