internal compiler error with gfortran 4.4-20081107

[kamaraju kusumanchi]

On Wed, Jan 14, 2009 at 5:25 PM, Tobias Schlüter
<tobias.schlueter@physik.uni-muenchen.de> wrote:
> Steve Kargl wrote:
>>
>> On Wed, Jan 14, 2009 at 05:17:38PM -0500, kamaraju kusumanchi wrote:
>>>>
>>>> You could try running f951 in a debugger to see where it dies.
>>>>
>>> What f951? I do not have anything named like that (not even f95, f95l)...
>>
>> It's the compiler!  You have it.
>>
>> find /usr/local -name f951
>>
>> Use an appropriate path for /usr/local if gfortran is
>> installed in a location other than the /usr/local
>> hierarchy.
>>
> And use gfortran -v in your compilation to see how it's invoked.
>

The compiler is in ~/software/myroot/gcc-4.4-20081107/bin/gfortran . I
found out that there is a f95l in
~/software/myroot/gcc-4.4-20081107/libexec/gcc/sparc-sun-solaris2.8/4.4.0/f951
.

I try to load it up in dbx. That also gives an internal error.

> dbx ~/software/myroot/gcc-4.4-20081107/libexec/gcc/sparc-sun-solaris2.8/4.4.0/f951
For information about new features see `help changes'
To remove this message, put `dbxenv suppress_startup_message 7.1' in your .dbxrc
Reading f951
Reading ld.so.1
Reading libmpfr.so.1
Reading libgmp.so.3
Reading libc.so.1
Reading libdl.so.1
Reading libc_psr.so.1

dbx: internal error: signal SIGSEGV (no mapping at the fault address)
dbx's coredump will appear in /tmp
Abort


It could that f95l is not compiled with debugging symbols in it. Is
that the case?