[Bug fortran/60147] New: [Regression] ICE with -fdump-tree-original and NAMELIST

[Bug fortran/60147] New: [Regression] ICE with -fdump-tree-original and NAMELIST

[kargl at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60147

            Bug ID: 60147
           Summary: [Regression] ICE with -fdump-tree-original and
                    NAMELIST
           Product: gcc
           Version: 4.9.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: fortran
          Assignee: unassigned at gcc dot gnu.org
          Reporter: kargl at gcc dot gnu.org

program foo
   implicit none
   character(len=5) :: s = 'tue'
   namelist /mylist/ s
end program

% gfc4x -o z -fdump-tree-original g.f90
g.f90: In function 'foo':
g.f90:5:0: internal compiler error: Segmentation fault
 end program
 ^
no stack trace because unwind library not available
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.

(gdb) run -fdump-tree-original g.f90
Starting program:
/usr/home/sgk/work/4x/libexec/gcc/x86_64-unknown-freebsd11.0/4.9.0/f951
-fdump-tree-original g.f90
 foo
Program received signal SIGSEGV, Segmentation fault.
0x0000000000b094a2 in dump_generic_node (buffer=0x18ccc00 <buffer>, 
    node=0x202f1f4c8, spc=2, flags=512, is_stmt=false)
    at ../../gcc4x/gcc/tree-pretty-print.c:1495
1495    ../../gcc4x/gcc/tree-pretty-print.c: No such file or directory.
(gdb) bt
#0  0x0000000000b094a2 in dump_generic_node (buffer=0x18ccc00 <buffer>, 
    node=0x202f1f4c8, spc=2, flags=512, is_stmt=false)
    at ../../gcc4x/gcc/tree-pretty-print.c:1495
#1  0x0000000000b18de5 in print_declaration (buffer=0x18ccc00 <buffer>, 
    t=0x202dff7e8, spc=2, flags=512)
    at ../../gcc4x/gcc/tree-pretty-print.c:2759
#2  0x0000000000b0a2d4 in dump_generic_node (buffer=0x18ccc00 <buffer>, 
    node=0x202dec510, spc=0, flags=512, is_stmt=true)
    at ../../gcc4x/gcc/tree-pretty-print.c:1723
#3  0x0000000000b1969e in print_generic_stmt_indented (file=<optimized out>, 
    t=0x202dec510, flags=512, indent=0)
    at ../../gcc4x/gcc/tree-pretty-print.c:157
#4  0x0000000000a845b9 in dump_function_to_file (fndecl=<optimized out>, 
    file=0x202bb3370, flags=512) at ../../gcc4x/gcc/tree-cfg.c:7117
#5  0x0000000000a9de89 in dump_function (phase=5, fn=0x202f24b00)
    at ../../gcc4x/gcc/tree-dump.c:170
#6  0x0000000000607c33 in gfc_generate_function_code (ns=0x203459600)
    at ../../gcc4x/gcc/fortran/trans-decl.c:5706
#7  0x000000000059a0b7 in translate_all_program_units (
    main_in_tu=<optimized out>, gfc_global_ns_list=0x203459600)
    at ../../gcc4x/gcc/fortran/parse.c:4536
#8  gfc_parse_file () at ../../gcc4x/gcc/fortran/parse.c:4733
#9  0x00000000005d7366 in gfc_be_parse_file ()
    at ../../gcc4x/gcc/fortran/f95-lang.c:188
#10 0x0000000000a4f136 in compile_file () at ../../gcc4x/gcc/toplev.c:548
#11 0x0000000000a4f9f3 in do_compile () at ../../gcc4x/gcc/toplev.c:1914
#12 0x0000000000a50562 in toplev_main (argc=3, argv=0x7fffffffd4c8)
    at ../../gcc4x/gcc/toplev.c:1990
#13 0x0000000000528daf in _start ()

[Bug fortran/60147] [4.9 Regression] ICE with -fdump-tree-original and NAMELIST

[dominiq at lps dot ens.fr]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60147

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2014-02-11
                 CC|                            |burnus at gcc dot gnu.org
            Summary|[Regression] ICE with       |[4.9 Regression] ICE with
                   |-fdump-tree-original and    |-fdump-tree-original and
                   |NAMELIST                    |NAMELIST
     Ever confirmed|0                           |1

--- Comment #1 from Dominique d'Humieres <dominiq at lps dot ens.fr> ---
Confirmed r205305 is OK and r205798 gives the ICE. Usual suspect is r205679.

[Bug fortran/60147] [4.9 Regression] ICE with -fdump-tree-original and NAMELIST

[dominiq at lps dot ens.fr]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60147

--- Comment #2 from Dominique d'Humieres <dominiq at lps dot ens.fr> ---
> Usual suspect is r205679.

It is the culprit: r205678 is OK, r205679 gives the ICE.

[Bug middle-end/60147] [4.9 Regression] ICE with -fdump-tree-original and NAMELIST

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60147

Tobias Burnus <burnus at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |ice-on-valid-code
          Component|fortran                     |middle-end
   Target Milestone|---                         |4.9.0

[Bug middle-end/60147] [4.9 Regression] ICE with -fdump-tree-original and NAMELIST

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60147

Tobias Burnus <burnus at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED
           Assignee|unassigned at gcc dot gnu.org      |burnus at gcc dot gnu.org

--- Comment #3 from Tobias Burnus <burnus at gcc dot gnu.org> ---
Created attachment 32197
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=32197&action=edit
draft patch

[Bug middle-end/60147] [4.9 Regression] ICE with -fdump-tree-original and NAMELIST

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60147

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P1

[Bug middle-end/60147] [4.9 Regression] ICE with -fdump-tree-original and NAMELIST

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60147

--- Comment #4 from Tobias Burnus <burnus at gcc dot gnu.org> ---
Author: burnus
Date: Thu Feb 27 23:24:20 2014
New Revision: 208209

URL: http://gcc.gnu.org/viewcvs?rev=208209&root=gcc&view=rev
Log:
2014-02-28  Tobias Burnus  <burnus@net-b.de>

        PR middle-end/60147
        * tree-pretty-print.c (dump_generic_node, print_declaration):
        Handle NAMELIST_DECL.


Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/tree-pretty-print.c

[Bug middle-end/60147] [4.9 Regression] ICE with -fdump-tree-original and NAMELIST

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60147

Tobias Burnus <burnus at gcc dot gnu.org> changed:

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

--- Comment #5 from Tobias Burnus <burnus at gcc dot gnu.org> ---
FIXED on the trunk.

Thanks, Steve, for the report and sorry for the breakage!

[Bug middle-end/60147] [4.9 Regression] ICE with -fdump-tree-original and NAMELIST

[sgk at troutmask dot apl.washington.edu]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60147

--- Comment #6 from Steve Kargl <sgk at troutmask dot apl.washington.edu> ---
On Thu, Feb 27, 2014 at 11:26:13PM +0000, burnus at gcc dot gnu.org wrote:
> FIXED on the trunk.
> 
> Thanks, Steve, for the report and sorry for the breakage!
> 

Thanks.  I don't use namelists.  I stumbled across
the problem when debugging one of the namelist IO
issues jerryd recently fixed.