[Bug target/63815] New: g++.dg/other/pr53811.C fails on darwin14

[Bug target/63815] New: g++.dg/other/pr53811.C fails on darwin14

[fxcoudert at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63815

            Bug ID: 63815
           Summary: g++.dg/other/pr53811.C fails on darwin14
           Product: gcc
           Version: 5.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: fxcoudert at gcc dot gnu.org

g++.dg/other/pr53811.C ICE's with trunk (rev. 217343) on darwin14 with
-mcmodel=large. Minimal reproducer is:

------------------------------
struct ICCStringClass
{
  virtual void CreateString (void) = 0;
};

struct AGSCCDynamicObject
{
  virtual void Unserialize (void) = 0;
};


struct ScriptString:AGSCCDynamicObject, ICCStringClass
{
  virtual void CreateString (void);
};

void CreateNewScriptString (void);

void
ScriptString::CreateString (void)
{
  CreateNewScriptString ();
}
------------------------------

which says:

pr53811.C: In member function ‘virtual void
ScriptString::_ZThn8_N12ScriptString12CreateStringEv()’:
pr53811.C:23:1: error: insn does not satisfy its constraints:
 }
 ^
(insn 4 3 5 (parallel [
            (set (reg:DI 39 r10)
                (plus:DI (reg:DI 39 r10)
                    (reg:DI 87)))
            (clobber (reg:CC 17 flags))
        ]) pr53811.C:14 221 {*adddi_1}
     (nil))


and the backtrace is:

  * frame #0: 0x0000000100ee03f0 cc1plus`internal_error(char const*, ...)
    frame #1: 0x0000000100ede966 cc1plus`fancy_abort(char const*, int, char
const*) + 38
    frame #2: 0x0000000100aca9de cc1plus`_fatal_insn(char const*, rtx_def
const*, char const*, int, char const*) + 78
    frame #3: 0x0000000100acaa02 cc1plus`_fatal_insn_not_found(rtx_def const*,
char const*, int, char const*) + 34
    frame #4: 0x00000001007c1509 cc1plus`final_scan_insn(rtx_insn*, __sFILE*,
int, int, int*) + 4953
    frame #5: 0x00000001007c1703 cc1plus`final(rtx_insn*, __sFILE*, int) + 195
    frame #6: 0x0000000100e96838 cc1plus`x86_output_mi_thunk(__sFILE*,
tree_node*, long long, long long, tree_node*) + 872
    frame #7: 0x000000010069f17c cc1plus`cgraph_node::expand_thunk(bool, bool)
+ 2156
    frame #8: 0x00000001006a0163
cc1plus`cgraph_node::assemble_thunks_and_aliases() + 259
    frame #9: 0x00000001006a0386 cc1plus`cgraph_node::expand() + 518
    frame #10: 0x00000001006a11af cc1plus`output_in_order(bool) + 559
    frame #11: 0x00000001006a16fa cc1plus`symbol_table::compile() + 890
    frame #12: 0x00000001006a375e
cc1plus`symbol_table::finalize_compilation_unit() + 110
    frame #13: 0x00000001001062e0 cc1plus`cp_write_global_declarations() + 4480
    frame #14: 0x0000000100b3cebf cc1plus`compile_file() + 143
    frame #15: 0x0000000100ff1ae4 cc1plus`toplev::main(int, char**) + 3764
    frame #16: 0x0000000100ff210c cc1plus`main + 44
>From gcc-bugs-return-466339-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Tue Nov 11 11:23:23 2014
Return-Path: <gcc-bugs-return-466339-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 25999 invoked by alias); 11 Nov 2014 11:23:22 -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 25971 invoked by uid 48); 11 Nov 2014 11:23:19 -0000
From: "fxcoudert at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/63815] g++.dg/other/pr53811.C fails on darwin14
Date: Tue, 11 Nov 2014 11:23:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: target
X-Bugzilla-Version: 5.0
X-Bugzilla-Keywords: ice-on-valid-code
X-Bugzilla-Severity: normal
X-Bugzilla-Who: fxcoudert 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: cf_gcctarget bug_status keywords cf_reconfirmed_on cc cf_gcchost everconfirmed cf_gccbuild
Message-ID: <bug-63815-4-7VtgP6HSSL@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-63815-4@http.gcc.gnu.org/bugzilla/>
References: <bug-63815-4@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: 2014-11/txt/msg00811.txt.bz2
Content-length: 768

