File: alloc_comp_constructor_4.f90

package info (click to toggle)
gcc-arm-none-eabi 15%3A14.2.rel1-1
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid, trixie
  • size: 1,099,328 kB
  • sloc: cpp: 3,627,108; ansic: 2,571,498; ada: 834,230; f90: 235,082; makefile: 79,231; asm: 74,984; xml: 51,692; exp: 39,736; sh: 33,298; objc: 15,629; python: 15,069; fortran: 14,429; pascal: 7,003; awk: 5,070; perl: 3,106; ml: 285; lisp: 253; lex: 204; haskell: 135
file content (16 lines) | stat: -rw-r--r-- 533 bytes parent folder | download | duplicates (3) 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
 
! { dg-do run }
! Tests the fix for PR32795, which was primarily about memory leakage is
! certain combinations of alloctable components and constructors. This test
! which appears in comment #2 of the PR has the advantage of a wrong
! numeric result which is symptomatic.
!
! Contributed by Tobias Burnus <burnus@gcc.gnu.org>
!
  type :: a
    integer, allocatable :: i(:)
  end type a
  type(a) :: x, y
  x = a ([1, 2, 3])
  y = a (x%i(:))  ! used to cause a memory leak and wrong result
  if (any (x%i .ne. [1, 2, 3])) STOP 1
end