j**********t
发帖数: 12 | 1 Hi All,
Thanks for your attention.
My fortran code produced some NaN in the output file, which probably means
some invalid operations incurred, such as 0.0/0.0, I wonder if anybody can
tell me how to detect the source(or location) of the problem? I remember if
the code is compiled by Lahey, we can do it by adding some option when
compile(but forgot what kind of option). Unfortulately, I am using ifort now
. Any information will be highly appreciated.thanks | s**i
发帖数: 381 | 2 man ifort
and look for floating point exceptions (FPE)
if
now
【在 j**********t 的大作中提到】
: Hi All,
: Thanks for your attention.
: My fortran code produced some NaN in the output file, which probably means
: some invalid operations incurred, such as 0.0/0.0, I wonder if anybody can
: tell me how to detect the source(or location) of the problem? I remember if
: the code is compiled by Lahey, we can do it by adding some option when
: compile(but forgot what kind of option). Unfortulately, I am using ifort now
: . Any information will be highly appreciated.thanks
| j**********t
发帖数: 12 | 3 Thanks a lot. The problem was solved by following your suggestion.
【在 s**i 的大作中提到】
: man ifort
: and look for floating point exceptions (FPE)
:
: if
: now
| O******e
发帖数: 734 | 4 ifort also has an intrinsic function isnan() to test for NaN, returning a
logical result.
if
now
【在 j**********t 的大作中提到】
: Hi All,
: Thanks for your attention.
: My fortran code produced some NaN in the output file, which probably means
: some invalid operations incurred, such as 0.0/0.0, I wonder if anybody can
: tell me how to detect the source(or location) of the problem? I remember if
: the code is compiled by Lahey, we can do it by adding some option when
: compile(but forgot what kind of option). Unfortulately, I am using ifort now
: . Any information will be highly appreciated.thanks
|
|