https://gcc.gnu.org/bugzilla/show_bug.cgi?idc815

Francois-Xavier Coudert <fxcoudert at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Target|                            |x86_64-apple-darwin14
             Status|UNCONFIRMED                 |NEW
           Keywords|                            |ice-on-valid-code
   Last reconfirmed|                            |2014-11-11
                 CC|                            |iains at gcc dot gnu.org
               Host|                            |x86_64-apple-darwin14
     Ever confirmed|0                           |1
              Build|                            |x86_64-apple-darwin14

[Bug target/63815] g++.dg/other/pr53811.C fails on darwin14

[dominiq at lps dot ens.fr]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63815

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

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

--- Comment #1 from Dominique d'Humieres <dominiq at lps dot ens.fr> ---
This is due to r216154.

[Bug target/63815] g++.dg/other/pr53811.C fails on darwin14

[dominiq at lps dot ens.fr]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63815

--- Comment #2 from Dominique d'Humieres <dominiq at lps dot ens.fr> ---
This appeared between revisions r216154 (OK) and r216633 (ICE).

[Bug target/63815] g++.dg/other/pr53811.C fails on darwin14

[dominiq at lps dot ens.fr]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63815

--- Comment #3 from Dominique d'Humieres <dominiq at lps dot ens.fr> ---
> This appeared between revisions r216154 (OK) and r216633 (ICE).

Wrong PR, sorry.

[Bug target/63815] g++.dg/other/pr53811.C fails with -mcmodel=large -fpic

[hjl.tools at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63815

--- Comment #6 from H.J. Lu <hjl.tools at gmail dot com> ---
Large model thunk with PIC never worked.  GCC 4.8 generates:

_ZThn8_N12ScriptString12CreateStringEPKc:
.LFB1:
    .cfi_startproc
    subq    $8, %rdi
    movabsq    $.LTHUNK0@GOTOFF, %r10
    addq    %rbx, %r10
    jmp    *%r10
    .cfi_endproc


It never sets up GOT base pointer.

[Bug target/63815] g++.dg/other/pr53811.C fails with -mcmodel=large -fpic

[hjl.tools at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63815

--- Comment #7 from H.J. Lu <hjl.tools at gmail dot com> ---
Created attachment 33942
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=33942&action=edit
A patch

[Bug target/63815] g++.dg/other/pr53811.C fails with -mcmodel=large -fpic

[ubizjak at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63815

--- Comment #9 from Uroš Bizjak <ubizjak at gmail dot com> ---
(In reply to Uroš Bizjak from comment #8)
> (In reply to H.J. Lu from comment #7)
> > Created attachment 33942 [details]
> > A patch
> 
> Please also add a testcase that will exercise this functionality.

... runtime testcase ...
>From gcc-bugs-return-466483-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Wed Nov 12 08:43:17 2014
Return-Path: <gcc-bugs-return-466483-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 10437 invoked by alias); 12 Nov 2014 08:43:16 -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 9516 invoked by uid 48); 12 Nov 2014 08:43:06 -0000
From: "ubizjak at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/63815] g++.dg/other/pr53811.C fails with -mcmodel=large -fpic
Date: Wed, 12 Nov 2014 08:43:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: target
X-Bugzilla-Version: 5.0
X-Bugzilla-Keywords: ice-on-valid-code
X-Bugzilla-Severity: normal
X-Bugzilla-Who: ubizjak 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: ---
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields:
Message-ID: <bug-63815-4-p8JhxjMD0S@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-63815-4@http.gcc.gnu.org/bugzilla/>
References: <bug-63815-4@http.gcc.gnu.org/bugzilla/>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/
Auto-Submitted: auto-generated
MIME-Version: 1.0
X-SW-Source: 2014-11/txt/msg00955.txt.bz2
Content-length: 274

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63815

--- Comment #8 from Uroš Bizjak <ubizjak at gmail dot com> ---
(In reply to H.J. Lu from comment #7)
> Created attachment 33942 [details]
> A patch

