> After some thinking I realised that the reliance on signal delivery to 
> work properly to trap non-executable stack may actually be a problem for 
> bare-iron targets.

IMO, I think we can start worrying about those when we actually
encounter the problem; and I am assuming that this is not going to
be specific to mips.

Are we still good to go with this patch? Attached is the latest
version. For any additional comments that you'd like to be added
(in particular, with respect to what you just pointed out), I suggest
add them as a followup patch.  This way, this one is out of the way,
and we can just focus on the comments themselves.

I am also adding a second patch, which shows the changes I made
in this iteration.

Tested on mips-irix, no regression.

Thanks,
-- 
Joel