[Bug libgomp/65362] New: OpenACC compilation on Tegra K1 (ARM)

[Bug libgomp/65362] New: OpenACC compilation on Tegra K1 (ARM)

[zziolko at clemson dot edu]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65362

            Bug ID: 65362
           Summary: OpenACC compilation on Tegra K1 (ARM)
           Product: gcc
           Version: 5.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: libgomp
          Assignee: unassigned at gcc dot gnu.org
          Reporter: zziolko at clemson dot edu
                CC: jakub at gcc dot gnu.org

Created attachment 34992
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=34992&action=edit
gfortran -v -save-temps -fopenacc abort-2.f90 output

The GCC 5.0 snapshot gcc-5-20150301 builds successfully on Tegra K1 processor
(Ubuntu 14.04 Linux on NVIDIA's Jetson TK1 platform). All attempts to compile c
or fortran code with -fopenacc fail.  

For example code abort-2.f90 (from the testsuite):

program main
  implicit none

  integer :: argc
  argc = command_argument_count ()

  !$acc parallel copyin(argc)
  if (argc .ne. 0) then
     call abort
  end if
  !$acc end parallel

end program main 

compiled using gfortran -fopenacc abort-2.f90 gives a message


abort-2.f90: In function ‘MAIN__._omp_fn.0’:
abort-2.f90:8:0: internal compiler error: in
convert_memory_address_addr_space_1, at explow.c:292
   if (argc .ne. 0) then
 ^
0x1df1e1 convert_memory_address_addr_space_1
    ../../gcc-5-20150301/gcc/explow.c:292
0x1df1e1 convert_memory_address_addr_space
    ../../gcc-5-20150301/gcc/explow.c:389
0x1df1e1 memory_address_addr_space(machine_mode, rtx_def*, unsigned char)
    ../../gcc-5-20150301/gcc/explow.c:403
0x1f1779 expand_expr_real_1(tree_node*, rtx_def*, machine_mode,
expand_modifier, rtx_def**, bool)
    ../../gcc-5-20150301/gcc/expr.c:9904
0x1ef439 expand_expr_real_1(tree_node*, rtx_def*, machine_mode,
expand_modifier, rtx_def**, bool)
    ../../gcc-5-20150301/gcc/expr.c:10171
0x1f8e83 store_expr_with_bounds(tree_node*, rtx_def*, int, bool, tree_node*)
    ../../gcc-5-20150301/gcc/expr.c:5385
0x1fe1a5 expand_assignment(tree_node*, tree_node*, bool)
    ../../gcc-5-20150301/gcc/expr.c:5154
0x1575e7 expand_gimple_stmt_1
    ../../gcc-5-20150301/gcc/cfgexpand.c:3385
0x1575e7 expand_gimple_stmt
    ../../gcc-5-20150301/gcc/cfgexpand.c:3481
0x158e4b expand_gimple_basic_block
    ../../gcc-5-20150301/gcc/cfgexpand.c:5470
0x15a649 execute
    ../../gcc-5-20150301/gcc/cfgexpand.c:6088


Similar message for the rest of the fortran or c tests. 

Marcin
>From gcc-bugs-return-479818-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Mon Mar 09 15:16:34 2015
Return-Path: <gcc-bugs-return-479818-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 10042 invoked by alias); 9 Mar 2015 15:16:34 -0000
Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc-bugs.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc-bugs/>
List-Post: <mailto:gcc-bugs@gcc.gnu.org>
List-Help: <mailto:gcc-bugs-help@gcc.gnu.org>
Sender: gcc-bugs-owner@gcc.gnu.org
Delivered-To: mailing list gcc-bugs@gcc.gnu.org
Received: (qmail 9983 invoked by uid 48); 9 Mar 2015 15:16:30 -0000
From: "ramana at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/65121] [5 regression] long_call attribute broken weak symbol arm*
Date: Mon, 09 Mar 2015 15:16:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: target
X-Bugzilla-Version: 5.0
X-Bugzilla-Keywords:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: ramana at gcc dot gnu.org
X-Bugzilla-Status: RESOLVED
X-Bugzilla-Priority: P1
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: 5.0
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields: bug_status resolution
Message-ID: <bug-65121-4-Chkh0kN8zo@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-65121-4@http.gcc.gnu.org/bugzilla/>
References: <bug-65121-4@http.gcc.gnu.org/bugzilla/>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 7bit
X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/
Auto-Submitted: auto-generated
MIME-Version: 1.0
X-SW-Source: 2015-03/txt/msg00962.txt.bz2
Content-length: 794

https://gcc.gnu.org/bugzilla/show_bug.cgi?ide121

Ramana Radhakrishnan <ramana at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|---                         |FIXED

--- Comment #4 from Ramana Radhakrishnan <ramana at gcc dot gnu.org> ---

https://gcc.gnu.org/ml/gcc-cvs/2015-03/msg00140.html should have fixed it. My
auto-tester is down so I can't see the "fixed" testresults however
https://gcc.gnu.org/ml/gcc-testresults/2015-03/msg00993.html shows it's fixed.

Alex, please put PR numbers in your changelog and commit messages properly so
that folks know when things are fixed up.

regards
Ramana

[Bug libgomp/65362] OpenACC compilation on Tegra K1 (ARM)

[ramana at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65362

Ramana Radhakrishnan <ramana at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2015-04-27
                 CC|                            |ramana at gcc dot gnu.org
     Ever confirmed|0                           |1
           Severity|normal                      |enhancement

--- Comment #1 from Ramana Radhakrishnan <ramana at gcc dot gnu.org> ---
The feature just isn't there - patches welcome :)