[Bug tree-optimization/59918] New: [4.9 Regression] ICE in record_target_from_binfo, at ipa-devirt.c:693

[Bug tree-optimization/59918] New: [4.9 Regression] ICE in record_target_from_binfo, at ipa-devirt.c:693

[doko at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59918

            Bug ID: 59918
           Summary: [4.9 Regression] ICE in record_target_from_binfo, at
                    ipa-devirt.c:693
           Product: gcc
           Version: 4.9.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: doko at gcc dot gnu.org

Created attachment 31931
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=31931&action=edit
preprocessed source

seen with trunk 20140122, works with the 4.8 branch. Lowering to -O1 works
around the issue.

$ g++ -c -g -O2 dll.ii 
In file included from seckey.h:8:0,
                 from rijndael.h:7,
                 from aes.h:4,
                 from dll.h:11,
                 from dll.cpp:6:
cryptlib.h: In member function 'bool CryptoPP::NameValuePairs::GetValue(const
char*, T&) const [with T = int]':
cryptlib.h:263:46: internal compiler error: in record_target_from_binfo, at
ipa-devirt.c:693
   return GetVoidValue(name, typeid(T), &value);
                                              ^
Please submit a full bug report,
with preprocessed source if appropriate.

[Bug ipa/59918] [4.9 Regression] ICE in record_target_from_binfo, at ipa-devirt.c:693

[trippels at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59918

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2014-01-23
                 CC|                            |hubicka at gcc dot gnu.org,
                   |                            |trippels at gcc dot gnu.org
          Component|tree-optimization           |ipa
     Ever confirmed|0                           |1

--- Comment #1 from Markus Trippelsdorf <trippels at gcc dot gnu.org> ---
markus@x4 tmp % cat test.ii
class NameValuePairs {
public:
  ~NameValuePairs();
  virtual bool GetVoidValue(const char *, const std::type_info &valueType,
                            void *) const;
};
class B : NameValuePairs {};
class F : virtual B {};
class G : virtual B {};
class ASN1Object {
public:
  virtual ~ASN1Object();
};
template <class, class BASE> class GetValueHelperClass {
public:
  GetValueHelperClass(int *, int *, const std::type_info &valueType, void *,
                      NameValuePairs *searchFirst) {
    searchFirst->GetVoidValue(0, valueType, 0);
  }
  void Assignable();
};
template <class BASE, class T>
GetValueHelperClass<T, int>
GetValueHelper(T, int, const std::type_info &valueType, void *) {
  GetValueHelperClass<T, BASE>(0, 0, valueType, 0, 0);
}
template <class BASE> class C : ASN1Object, BASE {};
class DL_GroupParameters : F {};
template <class GP> class H : C<F> {
  GP m_groupParameters;
};
template <class GP> class DL_PrivateKeyImpl : H<GP> {};
template <class BASE> class D : BASE {};
class ECP;
template <class EC> class DL_GroupParameters_EC : DL_GroupParameters {
  bool GetVoidValue(const char *, const std::type_info &valueType,
                    void *) const;
};
class A : DL_PrivateKeyImpl<DL_GroupParameters_EC<ECP> > {};
template class D<A>;
class I : C<G>, C<F> {};
template <class EC>
bool DL_GroupParameters_EC<EC>::GetVoidValue(const char *,
                                             const std::type_info &valueType,
                                             void *) const {
  GetValueHelper<DL_GroupParameters>(0, 0, valueType, 0).Assignable();
}

