Thus I can't build the new savecore, thus I cannot look at the crash 
dumps I get everytime I leave the latest built kernel running for more than
a minute or two, with

fatal integer divide fault in supervisor mode
trap type 8 code 0 eip f811301b cs 8 eflags 1028
2 cr2 1fbac cpl e000cefa
panic: trap

Actually it seems to be triggered by network activity into or out of the
machine, but i could be wrong. It usually seems to do this within a few
seconds of my telnetting into the machine. However the machine is being
a ppp gateway and that doesn't seem to crash it. *shrug*

It does make current totally unusable though.