segfault within CPU_TIME (was: libgfortran/intrinsics/cpu_time.c: No such file or directory.)

François-Xavier Coudert fxcoudert@gmail.com
Wed Aug 22 09:30:00 GMT 2007


> Is this a problem of the cpu_time intrinsic or does the error actually
> already occur before?

This is known as PR30780
(http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30780): the library
cpu_time function itself features a conversion with loss of precision
on standard targets. As Anrew said in the PR, it's hard to fix because
"there are no functions which convert from a floating point to an
integer that might not raise a signal.  This is why setting trapping
on
precision is almost something you never want to do."

Moreover, Jerry indicated in the PR that some of the I/O code might be
affected similarly.

So, to play it safe, I think it's best not to use -ffpe-trap=invalid
for now (-ffpe-trap=zero and others should be fine).

FX



More information about the Fortran mailing list