[Bug c/59776] New: gcc -g -O1 ICE in expand_debug_locations, at cfgexpand.c:3865

[Bug c/59776] New: gcc -g -O1 ICE in expand_debug_locations, at cfgexpand.c:3865

[zeccav at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59776

            Bug ID: 59776
           Summary: gcc -g -O1 ICE in expand_debug_locations, at
                    cfgexpand.c:3865
           Product: gcc
           Version: 4.8.2
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
          Assignee: unassigned at gcc dot gnu.org
          Reporter: zeccav at gmail dot com

/* gcc -g -O1 ICE in expand_debug_locations, at cfgexpand.c:3865 */
/* gcc 4.8.2-7 20131212 */
typedef struct {
  float re,im;
} Complex;
void sub_(Complex *var_Dummy)
{
Complex var_;
auto Complex dummy_;
var_ =  *var_Dummy;
 *(((int *)&(dummy_.re))) = 0;
 *(((int *)&(dummy_.im))) = 0;
dummy_ = var_;
}
/*
gccerr8.c: In function ‘sub_’:
gccerr8.c:6:6: internal compiler error: in expand_debug_locations, at
cfgexpand.c:3865
 void sub_(var_Dummy)
      ^
Please submit a full bug report, with preprocessed source if appropriate.
See <http://bugzilla.redhat.com/bugzilla> for instructions.
Preprocessed source stored into /tmp/ccrC9Ofk.out file, please attach this to
your bugreport.

// /usr/libexec/gcc/x86_64-redhat-linux/4.8.2/cc1 -quiet gccerr8.c -quiet
-dumpbase gccerr8.c -mtune=generic -march=x86-64 -auxbase gccerr8 -g -O1 -o -
-frandom-seed=0
# 1 "gccerr8.c"
# 1 "/home/vitti/f95/cc//"
# 1 "<command-line>"
# 1 "/usr/include/stdc-predef.h" 1 3 4
# 1 "<command-line>" 2
# 1 "gccerr8.c"
typedef struct {
  float re,im;
} Complex;
void sub_(Complex *var_Dummy)
{
Complex var_;
auto Complex dummy_;
var_ = *var_Dummy;
 *(((int *)&(dummy_.re))) = 0;
 *(((int *)&(dummy_.im))) = 0;
dummy_ = var_;
}
*/
>From gcc-bugs-return-440138-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Sun Jan 12 18:02:39 2014
Return-Path: <gcc-bugs-return-440138-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 11622 invoked by alias); 12 Jan 2014 18:02:38 -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 11586 invoked by uid 48); 12 Jan 2014 18:02:35 -0000
From: "mpolacek at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug middle-end/59776] [4.8/4.9 Regression] gcc -g -O1 ICE in expand_debug_locations, at cfgexpand.c:3865
Date: Sun, 12 Jan 2014 18:02:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: middle-end
X-Bugzilla-Version: 4.8.2
X-Bugzilla-Keywords:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: mpolacek at gcc dot gnu.org
X-Bugzilla-Status: NEW
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: 4.8.4
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields: bug_status cf_reconfirmed_on cc component target_milestone short_desc everconfirmed
Message-ID: <bug-59776-4-hbnRXkJFCM@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-59776-4@http.gcc.gnu.org/bugzilla/>
References: <bug-59776-4@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: 2014-01/txt/msg01280.txt.bz2
Content-length: 953

http://gcc.gnu.org/bugzilla/show_bug.cgi?idY776

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2014-01-12
                 CC|                            |mpolacek at gcc dot gnu.org
          Component|c                           |middle-end
   Target Milestone|---                         |4.8.4
            Summary|gcc -g -O1 ICE in           |[4.8/4.9 Regression] gcc -g
                   |expand_debug_locations, at  |-O1 ICE in
                   |cfgexpand.c:3865            |expand_debug_locations, at
                   |                            |cfgexpand.c:3865
     Ever confirmed|0                           |1

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

[Bug middle-end/59776] [4.8/4.9 Regression] gcc -g -O1 ICE in expand_debug_locations, at cfgexpand.c:3865

