[Bug fortran/38602] [regression 4.4] segfault - optional arguments, generics, interface problem, iso_varying_string ??
-- rguenth at gcc dot gnu dot org changed: What|Removed |Added Target Milestone|4.3.4 |4.3.3 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38602
[Bug fortran/38602] [regression 4.4] segfault - optional arguments, generics, interface problem, iso_varying_string ??
--- Comment #17 from pault at gcc dot gnu dot org 2008-12-23 09:26 --- (In reply to comment #12) (In reply to comment #11) Patch committed, fixed on trunk, 4.4 I think it was the wrong PR number (35780 instead of PR34820). Dang it! Sorry about that, I used the ChangeLogs from 35780 as a template and forgot to update the numbers. Thanks for helping out. Paul -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38602
[Bug fortran/38602] [regression 4.4] segfault - optional arguments, generics, interface problem, iso_varying_string ??
--- Comment #5 from mikael at gcc dot gnu dot org 2008-12-22 18:51 --- Reduced: program test_iso implicit none type varying_string character(LEN=1), dimension(:), allocatable :: chars end type varying_string CALL get_unit_set_ch () contains subroutine get_unit_set_CH (separator) type(varying_string), intent(out), optional :: separator end subroutine get_unit_set_CH end -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38602
[Bug fortran/38602] [regression 4.4] segfault - optional arguments, generics, interface problem, iso_varying_string ??
--- Comment #6 from mikael at gcc dot gnu dot org 2008-12-22 18:54 ---
(In reply to comment #5)
Generated code:
get_unit_set_ch (struct varying_string * separator)
{
if (separator-chars.data != 0B)
{
__builtin_free (separator-chars.data);
}
separator-chars.data = 0B;
}
test_iso ()
{
static integer(kind=4) options.0[8] = {68, 255, 0, 0, 0, 1, 0, 1};
static void get_unit_set_ch (struct varying_string *);
_gfortran_set_options (8, (void *) options.0);
get_unit_set_ch (0B);
}
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38602
[Bug fortran/38602] [regression 4.4] segfault - optional arguments, generics, interface problem, iso_varying_string ??
--- Comment #7 from mikael at gcc dot gnu dot org 2008-12-22 18:59 --- This was probably a bad idea: - tmp = gfc_deallocate_alloc_comp (e-ts.derived, tmp, parm_rank); - if (e-expr_type == EXPR_VARIABLE e-symtree-n.sym-attr.optional) - tmp = build3_v (COND_EXPR, gfc_conv_expr_present (e-symtree-n.sym), - tmp, build_empty_stmt ()); -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38602
[Bug fortran/38602] [regression 4.4] segfault - optional arguments, generics, interface problem, iso_varying_string ??
--- Comment #8 from pault at gcc dot gnu dot org 2008-12-22 19:39 ---
Created an attachment (id=16969)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16969action=view)
patch for the PR
This is a patch that I cannot commit until January 3rd at the earliest. Please
could you do the honours, Jerry? It runs all of alloc_comp*.f90 correctly but
I have NOT regtested it.
Paul
2008-12-22 Paul Thomas pa...@gcc.gnu.org
PR fortran/35780
* trans-decl.c (init_intent_out_dt): Allow for optional args.
2008-12-22 Paul Thomas pa...@gcc.gnu.org
PR fortran/35780
* gfortran.dg/alloc_comp_optional_1.f90: New test.
! { dg-do run }
! Tests the fix for PR38602, a regression caused by a modification
! to the nulling of INTENT_OUT dummies with allocatable components
! that caused a segfault with optional arguments.
!
! Contributed by David Kinniburgh davidkinnibu...@yahoo.co.uk
!
program test_iso
type ivs
character(LEN=1), dimension(:), allocatable :: chars
end type ivs
type(ivs) :: v_str
integer :: i
call foo(v_str, i)
if (v_str%chars(1) .ne. a) call abort
if (i .ne. 0) call abort
call foo(flag = i)
if (i .ne. 1) call abort
contains
subroutine foo (arg, flag)
type(ivs), optional, intent(out) :: arg
integer :: flag
if (present(arg)) then
arg = ivs([(char(i+96), i = 1,10)])
flag = 0
else
flag = 1
end if
end subroutine
end
--
pault at gcc dot gnu dot org changed:
What|Removed |Added
AssignedTo|unassigned at gcc dot gnu |pault at gcc dot gnu dot org
|dot org |
Status|NEW |ASSIGNED
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38602
[Bug fortran/38602] [regression 4.4] segfault - optional arguments, generics, interface problem, iso_varying_string ??
--- Comment #9 from pault at gcc dot gnu dot org 2008-12-22 20:03 --- Created an attachment (id=16970) -- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16970action=view) The real patch for the PR without the extra bits:-) Have eliminated the extra bits on the patch - sorry about that! Paul -- pault at gcc dot gnu dot org changed: What|Removed |Added Attachment #16969|0 |1 is obsolete|| http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38602
[Bug fortran/38602] [regression 4.4] segfault - optional arguments, generics, interface problem, iso_varying_string ??
--- Comment #10 from jvdelisle at gcc dot gnu dot org 2008-12-22 20:53 --- Regression testing now and will commit when complete. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38602
[Bug fortran/38602] [regression 4.4] segfault - optional arguments, generics, interface problem, iso_varying_string ??
