[moved to tech-userlevel because it doesn't really have anything to do
with the kernel any longer]

>> [...duplicate header inclusion...]

> The problem with removing "duplicate" includes is that you make
> include files not-self-sufficient.

Yes, except that NetBSD does not consider that a problem.  (Some time
ago I tried to offer fixes for such and was told that; more recently
someone else reported similar experiences.)

> You then end up with an irritating dance of "add the needed include"
> when making a new file.

Yes - and worse, "get the includes in the right order"; it's not enough
to include the right files, you have to get the order right too.

> I'd be much more interested in a tool which can figure out if we use
> nothing out of an include file. ;-)

Yes, me too.  I've occasionally considered trying to make gcc do that,
but so far I anticipate it being intimidating enough that I haven't
actually tried.

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse@rodents.montreal.qc.ca
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B