Patch to allow targets to prevent inlining

[Nick Clifton]

Hi Doug,

: You're working too hard Nick.  This goes to gcc-patches (or
: whatever).

Yeah - Ian noticed that as well.  Sorry guys!  I have resubmitted the
patch to the appropriate list.

: But, while we're talkin' :-),
: How about making `naked' a standard attribute?
: Then you could do away with this because you could have gcc
: proper not allow inlining of naked fns.

Well it would be nice to have it as a standard attribute, but at the
moment only two back ends that I know of actually use it (ARM and
MCore).  Besides, just because it wouldn't be needed for naked
functions any more, there is no reason to suppose that individual
targets might not have other reasons for suppressing inlining.

Cheers
	Nick