Tom Lane writes:
> Not sure; apparently it depends on optimization level, so I'd be
> hesitant to assume that any short test case would reliably expose
> the problem. Also, aren't you trying to avoid run-time tests in
> configure?
If the test doesn't use any library function's run-time behavior, you can
usually do something like
main() {
int a[(2.0+2.0==4.0)?1:-1]
}
This will fail to compile if the floating-point arithmetic is broken.
Otherwise a good solution might be to print a warning if configure detects
the flag. Or we can strip it out unconditionally, but that seems wrong.
The user should be made aware of the problem.
--
Peter Eisentraut peter_e@gmx.net
