The trap handler returns different information after this date for
numeric error conditions. Since kern.osreldate is not a reliable
indicator of when this change occurred, we revert to checking for
the presence of a particular #define in <machine/trap.h> to decide
if the SML/NJ runtime sources need to use the newer interface.
Submitted by: Tom <tcrimi@andrew.cmu.edu> (in part)
FWIW, checkout of these things took 5+hrs, staying on the local
.freebsd.org net w/o hitting the 'net at all.
As promised,
$ time cvs ci
real 67m51.701s
user 0m1.250s
sys 0m5.345s