--- Comment #11 from jvdelisle at gcc dot gnu dot org 2008-12-22 23:39 --- Patch committed, fixed on trunk, 4.4 -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38602
[Bug fortran/38602] [regression 4.4] segfault - optional arguments, generics, interface problem, iso_varying_string ??
--- Comment #12 from mikael at gcc dot gnu dot org 2008-12-23 00:18 --- (In reply to comment #11) Patch committed, fixed on trunk, 4.4 I think it was the wrong PR number (35780 instead of PR34820). -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38602
[Bug fortran/38602] [regression 4.4] segfault - optional arguments, generics, interface problem, iso_varying_string ??
--- Comment #13 from jvdelisle at verizon dot net 2008-12-23 01:11 --- Subject: Re: [regression 4.4] segfault - optional arguments, generics, interface problem, iso_varying_string ?? mikael at gcc dot gnu dot org wrote: --- Comment #12 from mikael at gcc dot gnu dot org 2008-12-23 00:18 --- (In reply to comment #11) Patch committed, fixed on trunk, 4.4 I think it was the wrong PR number (35780 instead of PR34820). Yes, I will fix the log. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38602
[Bug fortran/38602] [regression 4.4] segfault - optional arguments, generics, interface problem, iso_varying_string ??
--- Comment #14 from jvdelisle at gcc dot gnu dot org 2008-12-23 01:33 --- Subject: Bug 38602 Author: jvdelisle Date: Tue Dec 23 01:32:33 2008 New Revision: 142898 URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=142898 Log: 2008-12-22 Paul Thomas pa...@gcc.gnu.org PR fortran/38602 * trans-decl.c (init_intent_out_dt): Allow for optional args. Modified: branches/gcc-4_3-branch/gcc/fortran/ChangeLog branches/gcc-4_3-branch/gcc/fortran/trans-decl.c -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38602
[Bug fortran/38602] [regression 4.4] segfault - optional arguments, generics, interface problem, iso_varying_string ??
--- Comment #15 from jvdelisle at gcc dot gnu dot org 2008-12-23 01:35 --- Subject: Bug 38602 Author: jvdelisle Date: Tue Dec 23 01:34:13 2008 New Revision: 142899 URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=142899 Log: 2008-12-22 Paul Thomas pa...@gcc.gnu.org PR fortran/38602 * gfortran.dg/alloc_comp_optional_1.f90: New test. Added: branches/gcc-4_3-branch/gcc/testsuite/gfortran.dg/alloc_comp_optional_1.f90 Modified: branches/gcc-4_3-branch/gcc/testsuite/ChangeLog -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38602
[Bug fortran/38602] [regression 4.4] segfault - optional arguments, generics, interface problem, iso_varying_string ??
--- Comment #16 from jvdelisle at gcc dot gnu dot org 2008-12-23 02:10 --- Fixed on 4.3, closing. Thanks everyone for the help with this. -- jvdelisle at gcc dot gnu dot org changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution||FIXED Target Milestone|--- |4.3.4 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38602
[Bug fortran/38602] [regression 4.4] segfault - optional arguments, generics, interface problem, iso_varying_string ??
--- Comment #2 from kargl at gcc dot gnu dot org 2008-12-22 00:28 --- Please attach long files instead of embedded the code within the bug report. It makes it harder to copy and paste from a browser. The code works with 4.3.3, so this is a regression. Here's a reduced testcase that needs to be linked to iso_varying_string. Save this as lkl.f90 and compile. program test_iso use iso_varying_string implicit none type(varying_string) :: v_str integer :: i, ios character(len=1) :: newlinechar=; open (unit=1,file='lkl.f90', status='OLD') i = 0 do CALL get(unit=1,string=v_str, set=newlinechar, iostat=ios) if (ios.GT.0) stop 'Error on vs2 read' if (ios.EQ.-1) exit i = i + 1 print *, i, trim(char(v_str)) end do end -- kargl at gcc dot gnu dot org changed: What|Removed |Added Status|UNCONFIRMED |NEW Ever Confirmed|0 |1 Priority|P3 |P4 Last reconfirmed|-00-00 00:00:00 |2008-12-22 00:28:42 date|| Summary|run time error (crash) -|[regression 4.4] segfault - |optional arguments, |optional arguments, |generics, interface problem,|generics, interface problem, |iso_varying_string ?? |iso_varying_string ?? http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38602
[Bug fortran/38602] [regression 4.4] segfault - optional arguments, generics, interface problem, iso_varying_string ??
--- Comment #3 from kargl at gcc dot gnu dot org 2008-12-22 00:30 --- Created an attachment (id=16956) -- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=16956action=view) iso_varying_string code. Here's the iso_varying_string.f90 code as an attachment. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38602
[Bug fortran/38602] [regression 4.4] segfault - optional arguments, generics, interface problem, iso_varying_string ??
--- Comment #4 from jvdelisle at gcc dot gnu dot org 2008-12-22 05:18 --- The test case passes at r142147 and fails at r142148 The patch to pr34820 caused this regression PR fortran/34820 * trans-expr.c (gfc_conv_function_call): Remove all code to deallocate intent out derived types with allocatable components. (gfc_trans_assignment_1): An assignment from a scalar to an array of derived types with allocatable components, requires a deep copy to each array element and deallocation of the converted rhs expression afterwards. * trans-array.c : Minor whitespace. * trans-decl.c (init_intent_out_dt): Add code to deallocate allocatable components of derived types with intent out. (generate_local_decl): If these types are unused, set them referenced anyway but allow the uninitialized warning. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38602