[mpolacek at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59776

--- Comment #2 from Marek Polacek <mpolacek at gcc dot gnu.org> ---
struct S { float f, g; };

void
sub_ (struct S *p)
{
  struct S s1, s2;
  s1 = *p;
  *(int *) &s2.f = 0;
  s2 = s1;
}

[Bug middle-end/59776] [4.8/4.9 Regression] gcc -g -O1 ICE in expand_debug_locations, at cfgexpand.c:3865

[zeccav at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59776

--- Comment #3 from Vittorio Zecca <zeccav at gmail dot com> ---
Missing right brace at end of code.

[Bug middle-end/59776] [4.8/4.9 Regression] gcc -g -O1 ICE in expand_debug_locations, at cfgexpand.c:3865

[mpolacek at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59776

--- Comment #4 from Marek Polacek <mpolacek at gcc dot gnu.org> ---
(In reply to Vittorio Zecca from comment #3)
> Missing right brace at end of code.

What do you mean?

[Bug middle-end/59776] [4.8/4.9 Regression] gcc -g -O1 ICE in expand_debug_locations, at cfgexpand.c:3865

[zeccav at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59776

--- Comment #5 from Vittorio Zecca <zeccav at gmail dot com> ---
I am sorry I was not clear enough, in your shorter test case, after "s2 = s1;"
there is a right brace "}" missing.

[Bug middle-end/59776] [4.8/4.9 Regression] gcc -g -O1 ICE in expand_debug_locations, at cfgexpand.c:3865

[mpolacek at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59776

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

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

--- Comment #6 from Marek Polacek <mpolacek at gcc dot gnu.org> ---
Started with r195015.

[Bug middle-end/59776] [4.8/4.9 Regression] gcc -g -O1 ICE in expand_debug_locations, at cfgexpand.c:3865

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59776

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P2

[Bug middle-end/59776] [4.8/4.9 Regression] gcc -g -O1 ICE in expand_debug_locations, at cfgexpand.c:3865

[mpolacek at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59776

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P2                          |P1
            Version|4.8.2                       |4.9.0
   Target Milestone|4.8.4                       |4.9.0

[Bug middle-end/59776] [4.8/4.9 Regression] gcc -g -O1 ICE in expand_debug_locations, at cfgexpand.c:3865

[rth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59776

--- Comment #7 from Richard Henderson <rth at gcc dot gnu.org> ---
Definitely SRA's fault.  See

  http://gcc.gnu.org/ml/gcc-patches/2014-02/msg00489.html

for analysis, and the hackiest of patches.

[Bug middle-end/59776] [4.8/4.9 Regression] gcc -g -O1 ICE in expand_debug_locations, at cfgexpand.c:3865

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59776

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

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

--- Comment #8 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Created attachment 32104
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=32104&action=edit
gcc49-pr59776.patch

So, do we want to do this instead?  In this particular case, even no
DW_OP_GNU_reinterpret is needed.

[Bug middle-end/59776] [4.8/4.9 Regression] gcc -g -O1 ICE in expand_debug_locations, at cfgexpand.c:3865

[jamborm at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59776

--- Comment #9 from Martin Jambor <jamborm at gcc dot gnu.org> ---
(In reply to Jakub Jelinek from comment #8)
> Created attachment 32104 [details]
> gcc49-pr59776.patch
> 
> So, do we want to do this instead?  In this particular case, even no
> DW_OP_GNU_reinterpret is needed.

Yes, thanks a lot.

[Bug middle-end/59776] [4.8/4.9 Regression] gcc -g -O1 ICE in expand_debug_locations, at cfgexpand.c:3865

[rth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59776

--- Comment #10 from Richard Henderson <rth at gcc dot gnu.org> ---
(In reply to Jakub Jelinek from comment #8)
> So, do we want to do this instead?  In this particular case, even no
> DW_OP_GNU_reinterpret is needed.

Well, except for the whole dhrs = NULL_TREE thing on the previous line.
Either we need a check for that on the IF, or that line really needs to
be a gcc_unreachable.

