------- Comment #5 from ray at ultramarine dot com 2007-02-06 17:00 -------
Subject: Re: Internal Complier Error
On Tue, 6 Feb 2007, burnus at gcc dot gnu dot org wrote:
>
>
> ------- Comment #4 from burnus at gcc dot gnu dot org 2007-02-06 16:34
> -------
> Not that it much helps, but with today's gcc under x86_64-unknown-linux-gnu it
> does not crash.
> Maybe someone with Intel Mac can reproduce it.
>
> Oherwise:
> Compile with the "-v" option, e.g. gfortran -v -O3 env_grid.f
> After the line "gcc-Version 4.3.0 20070206 (experimental)" is the actual
> compilation command. On my computer it looks like:
> /projects/tob/gcc-trunk/libexec/gcc/x86_64-unknown-linux-gnu/4.3.0/f951
> env_grid.f -ffixed-form ...
>
> Run now (replace the argument with yours of above)
> gdb /projects/tob/gcc-trunk/libexec/gcc/x86_64-unknown-linux-gnu/4.3.0/f951
> (gdb) b real_abort
> (gdb) b fancy_abort
> (gdb) run env_grid.f -ffixed-form (... and the rest of the options)
>
> When it stops, enter
> (gdb) bt
>
> This gives a backtrace and a hint where the error occurs. What exact version
> of
> gfortran did you try? Do you know by chance whether this is a new problem or
> was present in a previous version of gfortran 4.3?
If you will look in the "log" file it contains the results of
both uname -a and gfortran -v. It says
gcc version 4.3.0 20070104 (experimental)
I tried the above, but did not get any trace. Message said no stack.
I have no idea if this is an old or a new problem. I am mostly using
4.1 on solaris, linux, and mac (power pc). This is the first time
I have seen the problem.
If it helps, I got around the problem by making the code from line
105 to 117 into a subroutine. Also, the problem does not exist without
-O3.
Sorry I could not be of more help,
Ray
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30713
