wmb vs mmiowb

[Jesse Barnes]

> > Yeah, they keep threatening to use this instead, but I'm not sure
> > how easy it would be.  Also they may have more devices/drivers to
> > worry about than sn2, so maybe changing over would mean too much
> > driver debugging (well auditing really since it's not that hard to
> > know where to put them).  Irix actually had an io_unlock() routine
> > that did this implicitly, but iirc that was shot down for Linux...
>
> Why was it shot down? Seems like a pretty good idea to me ;)

Well, like Linus said, it had some significant downsides (though I think 
Irix had fewer lock types, so the multiplicative effect wasn't so bad 
there).

> I'm clueless when it comes to drivers, but I see a lot of mmiowb()
> that are not paired with spin_unlock. How are these obvious? (ie.
> what is the pattern?) It looks like some might be lockless FIFOs (or
> maybe I'm just not aware of where the locks are). Can you just
> quickly illustrate the problem being solved?

Wow, it certainly has proliferated since it was added to the tree. :)

I didn't audit all the uses, but it seems like many of them get it 
right, i.e. mmiowb() before spin_unlock() where PIO has been done.  I'd 
have to look carefully to see whether lockless usages are correct, it's 
likely they're not.

Jesse