[Bug ipa/65722] New: [5 Regression] ICE: tree check: expected function_decl, have var_decl in equals_wpa, at ipa-icf.c:1679 with -fipa-icf -fno-rtti

[Bug ipa/65722] New: [5 Regression] ICE: tree check: expected function_decl, have var_decl in equals_wpa, at ipa-icf.c:1679 with -fipa-icf -fno-rtti

[zsojka at seznam dot cz]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65722

            Bug ID: 65722
           Summary: [5 Regression] ICE: tree check: expected
                    function_decl, have var_decl in equals_wpa, at
                    ipa-icf.c:1679 with -fipa-icf -fno-rtti
           Product: gcc
           Version: 5.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: ipa
          Assignee: unassigned at gcc dot gnu.org
          Reporter: zsojka at seznam dot cz

Created attachment 35280
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=35280&action=edit
reduced testcase

Compiler output:
$ gcc -O -fipa-icf -fno-rtti testcase.C 
testcase.C:18:1: internal compiler error: tree check: expected function_decl,
have var_decl in equals_wpa, at ipa-icf.c:1679
 }
 ^
0x10802bc tree_check_failed(tree_node const*, char const*, int, char const*,
...)
        /mnt/svn/gcc-trunk/gcc/tree.c:9297
0x1893327 tree_check(tree_node*, char const*, int, char const*, tree_code)
        /mnt/svn/gcc-trunk/gcc/tree.h:2846
0x1893327 ipa_icf::sem_variable::equals_wpa(ipa_icf::sem_item*,
hash_map<symtab_node*, ipa_icf::sem_item*, default_hashmap_traits>&)
        /mnt/svn/gcc-trunk/gcc/ipa-icf.c:1679
0x188f0b2 ipa_icf::sem_item_optimizer::subdivide_classes_by_equality(bool)
        /mnt/svn/gcc-trunk/gcc/ipa-icf.c:2657
0x189a4ea ipa_icf::sem_item_optimizer::execute()
        /mnt/svn/gcc-trunk/gcc/ipa-icf.c:2405
0x189bf56 ipa_icf_driver
        /mnt/svn/gcc-trunk/gcc/ipa-icf.c:3306
0x189bf56 ipa_icf::pass_ipa_icf::execute(function*)
        /mnt/svn/gcc-trunk/gcc/ipa-icf.c:3353
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.

$ gcc -v                                
Using built-in specs.
COLLECT_GCC=/mnt/svn/gcc-trunk/binary-latest/bin/gcc
COLLECT_LTO_WRAPPER=/mnt/svn/gcc-trunk/binary-221940-lto-fortran-checking-yes-rtl-df/libexec/gcc/x86_64-unknown-linux-gnu/5.0.0/lto-wrapper
Target: x86_64-unknown-linux-gnu
Configured with: /mnt/svn/gcc-trunk//configure --enable-checking=yes,rtl,df
--enable-languages=c,c++,lto,fortran
--prefix=/mnt/svn/gcc-trunk/binary-221940-lto-fortran-checking-yes-rtl-df/
--without-cloog --without-ppl
Thread model: posix
gcc version 5.0.0 20150409 (experimental) (GCC) 

Marking as regression since "-O2 -fno-rtti" triggers the issue.

[Bug ipa/65722] [5 Regression] ICE: tree check: expected function_decl, have var_decl in equals_wpa, at ipa-icf.c:1679 with -fipa-icf -fno-rtti

[trippels at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65722

Markus Trippelsdorf <trippels at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2015-04-09
                 CC|                            |marxin at gcc dot gnu.org,
                   |                            |trippels at gcc dot gnu.org
     Ever confirmed|0                           |1

--- Comment #1 from Markus Trippelsdorf <trippels at gcc dot gnu.org> ---
Confirmed.

[Bug ipa/65722] [5 Regression] ICE: tree check: expected function_decl, have var_decl in equals_wpa, at ipa-icf.c:1679 with -fipa-icf -fno-rtti

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65722

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |ice-checking
           Priority|P3                          |P1
   Target Milestone|---                         |5.0

--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> ---
Looks like compare_cgraph_references is incomplete.