[Bug middle-end/59776] [4.8/4.9 Regression] gcc -g -O1 ICE in expand_debug_locations, at cfgexpand.c:3865

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59776

--- Comment #11 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
So:
--- gcc/tree-sra.c.jj   2014-02-08 00:53:46.000000000 +0100
+++ gcc/tree-sra.c      2014-02-11 14:31:51.469937602 +0100
@@ -2950,6 +2950,10 @@ load_assign_lhs_subreplacements (struct
                                                  lacc);
              else
                drhs = NULL_TREE;
+             if (drhs
+                 && !useless_type_conversion_p (lacc->type, TREE_TYPE (drhs)))
+               drhs = fold_build1_loc (loc, VIEW_CONVERT_EXPR,
+                                       lacc->type, drhs);
              ds = gimple_build_debug_bind (get_access_replacement (lacc),
                                            drhs, gsi_stmt (*old_gsi));
              gsi_insert_after (new_gsi, ds, GSI_NEW_STMT);
?

[Bug middle-end/59776] [4.8/4.9 Regression] gcc -g -O1 ICE in expand_debug_locations, at cfgexpand.c:3865

[rth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59776

--- Comment #12 from Richard Henderson <rth at gcc dot gnu.org> ---
Works for me.

[Bug middle-end/59776] [4.8/4.9 Regression] gcc -g -O1 ICE in expand_debug_locations, at cfgexpand.c:3865

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59776

--- Comment #13 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Author: jakub
Date: Tue Feb 11 17:46:34 2014
New Revision: 207695

URL: http://gcc.gnu.org/viewcvs?rev=207695&root=gcc&view=rev
Log:
    PR debug/59776
    * tree-sra.c (load_assign_lhs_subreplacements): Add VIEW_CONVERT_EXPR
    around drhs if type conversion to lacc->type is not useless.

    * gcc.dg/guality/pr59776.c: New test.

Added:
    trunk/gcc/testsuite/gcc.dg/guality/pr59776.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/testsuite/ChangeLog
    trunk/gcc/tree-sra.c

[Bug middle-end/59776] [4.8 Regression] gcc -g -O1 ICE in expand_debug_locations, at cfgexpand.c:3865

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59776

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P1                          |P2
             Status|NEW                         |ASSIGNED
           Assignee|rth at gcc dot gnu.org             |jakub at gcc dot gnu.org
   Target Milestone|4.9.0                       |4.8.3
            Summary|[4.8/4.9 Regression] gcc -g |[4.8 Regression] gcc -g -O1
                   |-O1 ICE in                  |ICE in
                   |expand_debug_locations, at  |expand_debug_locations, at
                   |cfgexpand.c:3865            |cfgexpand.c:3865

--- Comment #14 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Fixed for the trunk so far.

[Bug middle-end/59776] [4.8 Regression] gcc -g -O1 ICE in expand_debug_locations, at cfgexpand.c:3865

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59776

--- Comment #15 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Author: jakub
Date: Thu Mar  6 07:59:29 2014
New Revision: 208366

URL: http://gcc.gnu.org/viewcvs?rev=208366&root=gcc&view=rev
Log:
    Backport from mainline
    2014-02-11  Richard Henderson  <rth@redhat.com>
            Jakub Jelinek  <jakub@redhat.com>

    PR debug/59776
    * tree-sra.c (load_assign_lhs_subreplacements): Add VIEW_CONVERT_EXPR
    around drhs if type conversion to lacc->type is not useless.

    * gcc.dg/guality/pr59776.c: New test.

Added:
    branches/gcc-4_8-branch/gcc/testsuite/gcc.dg/guality/pr59776.c
Modified:
    branches/gcc-4_8-branch/gcc/ChangeLog
    branches/gcc-4_8-branch/gcc/testsuite/ChangeLog
    branches/gcc-4_8-branch/gcc/tree-sra.c

[Bug middle-end/59776] [4.8 Regression] gcc -g -O1 ICE in expand_debug_locations, at cfgexpand.c:3865

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59776

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

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

--- Comment #16 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Fixed.