markus@x4 tmp % g++ -O2 -c test.ii
test.ii: In constructor ‘GetValueHelperClass< <template-parameter-1-1>,
BASE>::GetValueHelperClass(int*, int*, const std::type_info&, void*,
NameValuePairs*) [with <template-parameter-1-1> = int; BASE =
DL_GroupParameters]’:
test.ii:18:5: internal compiler error: in record_target_from_binfo, at
ipa-devirt.c:693
     searchFirst->GetVoidValue(0, valueType, 0);
     ^
0x9d0f2b record_target_from_binfo
        ../../gcc/gcc/ipa-devirt.c:693
0x9d0d2c record_target_from_binfo
        ../../gcc/gcc/ipa-devirt.c:728
0x9d0d2c record_target_from_binfo
        ../../gcc/gcc/ipa-devirt.c:728
0x9d0d2c record_target_from_binfo
        ../../gcc/gcc/ipa-devirt.c:728
0x9d0d2c record_target_from_binfo
        ../../gcc/gcc/ipa-devirt.c:728
0x9d133b possible_polymorphic_call_targets_1
        ../../gcc/gcc/ipa-devirt.c:755
0x9d13b4 possible_polymorphic_call_targets_1
        ../../gcc/gcc/ipa-devirt.c:762
0x9d13b4 possible_polymorphic_call_targets_1
        ../../gcc/gcc/ipa-devirt.c:762
0x9d13b4 possible_polymorphic_call_targets_1
        ../../gcc/gcc/ipa-devirt.c:762
0x9d3d32 possible_polymorphic_call_targets(tree_node*, long,
ipa_polymorphic_call_context, bool*, void**)
        ../../gcc/gcc/ipa-devirt.c:1346
0x99093b possible_polymorphic_call_targets(tree_node*, bool*, void**)
        ../../gcc/gcc/ipa-utils.h:135
0x98db4c gimple_fold_call
        ../../gcc/gcc/gimple-fold.c:1193
0x98db4c fold_stmt_1
        ../../gcc/gcc/gimple-fold.c:1306
0x9b01e3 gimplify_call_expr
        ../../gcc/gcc/gimplify.c:2427
0x9a3494 gimplify_expr(tree_node**, gimple_statement_base**,
gimple_statement_base**, bool (*)(tree_node*), int)
        ../../gcc/gcc/gimplify.c:7412
0x9a7206 gimplify_stmt(tree_node**, gimple_statement_base**)
        ../../gcc/gcc/gimplify.c:5334
0x9a353a gimplify_cleanup_point_expr
        ../../gcc/gcc/gimplify.c:5110
0x9a353a gimplify_expr(tree_node**, gimple_statement_base**,
gimple_statement_base**, bool (*)(tree_node*), int)
        ../../gcc/gcc/gimplify.c:7804
0x9a7206 gimplify_stmt(tree_node**, gimple_statement_base**)
        ../../gcc/gcc/gimplify.c:5334
0x9a7faa gimplify_bind_expr
        ../../gcc/gcc/gimplify.c:1072
Please submit a full bug report,
>From gcc-bugs-return-441310-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Thu Jan 23 14:32:46 2014
Return-Path: <gcc-bugs-return-441310-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 7269 invoked by alias); 23 Jan 2014 14:32:46 -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 7194 invoked by uid 55); 23 Jan 2014 14:32:41 -0000
From: "ygribov at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug sanitizer/57316] [4.8/4.9 regression] build failure in libsanitizer
Date: Thu, 23 Jan 2014 14:32:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: sanitizer
X-Bugzilla-Version: 4.8.0
X-Bugzilla-Keywords:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: ygribov at gcc dot gnu.org
X-Bugzilla-Status: NEW
X-Bugzilla-Priority: P2
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: 4.8.3
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields:
Message-ID: <bug-57316-4-itK2tvMUyD@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-57316-4@http.gcc.gnu.org/bugzilla/>
References: <bug-57316-4@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/msg02452.txt.bz2
Content-length: 673

http://gcc.gnu.org/bugzilla/show_bug.cgi?idW316

--- Comment #18 from ygribov at gcc dot gnu.org ---
Author: ygribov
Date: Thu Jan 23 14:32:05 2014
New Revision: 206966

URL: http://gcc.gnu.org/viewcvs?rev 6966&root=gcc&view=rev
Log:
2014-01-23  Yury Gribov  <y.gribov@samsung.com>
        Jakub Jelinek  <jakub@redhat.com>

    PR sanitizer/57316
    * configure.ac: Check for missing syscalls.
    * Makefile.am: Likewise.
    * configure: Regenerate.
    * Makefile.in: Regenerate.

Modified:
    trunk/libsanitizer/ChangeLog
    trunk/libsanitizer/Makefile.am
    trunk/libsanitizer/Makefile.in
    trunk/libsanitizer/configure
    trunk/libsanitizer/configure.ac

[Bug ipa/59918] [4.9 Regression] ICE in record_target_from_binfo, at ipa-devirt.c:693

[trippels at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59918

--- Comment #2 from Markus Trippelsdorf <trippels at gcc dot gnu.org> ---
A bit further reduced:

class NameValuePairs {
public:
  ~NameValuePairs();
  virtual bool m_fn1(const char *, const int &p2, void *) const;
};
class B : NameValuePairs {};
class F : virtual public B {};
class G : virtual B {};
class ASN1Object {
public:
  virtual ~ASN1Object();
};
class A {
public:
  NameValuePairs *GetValueHelperClass_searchFirst;
  A() { GetValueHelperClass_searchFirst->m_fn1(0, 0, 0); }
  void m_fn1();
};
template <class, class T> A GetValueHelper(T, int) { A(); }
template <class BASE> class C : ASN1Object, public BASE {};
template <class GP> class H : public C<F> {
  GP m_groupParameters;
};
template <class BASE> class D : BASE {};
template <class EC> class J : F {
  bool m_fn1(const char *, const int &p2, void *) const;
};
template class D<H<J<int> > >;
class I : C<G>, C<F> {};
template <class EC>
bool J<EC>::m_fn1(const char *, const int &p2, void *) const {
  GetValueHelper<F>(0, 0).m_fn1();
}