Please also add a testcase that will exercise this functionality.
>From gcc-bugs-return-466485-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Wed Nov 12 08:57:35 2014
Return-Path: <gcc-bugs-return-466485-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 28564 invoked by alias); 12 Nov 2014 08:57:35 -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 28544 invoked by uid 48); 12 Nov 2014 08:57:31 -0000
From: "trippels at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug preprocessor/63831] New: [5 Regression] r217292 causes segfauls with -MM
Date: Wed, 12 Nov 2014 08:57:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: new
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: preprocessor
X-Bugzilla-Version: 5.0
X-Bugzilla-Keywords:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: trippels at gcc dot gnu.org
X-Bugzilla-Status: UNCONFIRMED
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: bug_id short_desc product version bug_status bug_severity priority component assigned_to reporter
Message-ID: <bug-63831-4@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: 2014-11/txt/msg00957.txt.bz2
Content-length: 1245

https://gcc.gnu.org/bugzilla/show_bug.cgi?idc831

            Bug ID: 63831
           Summary: [5 Regression] r217292 causes segfauls with -MM
           Product: gcc
           Version: 5.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: preprocessor
          Assignee: unassigned at gcc dot gnu.org
          Reporter: trippels at gcc dot gnu.org

Starting with r217292 I get:

trippels@gcc1-power7 test % cat test.cpp
#include <1.h>

trippels@gcc1-power7 test % cat 1.h
#ifndef __has_attribute
#    define __has_attribute(x) 0
#endif

#if  __has_attribute(alloc_size)
#   define U_ALLOC_SIZE_ATTR(X) __attribute__ ((alloc_size(X)))
#endif

trippels@gcc1-power7 test % g++ -I ./ -MM test.cpp
<built-in>: internal compiler error: Segmentation fault
0x109b658b crash_signal
        ../../gcc/gcc/toplev.c:358
0x110b3a90 parse_has_attribute
        ../../gcc/libcpp/expr.c:2165
0x110b3a90 eval_token
        ../../gcc/libcpp/expr.c:1059
0x110b3a90 _cpp_parse_expr
        ../../gcc/libcpp/expr.c:1227
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.

[Bug target/63815] [5 Regression] g++.dg/other/pr53811.C fails with -mcmodel=large -fpic

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63815

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |5.0
            Summary|g++.dg/other/pr53811.C      |[5 Regression]
                   |fails with -mcmodel=large   |g++.dg/other/pr53811.C
                   |-fpic                       |fails with -mcmodel=large
                   |                            |-fpic

[Bug target/63815] [5 Regression] g++.dg/other/pr53811.C fails with -mcmodel=large -fpic

[hjl.tools at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63815

--- Comment #10 from H.J. Lu <hjl.tools at gmail dot com> ---
(In reply to Uroš Bizjak from comment #9)
> (In reply to Uroš Bizjak from comment #8)
> > (In reply to H.J. Lu from comment #7)
> > > Created attachment 33942 [details]
> > > A patch
> > 
> > Please also add a testcase that will exercise this functionality.
> 
> ... runtime testcase ...

Here is a testcase:

[hjl@gnu-6 tmp]$ cat foo.h 
struct ICCStringClass
{
  virtual int CreateString (int) = 0;
};

struct AGSCCDynamicObject
{
  virtual void Unserialize () = 0;
};


struct ScriptString:AGSCCDynamicObject, ICCStringClass
{
  virtual int CreateString (int);
  virtual void Unserialize ();
};
[hjl@gnu-6 tmp]$ cat foo1.cc 
#include "foo.h"

int
__attribute__ ((noinline))
CreateNewScriptString (int fromText, bool reAllocate = true)
{
  return fromText;
}

int
__attribute__ ((noinline))
ScriptString::CreateString (int fromText)
{
  return CreateNewScriptString (fromText);
}

void
__attribute__ ((noinline))
ScriptString::Unserialize ()
{
}
[hjl@gnu-6 tmp]$ cat foo2.cc 
#include "foo.h"

int
main ()
{
  ICCStringClass *x = new ScriptString;
  if (x->CreateString (1) != 1)
    __builtin_abort ();
  return 0;
}
[hjl@gnu-6 tmp]$ g++ -shared -fpic -O2 -mcmodel=large -o libfoo.so foo1.cc 
[hjl@gnu-6 tmp]$ g++ foo2.cc ./libfoo.so 
[hjl@gnu-6 tmp]$ ./a.out 
Segmentation fault
[hjl@gnu-6 tmp]$ 

