[Bug c/64663] New: ICE at -O1 and above with -g enabled on x86_64-linux-gnu

[Bug c/64663] New: ICE at -O1 and above with -g enabled on x86_64-linux-gnu

[su at cs dot ucdavis.edu]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64663

            Bug ID: 64663
           Summary: ICE at -O1 and above with -g enabled on
                    x86_64-linux-gnu
           Product: gcc
           Version: 5.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
          Assignee: unassigned at gcc dot gnu.org
          Reporter: su at cs dot ucdavis.edu

The following code causes an ICE when compiled with the current gcc trunk at
-O1 and above with -g enabled on x86_64-linux-gnu in both 32-bit and 64-bit
modes.

The ICE also affects all earlier versions of GCC since 4.6. 

$ gcc-trunk -v
Using built-in specs.
COLLECT_GCC=gcc-trunk
COLLECT_LTO_WRAPPER=/usr/local/gcc-trunk/libexec/gcc/x86_64-unknown-linux-gnu/5.0.0/lto-wrapper
Target: x86_64-unknown-linux-gnu
Configured with: ../gcc-trunk/configure --prefix=/usr/local/gcc-trunk
--enable-languages=c,c++ --disable-werror --enable-multilib
Thread model: posix
gcc version 5.0.0 20150118 (experimental) [trunk revision 219826] (GCC) 

$ 
$ gcc-trunk -O0 -g -c small.c
$ gcc-trunk -O1 -c small.c
$ 
$ gcc-trunk -O1 -g -c small.c
small.c: In function ‘fn1’:
small.c:13:1: internal compiler error: Segmentation fault
 }
 ^
0xa9f6af crash_signal
    ../../gcc-trunk/gcc/toplev.c:381
0x736a40 decl_piece_bitsize
    ../../gcc-trunk/gcc/dwarf2out.c:5072
0x766c47 dw_sra_loc_expr
    ../../gcc-trunk/gcc/dwarf2out.c:13943
0x7673e0 dw_loc_list
    ../../gcc-trunk/gcc/dwarf2out.c:14118
0x7673e0 loc_list_from_tree
    ../../gcc-trunk/gcc/dwarf2out.c:14562
0x76c5c3 add_location_or_const_value_attribute
    ../../gcc-trunk/gcc/dwarf2out.c:16070
0x753250 gen_variable_die
    ../../gcc-trunk/gcc/dwarf2out.c:19240
0x754ef9 gen_decl_die
    ../../gcc-trunk/gcc/dwarf2out.c:20969
0x7566a4 process_scope_var
    ../../gcc-trunk/gcc/dwarf2out.c:20484
0x756877 decls_for_scope
    ../../gcc-trunk/gcc/dwarf2out.c:20509
0x756f6a gen_subprogram_die
    ../../gcc-trunk/gcc/dwarf2out.c:18832
0x754f69 gen_decl_die
    ../../gcc-trunk/gcc/dwarf2out.c:20902
0x755f9c dwarf2out_decl
    ../../gcc-trunk/gcc/dwarf2out.c:21335
0x75636e dwarf2out_function_decl
    ../../gcc-trunk/gcc/dwarf2out.c:21343
0x7c2dd4 rest_of_handle_final
    ../../gcc-trunk/gcc/final.c:4521
0x7c2dd4 execute
    ../../gcc-trunk/gcc/final.c:4563
Please submit a full bug report,
with preprocessed source if appropriate.
Please include the complete backtrace with any bug report.
See <http://gcc.gnu.org/bugs.html> for instructions.
$ 


-----------------------------------


int a, b;

void
fn1 ()
{
  int c[9] = { 0 };
  for (;;)
    {
      a = (int) 3655990292 % 9;
      if (b && c[a])
    c[a] = 0;
    }
}
>From gcc-bugs-return-473787-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Mon Jan 19 04:55:47 2015
Return-Path: <gcc-bugs-return-473787-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 4021 invoked by alias); 19 Jan 2015 04:55:45 -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 3967 invoked by uid 48); 19 Jan 2015 04:55:38 -0000
From: "pangbw at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug rtl-optimization/64532] [4.9, 5 Regression] internal compiler error: Max. number of generated reload insns per insn is achieved (90)
Date: Mon, 19 Jan 2015 04:55:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: rtl-optimization
X-Bugzilla-Version: 5.0
X-Bugzilla-Keywords: ra
X-Bugzilla-Severity: major
X-Bugzilla-Who: pangbw at gmail dot com
X-Bugzilla-Status: NEW
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: 5.0
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields:
Message-ID: <bug-64532-4-TSXSM4AGZz@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-64532-4@http.gcc.gnu.org/bugzilla/>
References: <bug-64532-4@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: 2015-01/txt/msg01781.txt.bz2
Content-length: 586

https://gcc.gnu.org/bugzilla/show_bug.cgi?idd532

--- Comment #6 from baoshan <pangbw at gmail dot com> ---
After several days study to the code, I turn to feel the code is wrong. It
seems we should use "=t" instead of "=w" for 'y' because single float register
is expected here for "vcvt.f32.s32". From the document, "w" just means  VFP
floating-point register  but not distinguishing single, double or quad float
register:
https://gcc.gnu.org/onlinedocs/gcc/Machine-Constraints.html#Machine-Constraints

The constraint letter 't' is not documented in the GCC doc, this is a issue.

