[Bug fortran/51310] New: -finit-bla doesn't initialize *all* items of type bla to the requested constant.

[Bug fortran/51310] New: -finit-bla doesn't initialize *all* items of type bla to the requested constant.

[toon at moene dot org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51310

             Bug #: 51310
           Summary: -finit-bla doesn't initialize *all* items of type bla
                    to the requested constant.
    Classification: Unclassified
           Product: gcc
           Version: 4.6.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: fortran
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: toon@moene.org


The following source:

program nan
implicit none
real(8) :: joo(3)
joo(2) = 0
print *,joo
call sub(3)
end program nan
subroutine sub(n)
implicit none
integer, intent(in) :: n
real(8) :: a(n), var, b(3)
real(8), allocatable :: c(:)
!a(1) = 1
print *,'n=',n
print *,'a=',a
print *,'var=',var
print *,'b=',b
allocate(c(size(b)))
print *,'c=',c
a(1) = a(1) + 1
end subroutine sub

when compiled with gfortran -g -finit-real=snan outputs:

                       NaN   0.0000000000000000                            NaN
 n=           3
 a=   0.0000000000000000        0.0000000000000000        0.0000000000000000    
 var=                       NaN
 b=                       NaN                       NaN                      
NaN
 c=   0.0000000000000000        0.0000000000000000        0.0000000000000000    

Clearly, the allocatable c and the automatic array a are not initialized by the
compiler option.

1. Consider this an enhancement request - a colleague of mine was severely
   hampered in his search for a bug.

2. Consider this a bug report with respect to our documentation, which doesn't
   mention this exceptions at all.

[Bug fortran/51310] -finit-bla doesn't initialize *all* items of type bla to the requested constant.

[toon at moene dot org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51310

--- Comment #1 from Toon Moene <toon at moene dot org> 2011-11-29 19:37:19 UTC ---
Created attachment 25948
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=25948
Untested patch to the documentation

This is a completely untested patch to the documentation, as I cannot build gcc
trunk on Debian testing since August.

Perhaps someone else can review it as to:

- Correct texi syntax.
- Correct contents :-)

Thanks,

[Bug fortran/51310] -finit-bla doesn't initialize *all* items of type bla to the requested constant.

[toon at moene dot org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51310

Toon Moene <toon at moene dot org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
  Attachment #25948|0                           |1
        is obsolete|                            |
             Status|UNCONFIRMED                 |ASSIGNED
   Last reconfirmed|                            |2011-12-05
         AssignedTo|unassigned at gcc dot       |toon at moene dot org
                   |gnu.org                     |
     Ever Confirmed|0                           |1

--- Comment #2 from Toon Moene <toon at moene dot org> 2011-12-05 18:17:14 UTC ---
Created attachment 25996
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=25996
A patch to the documentation that actually passes a build

I'll submit this patch to fortran@gcc.gnu.org for approval.

After that's done, and the patch is committed, I'll leave the bug report open
to see if I can fix some of the shortcomings.

[Bug fortran/51310] -finit-bla doesn't initialize *all* items of type bla to the requested constant.

[toon at moene dot org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51310

--- Comment #3 from Toon Moene <toon at moene dot org> 2011-12-05 20:04:21 UTC ---
At first I thought that gfortran would initialize small local arrays to
whatever -finit-real indicated by making them static, instead of stack based.

However, perusing the assembler output of

      subroutine sub
      real a(3)
      print*,a
      end

(as a diff between gfortran -S and gfortran -S -finit-real=snan):

17c17,27
<     movq    $.LC0, -536(%rbp)
---
> 	movl	$1, %eax
> .L3:
> 	cmpq	$3, %rax
> 	jg	.L2
> 	leaq	-1(%rax), %rdx
> 	movl	$0x7fa00000, %ecx
> 	movl	%ecx, -16(%rbp,%rdx,4)
> 	addq	$1, %rax
> 	jmp	.L3
> .L2:

and checking with a small main program

      call sub
      end

that the output is really different:

$ ./a.out
  1.12103877E-44   0.0000000      9.80908925E-45

vs:

$ ./a.out
             NaN             NaN             NaN

convinced me otherwise.

