On Mon, 2015-07-13 at 16:53 +0100, Andrew Cooper wrote: > I could see about doing something based on BITS_PER_LONG which doesn't > have a visible ifdef, but is still ifdefary.
Given that it is the reason for the width differences in the old stream using sizeof(unsigned long) wouldn't seem massively inappropriate, indeed. I'm not sure it's especially necessary to launder that through bits_per_long though. Ian. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