[Bug ipa/65722] [5 Regression] ICE: tree check: expected function_decl, have var_decl in equals_wpa, at ipa-icf.c:1679 with -fipa-icf -fno-rtti

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65722

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

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

--- Comment #3 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Started with r221576.

[Bug ipa/65722] [5 Regression] ICE: tree check: expected function_decl, have var_decl in equals_wpa, at ipa-icf.c:1679 with -fipa-icf -fno-rtti

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65722

--- Comment #4 from Martin Liška <marxin at gcc dot gnu.org> ---
I've just finished patch and launched regression tests.
>From gcc-bugs-return-483268-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Fri Apr 10 09:34:37 2015
Return-Path: <gcc-bugs-return-483268-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 55606 invoked by alias); 10 Apr 2015 09:34:36 -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 55548 invoked by uid 48); 10 Apr 2015 09:34:33 -0000
From: "rguenth at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug ipa/65701] r221530 makes 187.facerec drop with -Ofast -flto
Date: Fri, 10 Apr 2015 09:34:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: ipa
X-Bugzilla-Version: 5.0
X-Bugzilla-Keywords: lto, missed-optimization
X-Bugzilla-Severity: normal
X-Bugzilla-Who: rguenth 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: ---
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields:
Message-ID: <bug-65701-4-6cKwZtLYaj@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-65701-4@http.gcc.gnu.org/bugzilla/>
References: <bug-65701-4@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: 2015-04/txt/msg00820.txt.bz2
Content-length: 1314

https://gcc.gnu.org/bugzilla/show_bug.cgi?ide701

--- Comment #13 from Richard Biener <rguenth at gcc dot gnu.org> ---
I am patching the head-64 and head-64-2006 testers with

Index: gcc/tree-vect-data-refs.c
==================================================================--- gcc/tree-vect-data-refs.c   (revision 221942)
+++ gcc/tree-vect-data-refs.c   (working copy)
@@ -1618,9 +1608,7 @@ vect_enhance_data_refs_alignment (loop_v
          peeling only if it may help to align other accesses in the loop.  */
       if (!first_store
          && !STMT_VINFO_SAME_ALIGN_REFS (
-                 vinfo_for_stmt (DR_STMT (dr0))).length ()
-          && vect_supportable_dr_alignment (dr0, false)
-              != dr_unaligned_supported)
+                 vinfo_for_stmt (DR_STMT (dr0))).length ())
         do_peeling = false;
     }


which makes the code match the comment (and which should prevent us
peeling loads for alignment in loops with no stores and when that peeling
only aligns a single load).  Full comment quoted:

      /* In case there are only loads with different unknown misalignments, use
         peeling only if it may help to align other accesses in the loop.  */

I suppose the test was meant to be != dr_unaligned_unsupported but that's
already checked in the guard of this block.

[Bug ipa/65722] [5 Regression] ICE: tree check: expected function_decl, have var_decl in equals_wpa, at ipa-icf.c:1679 with -fipa-icf -fno-rtti

[hubicka at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65722

--- Comment #5 from Jan Hubicka <hubicka at gcc dot gnu.org> ---
Author: hubicka
Date: Sun Apr 12 01:08:04 2015
New Revision: 222015

URL: https://gcc.gnu.org/viewcvs?rev=222015&root=gcc&view=rev
Log:

    Jan Hubicka  <hubicka@ucw.cz>
    Martin Liska  <mliska@suse.cz>

    PR ipa/65722
    * g++.dg/ipa/pr65722.C: New testcase.

    * ipa-icf.c (sem_item::compare_cgraph_references): function and
    variable can not match.
    (sem_item::update_hash_by_addr_refs): Fix handling of virtual tables.
    (sem_variable::equals_wpa): Fix checking of DECL_FINAL_P patch.

Added:
    trunk/gcc/testsuite/g++.dg/ipa/pr65722.C
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/ipa-icf.c
    trunk/gcc/testsuite/ChangeLog

[Bug ipa/65722] [5 Regression] ICE: tree check: expected function_decl, have var_decl in equals_wpa, at ipa-icf.c:1679 with -fipa-icf -fno-rtti

[hubicka at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65722

Jan Hubicka <hubicka at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |hubicka at gcc dot gnu.org
         Resolution|---                         |FIXED

--- Comment #6 from Jan Hubicka <hubicka at gcc dot gnu.org> ---
Fixed.