But I am having a hard time to add it to gcc testsuite.
>From gcc-bugs-return-466525-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Wed Nov 12 14:08:14 2014
Return-Path: <gcc-bugs-return-466525-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 3355 invoked by alias); 12 Nov 2014 14:08:14 -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 3303 invoked by uid 48); 12 Nov 2014 14:08:05 -0000
From: "trippels at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug middle-end/63761] [5 Regression] error: gimple_bb (stmt) is set to a wrong basic block
Date: Wed, 12 Nov 2014 14:08:00 -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: 5.0
X-Bugzilla-Keywords:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: trippels at gcc dot gnu.org
X-Bugzilla-Status: RESOLVED
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: thopre01 at gcc dot gnu.org
X-Bugzilla-Target-Milestone: 5.0
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields: bug_status resolution
Message-ID: <bug-63761-4-79exUCUXJs@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-63761-4@http.gcc.gnu.org/bugzilla/>
References: <bug-63761-4@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: 2014-11/txt/msg00997.txt.bz2
Content-length: 448

https://gcc.gnu.org/bugzilla/show_bug.cgi?idc761

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

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

--- Comment #6 from Markus Trippelsdorf <trippels at gcc dot gnu.org> ---
Fixed. Thanks.

[Bug target/63815] [5 Regression] g++.dg/other/pr53811.C fails with -mcmodel=large -fpic

[ubizjak at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63815

--- Comment #11 from Uroš Bizjak <ubizjak at gmail dot com> ---
(In reply to H.J. Lu from comment #10)

> But I am having a hard time to add it to gcc testsuite.

IMO, the following should be sufficient:

--cut here--
Index: g++.dg/other/pr63815.C
===================================================================
--- g++.dg/other/pr63815.C      (revision 0)
+++ g++.dg/other/pr63815.C      (working copy)
@@ -0,0 +1,50 @@
+// PR target/63815
+// { dg-do run { target { { i?86-*-* x86_64-*-* } && lp64 } } }
+// { dg-options "-mcmodel=large" }
+// { dg-additional-options "-fpic" { target fpic } }
+
+struct ICCStringClass
+{
+  virtual int CreateString (int) = 0;
+};
+
+struct AGSCCDynamicObject
+{
+  virtual void Unserialize () = 0;
+};
+
+struct ScriptString:AGSCCDynamicObject, ICCStringClass
+{
+  virtual int CreateString (int);
+  virtual void Unserialize ();
+};
+
+int
+__attribute__ ((noinline))
+CreateNewScriptString (int fromText, bool reAllocate = true)
+{
+  return fromText;
+}
+
+int
+__attribute__ ((noinline))
+ScriptString::CreateString (int fromText)
+{
+  return CreateNewScriptString (fromText);
+}
+
+void
+__attribute__ ((noinline))
+ScriptString::Unserialize ()
+{
+}
+
+int
+main ()
+{
+  ICCStringClass *x = new ScriptString;
+
+  if (x->CreateString (1) != 1)
+    __builtin_abort ();
+  return 0;
+}
--cut here--

Currently, this test crashes, and when fixed, there is little chance that the
(allocatable) PIC register from main equals r11 in the thunk.
>From gcc-bugs-return-466567-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Wed Nov 12 20:11:07 2014
Return-Path: <gcc-bugs-return-466567-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 30772 invoked by alias); 12 Nov 2014 20:11:07 -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 30713 invoked by uid 48); 12 Nov 2014 20:11:01 -0000
From: "hubicka at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug ipa/63671] [5 Regression] 21% tramp3d-v4 performance hit due to -fdevirtualize
Date: Wed, 12 Nov 2014 20:11:00 -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:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: hubicka at gcc dot gnu.org
X-Bugzilla-Status: ASSIGNED
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: hubicka at gcc dot gnu.org
X-Bugzilla-Target-Milestone: 5.0
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields:
Message-ID: <bug-63671-4-9BRVCmi69O@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-63671-4@http.gcc.gnu.org/bugzilla/>
References: <bug-63671-4@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: 2014-11/txt/msg01039.txt.bz2
Content-length: 1573

https://gcc.gnu.org/bugzilla/show_bug.cgi?idc671

--- Comment #9 from Jan Hubicka <hubicka at gcc dot gnu.org> ---
With early VRP (but also without) the inliner seems to now suffer from extreme
roundoff errors at badness.  With VRP the first uninlined function still has
badness 0:

