[kaffe] Build failure without 'mincore' nor 'madvise'

[Kiyo Inaba]

>Thanks for the patch, Kiyo. Since the code comes from GNU Classpath,
>I'd like to see it fixed upstream,

Thanks, I can wait for this happen on Classpath.
# BTW, Tim Bevan's patch with void cast worked for m68k/linux.

>as well. I've read
>http://citeseer.ist.psu.edu/928.html
>a while ago, and it advocates not using #ifdefs. I'm not sure what the
>most elegant
>solution is in this case, but it seems that all the 'replacement'
>functionality in that
>file follows a common pattern.

In general, I agree with the opinion on the original paper, especially
when I am a idealist (sometimes it is called as a 'Manager' ;->). The
replacement (or lack of) functionality shall be treated not to use
ifdef and we can keep entire source code 'syntactically' safe.

Only one exception I can imagine for using ifdef is to support compiler
(or its version) specific feature.

But from pragmatist's point of view, I don't hesitate to use ifdef
to support multiple platforms, because 'it is the quickest (and dirtiest)
solution' :-) This is similar for very old story about 'is goto harmful?'.

Kiyo