public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug fortran/66493] New: ICE on alternate return argument for typebound procedure
@ 2015-06-10 16:21 gerhard.steinmetz.fortran@t-online.de
  2015-06-10 21:18 ` [Bug fortran/66493] " dominiq at lps dot ens.fr
                   ` (7 more replies)
  0 siblings, 8 replies; 9+ messages in thread
From: gerhard.steinmetz.fortran@t-online.de @ 2015-06-10 16:21 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66493

            Bug ID: 66493
           Summary: ICE on alternate return argument for typebound
                    procedure
           Product: gcc
           Version: 5.1.1
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: fortran
          Assignee: unassigned at gcc dot gnu.org
          Reporter: gerhard.steinmetz.fortran@t-online.de
  Target Milestone: ---

This code with an alternate return argument (*) seems to be improper :
   module p
      type t
      contains
         procedure :: s
      end type
   contains
      subroutine s(*)
      end
   end module

and yields (with gfortran 5.1.1 on SUSE Linux 13.2, 64 bit) :
f951: internal compiler error: in resolve_typebound_procedure, at
fortran/resolve.c:12210


^ permalink raw reply	[flat|nested] 9+ messages in thread

* [Bug fortran/66493] ICE on alternate return argument for typebound procedure
  2015-06-10 16:21 [Bug fortran/66493] New: ICE on alternate return argument for typebound procedure gerhard.steinmetz.fortran@t-online.de
@ 2015-06-10 21:18 ` dominiq at lps dot ens.fr
  2015-06-10 22:21 ` kargl at gcc dot gnu.org
                   ` (6 subsequent siblings)
  7 siblings, 0 replies; 9+ messages in thread
From: dominiq at lps dot ens.fr @ 2015-06-10 21:18 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66493

Dominique d'Humieres <dominiq at lps dot ens.fr> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2015-06-10
     Ever confirmed|0                           |1

--- Comment #1 from Dominique d'Humieres <dominiq at lps dot ens.fr> ---
Confirmed from 4.6 up to trunk (6.0).


^ permalink raw reply	[flat|nested] 9+ messages in thread

* [Bug fortran/66493] ICE on alternate return argument for typebound procedure
  2015-06-10 16:21 [Bug fortran/66493] New: ICE on alternate return argument for typebound procedure gerhard.steinmetz.fortran@t-online.de
  2015-06-10 21:18 ` [Bug fortran/66493] " dominiq at lps dot ens.fr
@ 2015-06-10 22:21 ` kargl at gcc dot gnu.org
  2015-06-11  0:04 ` kargl at gcc dot gnu.org
                   ` (5 subsequent siblings)
  7 siblings, 0 replies; 9+ messages in thread
From: kargl at gcc dot gnu.org @ 2015-06-10 22:21 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66493

kargl at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |kargl at gcc dot gnu.org
           Assignee|unassigned at gcc dot gnu.org      |kargl at gcc dot gnu.org

--- Comment #2 from kargl at gcc dot gnu.org ---
I have a patch.  Just need to come up with a complete test.
With my patch the following runs to completion:

program foo
   integer :: cnt = 0
1  call bar(*10, *20)
   cnt = cnt + 1
   goto 30
10 cnt = cnt + 1
   if (cnt /= 3) call abort
   stop
20 cnt = cnt + 1
30 goto 1
end program foo

subroutine bar(*, *)
   implicit none
   integer, save :: i = 0
   if (i < 0) return 1
   if (i > 0) then
      i = -i
      return 2
   end if
   i = i + 1
end subroutine bar

Having never used type bound procedure, it may take a bit to convert this
into a testcase.


^ permalink raw reply	[flat|nested] 9+ messages in thread

* [Bug fortran/66493] ICE on alternate return argument for typebound procedure
  2015-06-10 16:21 [Bug fortran/66493] New: ICE on alternate return argument for typebound procedure gerhard.steinmetz.fortran@t-online.de
  2015-06-10 21:18 ` [Bug fortran/66493] " dominiq at lps dot ens.fr
  2015-06-10 22:21 ` kargl at gcc dot gnu.org
@ 2015-06-11  0:04 ` kargl at gcc dot gnu.org
  2015-06-11  0:04 ` kargl at gcc dot gnu.org
                   ` (4 subsequent siblings)
  7 siblings, 0 replies; 9+ messages in thread
From: kargl at gcc dot gnu.org @ 2015-06-11  0:04 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66493