Considering std::_Bit_reference& std::_Bit_reference::operator=(bool)/797 with
15 size
 to be inlined into <built-in>/47767 in /aux/hubicka/tramp3d-v4.cpp:38764
 Estimated badness is 0, frequency 0.71.
    Badness calculation for <built-in>/47767 -> std::_Bit_reference&
std::_Bit_reference::operator=(bool)/797
      size growth 8, time 5 inline hints: declared_inline
      0: guessed profile. frequency 0.709000, benefit 0.002000%, time w/o
inlining 500006, time w inlining 499996 overall growth 612 (current) 398
(original)
  not inlinable: <built-in>/47767 -> std::_Bit_reference&
std::_Bit_reference::operator=(bool)/797, --param inline-unit-growth limit
reached
   Estimating body: std::basic_ostream<char, _Traits>&
std::operator<<(std::basic_ostream<char, _Traits>&, const char*) [with _Traits
= std::char_traits<char>]/2076
   Known to be false: not inlined, op1 == 0B, op1 changed
   size:7 time:22

so inline decisions are basically random.  I tuned this few times, but it is
hard to balance the fixpoint arithmetic to not get into 0.  The function in
question is very small, but there are too many of them.

I wonder if we can't switch inliner to std::priority_queue and use sreal to
drive the priority queue? Or one can hold fractions and compare in wide int
calculations.

[Bug target/63815] [5 Regression] g++.dg/other/pr53811.C fails with -mcmodel=large -fpic

[iains at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63815

--- Comment #12 from Iain Sandoe <iains at gcc dot gnu.org> ---
if there's a guard for mcmodel=large support, that should be added
[since we have not had time to implement mcmodel=large for Darwin, I would
expect the test to fail there anyway]

[Bug target/63815] [5 Regression] g++.dg/other/pr53811.C fails with -mcmodel=large -fpic

[hjl.tools at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63815

--- Comment #14 from H.J. Lu <hjl.tools at gmail dot com> ---
(In reply to Uroš Bizjak from comment #11)
> 
> Currently, this test crashes, and when fixed, there is little chance that
> the (allocatable) PIC register from main equals r11 in the thunk.

Well, g++ 4.8 and 4.9 -fpic -mcmodel=large always generate

grep _GLOBAL_OFFSET_TABLE_ pr63815.s

    movabsq    $_GLOBAL_OFFSET_TABLE_-.L5, %r11
    movabsq    $_GLOBAL_OFFSET_TABLE_-.L8, %r11
    movabsq    $_GLOBAL_OFFSET_TABLE_-.L10, %r11
    movabsq    $_GLOBAL_OFFSET_TABLE_-.L12, %r11
    movabsq    $_GLOBAL_OFFSET_TABLE_-.L18, %r11

That is why I created libfoo.so since I couldn't get g++ 4.8 and 4.9 to
gail without libfoo.so.
>From gcc-bugs-return-466571-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Wed Nov 12 20:28:02 2014
Return-Path: <gcc-bugs-return-466571-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 15300 invoked by alias); 12 Nov 2014 20:28:02 -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 14650 invoked by uid 48); 12 Nov 2014 20:27:58 -0000
From: "ubizjak at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/63815] [5 Regression] g++.dg/other/pr53811.C fails with -mcmodel=large -fpic
Date: Wed, 12 Nov 2014 20:28:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: target
X-Bugzilla-Version: 5.0
X-Bugzilla-Keywords: ice-on-valid-code
X-Bugzilla-Severity: normal
X-Bugzilla-Who: ubizjak 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-63815-4-CmamzxAS46@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-63815-4@http.gcc.gnu.org/bugzilla/>
References: <bug-63815-4@http.gcc.gnu.org/bugzilla/>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/
Auto-Submitted: auto-generated
MIME-Version: 1.0
X-SW-Source: 2014-11/txt/msg01043.txt.bz2
Content-length: 503

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63815

--- Comment #13 from Uroš Bizjak <ubizjak at gmail dot com> ---
(In reply to Iain Sandoe from comment #12)
> if there's a guard for mcmodel=large support, that should be added
> [since we have not had time to implement mcmodel=large for Darwin, I would
> expect the test to fail there anyway]

