https://gcc.gnu.org/bugzilla/show_bug.cgi?id=125535
Bug ID: 125535
Summary: Wrong-code for implied-do with allocatable-component
derived type and transformational intrinsic
Product: gcc
Version: 16.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: fortran
Assignee: unassigned at gcc dot gnu.org
Reporter: jvdelisle at gcc dot gnu.org
Target Milestone: ---
When a nested implied-do array constructor calls a transformational intrinsic
(e.g. RESHAPE) and the element type is a derived type with allocatable
components, gfortran produces wrong runtime values or crashes with a
heap-use-after-free.
The argument temporaries are freed before the result's allocatable components
are deep-copied. Since RESHAPE does a shallow byte-copy, the result's
component pointers alias those of the temporaries, so freeing first yields
use-after-free.
module m
implicit none
type :: t
real, allocatable :: v(:)
end type
contains
pure function make (x) result (r)
real, intent(in) :: x(:)
type(t) :: r
r%v = x
end function
end module
program test
use m
implicit none
integer, parameter :: n = 3, k = 2
type(t), allocatable :: a(:,:)
real :: h(k, n)
integer :: i, j
h(1,:) = [10., 20., 30.]
h(2,:) = [11., 21., 31.]
a = reshape ([([(make (h(:,i)), j=1,1)], i=1,n)], [1,n])
if (any (a(1,1)%v /= h(:,1))) stop 1
if (any (a(1,2)%v /= h(:,2))) stop 2
if (any (a(1,3)%v /= h(:,3))) stop 3
end program
$ gfortran implied_do_alloc_comp_1.f90 && ./a.out
The program stops at STOP 1 or crashes with a
heap-use-after-free. It should run and exit cleanly.
This one could be related to 90519 and 96910, not sure