--- Comment #3 from kargl at gcc dot gnu.org ---
Created attachment 35748
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=35748&action=edit
test program without type bound procedure


^ permalink raw reply	[flat|nested] 9+ messages in thread

* [Bug fortran/66493] ICE on alternate return argument for typebound procedure
  2015-06-10 16:21 [Bug fortran/66493] New: ICE on alternate return argument for typebound procedure gerhard.steinmetz.fortran@t-online.de
                   ` (2 preceding siblings ...)
  2015-06-11  0:04 ` kargl at gcc dot gnu.org
@ 2015-06-11  0:04 ` kargl at gcc dot gnu.org
  2015-06-11  0:05 ` kargl at gcc dot gnu.org
                   ` (3 subsequent siblings)
  7 siblings, 0 replies; 9+ messages in thread
From: kargl at gcc dot gnu.org @ 2015-06-11  0:04 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66493

--- Comment #4 from kargl at gcc dot gnu.org ---
Created attachment 35749
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=35749&action=edit
test program with tbp


^ permalink raw reply	[flat|nested] 9+ messages in thread

* [Bug fortran/66493] ICE on alternate return argument for typebound procedure
  2015-06-10 16:21 [Bug fortran/66493] New: ICE on alternate return argument for typebound procedure gerhard.steinmetz.fortran@t-online.de
                   ` (3 preceding siblings ...)
  2015-06-11  0:04 ` kargl at gcc dot gnu.org
@ 2015-06-11  0:05 ` kargl at gcc dot gnu.org
  2015-06-11  0:10 ` kargl at gcc dot gnu.org
                   ` (2 subsequent siblings)
  7 siblings, 0 replies; 9+ messages in thread
From: kargl at gcc dot gnu.org @ 2015-06-11  0:05 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66493

--- Comment #5 from kargl at gcc dot gnu.org ---
Created attachment 35750
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=35750&action=edit
Patch that gets rid of ICE


^ permalink raw reply	[flat|nested] 9+ messages in thread

* [Bug fortran/66493] ICE on alternate return argument for typebound procedure
  2015-06-10 16:21 [Bug fortran/66493] New: ICE on alternate return argument for typebound procedure gerhard.steinmetz.fortran@t-online.de
                   ` (4 preceding siblings ...)
  2015-06-11  0:05 ` kargl at gcc dot gnu.org
@ 2015-06-11  0:10 ` kargl at gcc dot gnu.org
  2015-06-11  8:05 ` gerhard.steinmetz.fortran@t-online.de
  2015-06-11  8:22 ` gerhard.steinmetz.fortran@t-online.de
  7 siblings, 0 replies; 9+ messages in thread
From: kargl at gcc dot gnu.org @ 2015-06-11  0:10 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66493

kargl at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Assignee|kargl at gcc dot gnu.org           |unassigned at gcc dot gnu.org

--- Comment #6 from kargl at gcc dot gnu.org ---
Some who understands the internals of type bound procedures will need to
expand my patch.  The patch cures the ICE in comment #1.  I've attached
two test programs.  The 1st one does not use TBP.  The 2nd one does.  I
get

gfc6 -o z gfc6 -o z alternate_return_2.f90 && ./z
alternate_return_2.f90:30:17:

    call argh%bar(*10, *20)
                 1
Error: Syntax error in argument list at (1)


The reason is that whoever write the TBP code did not consider
the possibility of alternate return.  I don't blame them.  I
simply don't know how to fix it.


^ permalink raw reply	[flat|nested] 9+ messages in thread

* [Bug fortran/66493] ICE on alternate return argument for typebound procedure
  2015-06-10 16:21 [Bug fortran/66493] New: ICE on alternate return argument for typebound procedure gerhard.steinmetz.fortran@t-online.de
                   ` (5 preceding siblings ...)
  2015-06-11  0:10 ` kargl at gcc dot gnu.org
@ 2015-06-11  8:05 ` gerhard.steinmetz.fortran@t-online.de
  2015-06-11  8:22 ` gerhard.steinmetz.fortran@t-online.de
  7 siblings, 0 replies; 9+ messages in thread
From: gerhard.steinmetz.fortran@t-online.de @ 2015-06-11  8:05 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66493

--- Comment #7 from Gerhard Steinmetz <gerhard.steinmetz.fortran@t-online.de> ---
Interestingly, there are some tests in the test suite already,
especially altreturn_3.f90 and altreturn_8.f90.

