Antti Kantee <pooka@cs.hut.fi> writes:

> On Sun Jun 05 2005 at 16:28:32 -0400, Nathan J. Williams wrote:
> > Antti Kantee <pooka@cs.hut.fi> writes:
> > 
> > > Is that patch ok to commit, or was there a reason for doing two writes
> > > in the first place?
> > 
> > I did something similar for an evb with a 16-bit bus (plus a 32-bit
> > kluge) to one of these chips. I had trouble with byte-wite interrupt
> > registers, too. Try this:
> 
> You probably also have some header-patches you forgot to include.
> This didn't compile.

Oh, yeah, probably some flags for the 32-bit stuff. 

> But, since I don't have any troubles in using the chip with my arguably
> simpler patch, I'm just gonna commit it.

Sure. I was mostly hoping that my patch might point out some more
areas where 1-byte writes can profitably be removed.

        - Nathan