Re: [patch] the other __raw_writel's

B. James Phillippe (bryan@terran.org)
Fri, 20 Aug 1999 15:02:26 -0700 (PDT)


On Fri, 20 Aug 1999, Ralf Baechle wrote:

> On Fri, Aug 20, 1999 at 02:46:24PM -0700, B. James Phillippe wrote:
>
> > > On Thu, Aug 19, 1999 at 03:57:34PM -0700, B. James Phillippe wrote:
> > >
> > > > Isn't this because a "longword" is 32 bits? So the use of "q" for
> > > > "quadword" is appropriate for a 64 bit operation.
> > >
> > > How confusing, on MIPS q stands for quadword, 128 bit.
> >
> > That is confusing, and most unfortunate. Redfining the word (and thus
> > long and quad) is almost as broken as redefining the byte.
>
> But a byte is only a foggy defined entity of a few bits anyway. There
> were and are machines with != 8 bits per byte.

What matters is whether Linux runs on those machines. I think of the
byte/word issue much the same way I think of "unsigned long" in Linux; it's
used interchangeably with "void *" because it makes sense for all supported
(and likely supported) architectures. Of course, using the number itself
in the name doesn't hurt: write64, read16, etc. I'd take credit for the
suggestion if it hadn't already been posted. ;-)

-bp

--
# bryan at terran dot org
# http://www.terran.org/~bryan

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/