The latter is using a nopass attribute and compiles.
It still should be tested with some real runtime tests, as yours.

Some of the following constructed tests fail to compile.


$ cat zarc1_1.f90
module m
   implicit none
   type t
   contains
      procedure :: s1
      procedure :: s2
      generic :: s => s1, s2
   end type
contains
   subroutine s1(*)
   end
   subroutine s2(*, *)
   end
end module


$ cat zarc1_2_pass.f90
module m
   implicit none
   type t
   contains
      procedure, pass :: s1
      procedure, pass :: s2
      generic :: s => s1, s2
   end type
contains
   subroutine s1(*)
   end
   subroutine s2(*, *)
   end
end module


$ cat zarc1_3_nopass.f90
module m
   implicit none
   type t
   contains
      procedure, nopass :: s1
      procedure, nopass :: s2
      generic :: s => s1, s2
   end type
contains
   subroutine s1(*)
   end
   subroutine s2(*, *)
   end
end module



$ cat zarcg_1.f90
module m
   implicit none
   type t
   contains
      procedure :: s1
      procedure :: s2
      procedure :: s3
      generic :: s => s1, s2, s3
   end type
contains
   subroutine s1(x)
      class(t) :: x
   end
   subroutine s2(x, *)
      class(t) :: x
   end
   subroutine s3(x, *, *)
      class(t) :: x
   end
end module


$ cat zarcg_2_pass.f90
module m
   implicit none
   type t
   contains
      procedure, pass :: s1
      procedure, pass :: s2
      procedure, pass :: s3
      generic :: s => s1, s2, s3
   end type
contains
   subroutine s1(x)
      class(t) :: x
   end
   subroutine s2(x, *)
      class(t) :: x
   end
   subroutine s3(x, *, *)
      class(t) :: x
   end
end module


$ cat zarcg_3_nopass.f90
module m
   implicit none
   type t
   contains
      procedure, nopass :: s1
      procedure, nopass :: s2
      procedure, nopass :: s3
      generic :: s => s1, s2, s3
   end type
contains
   subroutine s1(x)
      class(t) :: x
   end
   subroutine s2(x, *)
      class(t) :: x
   end
   subroutine s3(x, *, *)
      class(t) :: x
   end
end module


^ permalink raw reply	[flat|nested] 9+ messages in thread

* [Bug fortran/66493] ICE on alternate return argument for typebound procedure
  2015-06-10 16:21 [Bug fortran/66493] New: ICE on alternate return argument for typebound procedure gerhard.steinmetz.fortran@t-online.de
                   ` (6 preceding siblings ...)
  2015-06-11  8:05 ` gerhard.steinmetz.fortran@t-online.de
@ 2015-06-11  8:22 ` gerhard.steinmetz.fortran@t-online.de
  7 siblings, 0 replies; 9+ messages in thread
From: gerhard.steinmetz.fortran@t-online.de @ 2015-06-11  8:22 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66493

--- Comment #8 from Gerhard Steinmetz <gerhard.steinmetz.fortran@t-online.de> ---
Case zarc1_2_pass.f90 should be definitely illegal code.


^ permalink raw reply	[flat|nested] 9+ messages in thread

end of thread, other threads:[~2015-06-11  8:22 UTC | newest]

Thread overview: 9+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-06-10 16:21 [Bug fortran/66493] New: ICE on alternate return argument for typebound procedure gerhard.steinmetz.fortran@t-online.de
2015-06-10 21:18 ` [Bug fortran/66493] " dominiq at lps dot ens.fr
2015-06-10 22:21 ` kargl at gcc dot gnu.org
2015-06-11  0:04 ` kargl at gcc dot gnu.org
2015-06-11  0:04 ` kargl at gcc dot gnu.org
2015-06-11  0:05 ` kargl at gcc dot gnu.org
2015-06-11  0:10 ` kargl at gcc dot gnu.org
2015-06-11  8:05 ` gerhard.steinmetz.fortran@t-online.de
2015-06-11  8:22 ` gerhard.steinmetz.fortran@t-online.de

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).