[Bug c/64663] ICE at -O1 and above with -g enabled on x86_64-linux-gnu

[mpolacek at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64663

Marek Polacek <mpolacek at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2015-01-19
                 CC|                            |mpolacek at gcc dot gnu.org
   Target Milestone|---                         |4.8.5
     Ever confirmed|0                           |1

--- Comment #1 from Marek Polacek <mpolacek at gcc dot gnu.org> ---
Confirmed.

[Bug c/64663] ICE at -O1 and above with -g enabled on x86_64-linux-gnu

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64663

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jakub at gcc dot gnu.org

--- Comment #2 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Started with r171450.

[Bug c/64663] ICE at -O1 and above with -g enabled on x86_64-linux-gnu

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64663

--- Comment #3 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Reduced testcase -O2 -g:
void
foo (void)
{
  int a[9];
  a[-8] = 0;
}
surprisingly doesn't ICE with -7 or -9 instead of -8, ICEs again with -16. 
Will have a look.

[Bug c/64663] [4.8/4.9/5 Regression] ICE at -O1 and above with -g enabled on x86_64-linux-gnu

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64663

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED
                 CC|jakub at gcc dot gnu.org           |
           Assignee|unassigned at gcc dot gnu.org      |jakub at gcc dot gnu.org
            Summary|ICE at -O1 and above with   |[4.8/4.9/5 Regression] ICE
                   |-g enabled on               |at -O1 and above with -g
                   |x86_64-linux-gnu            |enabled on x86_64-linux-gnu

[Bug debug/64663] [4.8/4.9/5 Regression] ICE at -O1 and above with -g enabled on x86_64-linux-gnu

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64663

--- Comment #4 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Created attachment 34484
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=34484&action=edit
gcc5-pr64663.patch

Untested fix.  Perhaps for the backport just the second hunk would be enough
(plus testcase).

[Bug debug/64663] [4.8/4.9/5 Regression] ICE at -O1 and above with -g enabled on x86_64-linux-gnu

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64663

--- Comment #5 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Author: jakub
Date: Tue Jan 20 18:29:58 2015
New Revision: 219906

URL: https://gcc.gnu.org/viewcvs?rev=219906&root=gcc&view=rev
Log:
    PR debug/64663
    * dwarf2out.c (decl_piece_node): Don't put bitsize into
    mode if bitsize <= 0.
    (decl_piece_bitsize, adjust_piece_list, add_var_loc_to_decl,
    dw_sra_loc_expr): Use HOST_WIDE_INT instead of int for bit
    sizes and positions.

    * gcc.dg/pr64663.c: New test.

Added:
    trunk/gcc/testsuite/gcc.dg/pr64663.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/dwarf2out.c
    trunk/gcc/testsuite/ChangeLog

[Bug debug/64663] [4.8/4.9 Regression] ICE at -O1 and above with -g enabled on x86_64-linux-gnu

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64663

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[4.8/4.9/5 Regression] ICE  |[4.8/4.9 Regression] ICE at
                   |at -O1 and above with -g    |-O1 and above with -g
                   |enabled on x86_64-linux-gnu |enabled on x86_64-linux-gnu

--- Comment #6 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Fixed on the trunk so far.

[Bug debug/64663] [4.8/4.9 Regression] ICE at -O1 and above with -g enabled on x86_64-linux-gnu

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64663

--- Comment #7 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Author: jakub
Date: Sun Feb  1 17:31:48 2015
New Revision: 220321

URL: https://gcc.gnu.org/viewcvs?rev=220321&root=gcc&view=rev
Log:
    Backported from mainline
    2015-01-20  Jakub Jelinek  <jakub@redhat.com>

    PR debug/64663
    * dwarf2out.c (decl_piece_node): Don't put bitsize into
    mode if bitsize <= 0.
    (decl_piece_bitsize, adjust_piece_list, add_var_loc_to_decl,
    dw_sra_loc_expr): Use HOST_WIDE_INT instead of int for bit
    sizes and positions.

    * gcc.dg/pr64663.c: New test.

Added:
    branches/gcc-4_9-branch/gcc/testsuite/gcc.dg/pr64663.c
Modified:
    branches/gcc-4_9-branch/gcc/ChangeLog
    branches/gcc-4_9-branch/gcc/dwarf2out.c
    branches/gcc-4_9-branch/gcc/testsuite/ChangeLog

[Bug debug/64663] [4.8 Regression] ICE at -O1 and above with -g enabled on x86_64-linux-gnu

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64663

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|---                         |FIXED
   Target Milestone|4.8.5                       |4.9.3
      Known to fail|                            |4.8.5

--- Comment #8 from Richard Biener <rguenth at gcc dot gnu.org> ---
Fixed for 4.9.3.