Re: [RFC PATCH] docs/memory-barriers.txt: Rewrite "KERNEL I/O BARRIER EFFECTS" section

From: Benjamin Herrenschmidt
Date: Mon Feb 11 2019 - 23:02:42 EST

On Mon, 2019-02-11 at 14:34 -0800, Linus Torvalds wrote:
> On Mon, Feb 11, 2019 at 9:30 AM Will Deacon <will.deacon@xxxxxxx> wrote:
> > +
> > + 1. All readX() and writeX() accesses to the same peripheral are ordered
> > + with respect to each other. For example, this ensures that MMIO register
> > + writes by the CPU to a particular device will arrive in program order.
> Hmm. I'd like more people look at strengthening this one wrt across
> CPUs and locking.
> Right now we document mmiowb(), but that "documentation" is really
> just a fairy tale. Very *very* few drivers actually do mmiowb() on
> their own.
> IOW, we should seriously just consider making the rule be that locking
> will order mmio too. Because that's practically the rule anyway.
> Powerpc already does it. IO within a locked region will serialize with the lock.

Yup. It's a bit ugly but I felt back then that getting drivers to use
mmiowb() properly was going to be a losing battle.