Bug 50174 - [4.7 Regression] ICE on derived type allocation
Summary: [4.7 Regression] ICE on derived type allocation
Status: RESOLVED DUPLICATE of bug 50050
Alias: None
Product: gcc
Classification: Unclassified
Component: fortran (show other bugs)
Version: 4.7.0
: P3 normal
Target Milestone: ---
Assignee: Not yet assigned to anyone
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2011-08-24 10:15 UTC by Philip Mason
Modified: 2011-08-25 09:15 UTC (History)
2 users (show)

See Also:
Host:
Target:
Build:
Known to work:
Known to fail:
Last reconfirmed:


Attachments
Simple module, contains type def + procedure (163 bytes, application/octet-stream)
2011-08-24 10:15 UTC, Philip Mason
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Philip Mason 2011-08-24 10:15:05 UTC
Created attachment 25086 [details]
Simple module, contains type def + procedure

Get ICE when compiling code with a function that allocates a derived type array. Compiled as "gfortran -c settee.f90". See test file attached.
Comment 1 janus 2011-08-24 11:34:30 UTC
Confirmed on 4.7 trunk. Works on 4.6.

Reduced test case:


  implicit none

  type settee
    real,allocatable :: seat(:)
  end type settee

  integer      :: is = 1, ie = 3
  type(settee),allocatable :: set1(:)

  allocate (set1(is:ie))

end
Comment 2 janus 2011-08-24 11:36:35 UTC
Probably related to PR 50050.
Comment 3 janus 2011-08-24 14:36:22 UTC
(In reply to comment #2)
> Probably related to PR 50050.

Apparently it is fixed by Mikael's patch at

http://gcc.gnu.org/ml/fortran/2011-08/msg00199.html
Comment 4 Tobias Burnus 2011-08-25 09:15:44 UTC
I marked it as duplicated of PR 50050 comment 7 as the error seems to be identically.

If the patch at http://gcc.gnu.org/ml/fortran/2011-08/msg00199.html does not fix the issue - or if the final commit does not fix the issue, feel free to reopen this PR.

Sorry for the breakage.

*** This bug has been marked as a duplicate of bug 50050 ***