catastrophic error: **Internal compiler error: segmentation violation signal raised**

[nnemec]

I am using ifort 11.1 20100414

Compiling the source file below with

ifort -O3 -check bounds -c pbc_lists_X.f90

gives the error message :

--------
: catastrophic error: **Internal compiler error: segmentation violation signal raised** Please report this error along with the circumstances in which it occurred in a Software Problem Report. Note: File and line given may not be explicit cause of this error.

compilation aborted for pbc_lists_X.f90 (code 3)
--------

Greetings,
Norbert Nemec

PS: Is there any obvious place to report internal compiler errors? These are obvious bug reports and as such very different from user support issues.


----- pbc_lists_X.f90 -----
module pbc_lists

implicit none

integer,public :: n_periodic, n_atoms
logical,public :: transport_calculation, transport_lead_calculation

contains

subroutine evaluate_number_of_supercells_needed_in_periodic_systems

implicit none

real*8 :: lenght_of_lattice_vector_l

real*8, dimension(3,3):: A

integer :: i_atom, i_latt, i_latt_2

if( transport_lead_calculation .or. transport_calculation)then
do i_latt = 1,3

end do
end if

if(transport_lead_calculation .or. transport_calculation)then
do i_atom = 1,n_atoms

end do
end if

do i_latt = n_periodic+1,3,1
do i_latt_2 = 1,i_latt-1,1
lenght_of_lattice_vector_l = (dot_product(A(:,i_latt_2),A(:,i_latt_2))) !

A(:,i_latt) = A(:,i_latt) - A(:,i_latt_2)* &
dot_product(A(:,i_latt),A(:,i_latt_2)) / lenght_of_lattice_vector_l

end do
end do

end subroutine evaluate_number_of_supercells_needed_in_periodic_systems

end module pbc_lists
----------------------------

[mecej4]

The error does not occur with the 11.1, Update-7 (20100806) version.

There is a new compiler release (Release 12) expected this month.

I have seen frequent postings about compiler errors in this Forum, and the Intel people have responded promptly to those postings. The majority of such errors are fixed in the next update of the compiler, so I tend to assume that this forum is a fine place for reporting bugs.