--- Comment #8 from eedelman at gcc dot gnu dot org 2006-01-31 11:12
---
(In reply to comment #7)
> The original code looks like
>
>subroutine foo(self)
> character(*) :: self
> pointer :: self
> nullify(self)
> allocate(self)
> self = " "
>end subro
--- Comment #7 from hjl at lucon dot org 2006-01-31 01:52 ---
The original code looks like
subroutine foo(self)
character(*) :: self
pointer :: self
nullify(self)
allocate(self)
self = " "
end subroutine
Is this code valid? How can I make it valid?
-
--- Comment #6 from pinskia at gcc dot gnu dot org 2006-01-31 01:15 ---
Subject: Re: FORTRAN segfault
On Jan 30, 2006, at 7:45 PM, hjl at lucon dot org wrote:
> Intel FORTRAN compiler has no problem with it.
Intel's Fortran compiler does not detect a lot of
invalid code, that does no
On Jan 30, 2006, at 7:45 PM, hjl at lucon dot org wrote:
Intel FORTRAN compiler has no problem with it.
Intel's Fortran compiler does not detect a lot of
invalid code, that does not make this code valid.
-- Pinski
--- Comment #5 from pinskia at gcc dot gnu dot org 2006-01-31 00:59 ---
Subject: Re: FORTRAN segfault
>
>
>
> --- Comment #4 from hjl at lucon dot org 2006-01-31 00:45 ---
> This code is extracted from a much larger program. Intel FORTRAN compiler
> has no problem with it.
>
>
>
> --- Comment #4 from hjl at lucon dot org 2006-01-31 00:45 ---
> This code is extracted from a much larger program. Intel FORTRAN compiler
> has no problem with it.
And what should it allocate a zero sized string?
-- Pinski
--- Comment #4 from hjl at lucon dot org 2006-01-31 00:45 ---
This code is extracted from a much larger program. Intel FORTRAN compiler
has no problem with it.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26038
--- Comment #3 from pinskia at gcc dot gnu dot org 2006-01-30 23:59 ---
If I change the program to use a constant size string, it works.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #2 from pinskia at gcc dot gnu dot org 2006-01-30 23:48 ---
This code is meaning less as far as I can tell but it is accepted by Lahey's
compilers.
Oh don't try redhat's branch please it is not something that is just wrong to
do with a modifed compiler at least in bug report
--- Comment #1 from hjl at lucon dot org 2006-01-30 22:39 ---
It happens on gcc 4.2, 4.1 and 4.0. But gcc-4.1-redhat is fine:
[EMAIL PROTECTED] cpu2006]$ /usr/gcc-4.1-redhat/bin/gcc -S foo.f90 -O2
[EMAIL PROTECTED] cpu2006]$ /usr/gcc-4.1-redhat/bin/gcc --version
gcc (GCC) 4.1.0 20060128
10 matches
Mail list logo