[Bug ipa/59918] [4.9 Regression] ICE in record_target_from_binfo, at ipa-devirt.c:693

[glisse at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59918

--- Comment #4 from Marc Glisse <glisse at gcc dot gnu.org> ---
(In reply to Markus Trippelsdorf from comment #2)
> A bit further reduced:

I want to play! ;-)

struct E {
  ~E();
  virtual void f() const;
};
struct B : E {};
struct G : virtual B {};
struct A {
  virtual ~A();
};
struct J : E {
  void f() const {
    E *p = 0;
    p->f();
  }
};
J h;
struct I : A, G, virtual B {};

[Bug ipa/59918] [4.9 Regression] ICE in record_target_from_binfo, at ipa-devirt.c:693

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59918

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.9.0

[Bug ipa/59918] [4.9 Regression] ICE in record_target_from_binfo, at ipa-devirt.c:693

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59918

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

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

[Bug ipa/59918] [4.9 Regression] ICE in record_target_from_binfo, at ipa-devirt.c:693

[hubicka at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59918

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Depends on|                            |59947

--- Comment #5 from Jan Hubicka <hubicka at gcc dot gnu.org> ---
Should be the same issue as PR59947

[Bug ipa/59918] [4.9 Regression] ICE in record_target_from_binfo, at ipa-devirt.c:693

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59918
Bug 59918 depends on bug 59947, which changed state.

Bug 59947 Summary: [4.9 Regression] Segmentation fault with #pragma GCC optimize ("O2"), ICE in get_odr_type
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59947

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

[Bug ipa/59918] [4.9 Regression] ICE in record_target_from_binfo, at ipa-devirt.c:693

[trippels at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59918

--- Comment #6 from Markus Trippelsdorf <trippels at gcc dot gnu.org> ---
(In reply to Jan Hubicka from comment #5)
> Should be the same issue as PR59947

No. Still happens after r207512.

[Bug ipa/59918] [4.9 Regression] ICE in record_target_from_binfo, at ipa-devirt.c:693

[hubicka at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59918

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED

--- Comment #7 from Jan Hubicka <hubicka at gcc dot gnu.org> ---
Confirmed. Never ending virtual inheritance fun!

[Bug ipa/59918] [4.9 Regression] ICE in record_target_from_binfo, at ipa-devirt.c:693

[hubicka at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59918

--- Comment #8 from Jan Hubicka <hubicka at gcc dot gnu.org> ---
This is just over-active sanity check
Index: ipa-devirt.c
===================================================================
--- ipa-devirt.c  (revision 207588)
+++ ipa-devirt.c  (working copy)
@@ -689,10 +689,7 @@ record_target_from_binfo (vec <cgraph_no
         we may not have its associated vtable.  This is not a problem, since
         we will walk it on the other path.  */
       if (!type_binfo)
-       {
-         gcc_assert (BINFO_VIRTUAL_P (binfo));
-         return;
-       }
+       return;
       tree inner_binfo = get_binfo_at_offset (type_binfo,
                              offset, otr_type);
       /* For types in anonymous namespace first check if the respective vtable

Are we getting to end of the non-invalid code ipa-devirt ICEs? :)

[Bug ipa/59918] [4.9 Regression] ICE in record_target_from_binfo, at ipa-devirt.c:693

[hubicka at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59918

--- Comment #9 from Jan Hubicka <hubicka at gcc dot gnu.org> ---
Author: hubicka
Date: Fri Feb  7 06:01:36 2014
New Revision: 207592

URL: http://gcc.gnu.org/viewcvs?rev=207592&root=gcc&view=rev
Log:
    PR ipa/59918
    * ipa-devirt.c (record_target_from_binfo): Remove overactive sanity check.
    * g++.dg/torture/pr59918.C: New testcase.

Added:
    trunk/gcc/testsuite/g++.dg/torture/pr59918.C
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/ipa-devirt.c
    trunk/gcc/testsuite/ChangeLog

[Bug ipa/59918] [4.9 Regression] ICE in record_target_from_binfo, at ipa-devirt.c:693

[trippels at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59918

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

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

--- Comment #10 from Markus Trippelsdorf <trippels at gcc dot gnu.org> ---
Fixed by r207592. Closing.