So one wonders why that doesn't work with variable length automatic arrays (or
why it would even be hard to initialize malloc'd automatic arrays /
allocatables).

This quest to be continued ("do you think coconuts migrate ?  No - but they
could be carried.").

[Bug fortran/51310] -finit-bla doesn't initialize *all* items of type bla to the requested constant.

[toon at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51310

--- Comment #4 from toon at gcc dot gnu.org 2011-12-08 18:38:39 UTC ---
Author: toon
Date: Thu Dec  8 18:38:32 2011
New Revision: 182127

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=182127
Log:
2011-12-08  Toon Moene  <toon@moene.org>

    PR fortran/51310
    * invoke.texi: Itemize the cases for which
    -finit-<type>=<constant> doesn't work.


Modified:
    trunk/gcc/fortran/ChangeLog
    trunk/gcc/fortran/invoke.texi

[Bug fortran/51310] -finit-bla doesn't initialize *all* items of type bla to the requested constant.

[toon at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51310

--- Comment #5 from toon at gcc dot gnu.org 2011-12-08 20:58:31 UTC ---
Author: toon
Date: Thu Dec  8 20:58:23 2011
New Revision: 182138

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=182138
Log:
2011-12-08  Toon Moene  <toon@moene.org>

    PR fortran/51310
    * invoke.texi: Itemize the cases for which
    -finit-<type>=<constant> doesn't work.


Modified:
    branches/gcc-4_6-branch/gcc/fortran/ChangeLog
    branches/gcc-4_6-branch/gcc/fortran/invoke.texi

[Bug fortran/51310] -finit-bla doesn't initialize *all* items of type bla to the requested constant.

[toon at moene dot org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51310

Toon Moene <toon at moene dot org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|normal                      |enhancement

--- Comment #6 from Toon Moene <toon at moene dot org> 2011-12-08 21:01:06 UTC ---
The bugs in the documentation have been fixed, so now it is an enhancement to a
set of debugging compiler flags.

[Bug fortran/51310] -finit-bla doesn't initialize *all* items of type bla to the requested constant.

[toon at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51310

--- Comment #7 from toon at gcc dot gnu.org 2011-12-15 18:26:08 UTC ---
Author: toon
Date: Thu Dec 15 18:26:02 2011
New Revision: 182384

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=182384
Log:
2011-12-15  Toon Moene  <toon@moene.org>

    PR fortran/51310
    * resolve.c (build_default_init_expr): Allow non-allocatable,
    non-compile-time-constant-shape arrays to have a default
    initializer.
    * invoke.texi: Delete the restriction on automatic arrays not


Modified:
    trunk/gcc/fortran/ChangeLog
    trunk/gcc/fortran/invoke.texi
    trunk/gcc/fortran/resolve.c

[Bug fortran/51310] -finit-bla doesn't initialize *all* items of type bla to the requested constant.

[toon at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51310

--- Comment #8 from toon at gcc dot gnu.org 2011-12-22 19:23:31 UTC ---
Author: toon
Date: Thu Dec 22 19:23:26 2011
New Revision: 182634

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=182634
Log:
2011-12-22  Toon Moene  <toon@moene.org>

    PR fortran/51310
    * resolve.c (build_default_init_expr): Allow non-allocatable,
    non-compile-time-constant-shape arrays to have a default
    initializer.
    * invoke.texi: Delete the restriction on automatic arrays not
    being initialized by -finit-<type>=<constant>.


Modified:
    branches/gcc-4_6-branch/gcc/fortran/ChangeLog
    branches/gcc-4_6-branch/gcc/fortran/invoke.texi
    branches/gcc-4_6-branch/gcc/fortran/resolve.c