ICE
Salvatore Filippone
salvatore.filippone@uniroma2.it
Tue Sep 5 15:49:00 GMT 2006
Quoting Paul Thomas <paulthomas2@wanadoo.fr>:
>
> HJ Lu found a problem with that version already => try today's; it's
> copper-plated and works.
>
> Thanks
>
> Paul
I hate being an ill-news bearer, but it isn't, and it doesn't.
Here is the output with a bootstrap from today's svn:
[sfilippo@localhost TEMP]$ gfortran -v
Using built-in specs.
Target: i686-pc-linux-gnu
Configured with: ../gcc/configure --prefix=/usr/local/gcc42
--enable-languages=c,c++,fortran
Thread model: posix
gcc version 4.2.0 20060905 (experimental)
[sfilippo@localhost TEMP]$ gfortran -c bug-fold.f90
bug-fold.f90: In function 'inner_sub':
bug-fold.f90:54: internal compiler error: in fold_convert, at
fold-const.c:2098
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://gcc.gnu.org/bugs.html> for instructions.
Took me a while to reduce to this size. I believe the code to be
valid; should this not be the case I'd appreciate being explained why
not, but even in that case an ICE is an ICE.
Do you want me to open a new PR? Or are you going to reopen the old one?
Thanks
Salvatore
-------------- next part --------------
module derived_type_mod
type foo_dtype
integer, pointer :: v1(:)=>null()
end type foo_dtype
end module derived_type_mod
Module tools
interface foo_d_sub
subroutine cdalv(m, v, i, desc_a, info, flag)
use derived_type_mod
Integer, intent(in) :: m,i, v(:)
integer, intent(in), optional :: flag
integer, intent(out) :: info
Type(foo_dtype), intent(out) :: desc_a
end subroutine cdalv
end interface
end module tools
subroutine foo_bar(a,p,info)
use derived_type_mod
implicit none
type(foo_dtype), intent(in) :: a
type(foo_dtype), intent(inout) :: p
integer, intent(out) :: info
info=0
call inner_sub(info)
return
contains
subroutine inner_sub(info)
use tools
implicit none
integer, intent(out) :: info
integer :: i, nt,iv(10)
i = 0
nt = 1
call foo_d_sub(nt,iv,i,p,info,flag=1)
return
end subroutine inner_sub
end subroutine foo_bar
More information about the Fortran
mailing list