Berkeley client + floating point exception

[fadden@uts.amdahl.com (Andy McFadden)]

I get a floating point exception every once in a while with the Berkeley
client.  From previous experience, I have a feeling this is because our
MAXDOUBLE is like 7e+75 instead of 1e+308 (64 bit vs 128 bit; it causes
ghostscript to bail on certain kinds of files).  My best guess so far is
the lag-o-meter, since it only dies when the meter is up and I hit a
particularly nasty chunk o' lag.

Anybody else seeing this?  Probably not, since it appears most workstations
these days have the larger IEEE floating point.

-- 
fadden@uts.amdahl.com (Andy McFadden)
[ Above opinions are mine, Amdahl has nothing to do with them, etc, etc. ]