> OK the variable alluded to above would have been better name
> fp_multiply_broken, but in order to be both terse and explicit how about
> fp_snm_broken? As in this snippet that could fit into Configure after the
> value of $i_math is determined (the eval and rm lines have been #commented
> out so that I could play with this script standalone, the echo >&4 line
> is commented too since I have not done the approriate exec in my login
> shell):
I feel that this is too specific an error with too specific parameters
to warrant a separate Configure test. How about a BEGIN test in the
library files?
--
$jhi++; # http://www.iki.fi/jhi/
# There is this special biologist word we use for 'stable'.
# It is 'dead'. -- Jack Cohen
