Re: memory-mapped i/o barrier

[Jesse Barnes]

On Mon, Jan 14, 2002 at 05:24:54PM +1100, Anton Blanchard wrote:
> Can loads/stores also complete out of order to IO? (the example just shows
> a store from one cpu passing one from another cpu)

I'm not sure what you mean, do you have an example?

> On ppc32/ppc64 this can happen, it is fixed up in the low level pci
> routines. Is there a case where you cant wrap it up in the low level
> routines like ppc32/ppc64?

You mean in each outX routine you essentially do an mmiob()?

Jesse