There is no guard for mcmodel that I know of, but

// { dg-do run { target { { i?86-*-linux* x86_64-*-linux* } && lp64 } } }

should also work.
>From gcc-bugs-return-466573-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Wed Nov 12 20:43:47 2014
Return-Path: <gcc-bugs-return-466573-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 5420 invoked by alias); 12 Nov 2014 20:43:46 -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 5384 invoked by uid 55); 12 Nov 2014 20:43:42 -0000
From: "paolo at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug c++/60420] [DR 1510] Bogus "error:=?UTF-8?Q? ‘const’ qualifiers cannot be applied to ‘int&’?=" with lambda
Date: Wed, 12 Nov 2014 20:43:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: c++
X-Bugzilla-Version: unknown
X-Bugzilla-Keywords:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: paolo at gcc dot gnu.org
X-Bugzilla-Status: ASSIGNED
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: paolo.carlini at oracle dot com
X-Bugzilla-Target-Milestone: ---
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields:
Message-ID: <bug-60420-4-vb6cUmScyI@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-60420-4@http.gcc.gnu.org/bugzilla/>
References: <bug-60420-4@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: 2014-11/txt/msg01045.txt.bz2
Content-length: 953

https://gcc.gnu.org/bugzilla/show_bug.cgi?id`420

--- Comment #7 from paolo at gcc dot gnu.org <paolo at gcc dot gnu.org> ---
Author: paolo
Date: Wed Nov 12 20:43:09 2014
New Revision: 217444

URL: https://gcc.gnu.org/viewcvs?rev!7444&root=gcc&view=rev
Log:
/cp
2014-11-12  Paolo Carlini  <paolo.carlini@oracle.com>

    DR 1510
    PR c++/60420
    * cp-tree.h (struct cp_decl_specifier_seq): Add decltype_p bool field.
    * decl.c (grokdeclarator): Use it.
    * parser.c (cp_parser_simple_type_specifier): Likewise.
    * pt.c (tsubst, case DECLTYPE_TYPE): Use tf_ignore_bad_quals.

/testsuite
2014-11-12  Paolo Carlini  <paolo.carlini@oracle.com>

    DR 1510
    PR c++/60420
    * g++.dg/cpp0x/decltype61.C: New.

Added:
    trunk/gcc/testsuite/g++.dg/cpp0x/decltype61.C
Modified:
    trunk/gcc/cp/ChangeLog
    trunk/gcc/cp/cp-tree.h
    trunk/gcc/cp/decl.c
    trunk/gcc/cp/parser.c
    trunk/gcc/cp/pt.c
    trunk/gcc/testsuite/ChangeLog

[Bug target/63815] [5 Regression] g++.dg/other/pr53811.C fails with -mcmodel=large -fpic

[hjl at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63815

--- Comment #15 from hjl at gcc dot gnu.org <hjl at gcc dot gnu.org> ---
Author: hjl
Date: Wed Nov 12 21:16:14 2014
New Revision: 217445

URL: https://gcc.gnu.org/viewcvs?rev=217445&root=gcc&view=rev
Log:
Initialize PIC register for large PIC model C++ thunk

gcc/

    PR target/63815
    * config/i386/i386.c (ix86_init_large_pic_reg): New.  Extracted
    from ...
    (ix86_init_pic_reg): Here.  Use ix86_init_large_pic_reg.
    (x86_output_mi_thunk): Set PIC register to %r11.  Call
    ix86_init_large_pic_reg to initialize PIC register.

gcc/testsuite/

    PR target/63815
    * g++.dg/other/pr63815.C: New test.

Added:
    trunk/gcc/testsuite/g++.dg/other/pr63815.C
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/config/i386/i386.c
    trunk/gcc/testsuite/ChangeLog

[Bug target/63815] [5 Regression] g++.dg/other/pr53811.C fails with -mcmodel=large -fpic

[fxcoudert at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63815

Francois-Xavier Coudert <fxcoudert at gcc dot gnu.org> changed:

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

--- Comment #17 from Francois-Xavier Coudert <fxcoudert at gcc dot gnu.org> ---
Yes, fixed indeed: https://gcc.gnu.org/ml/gcc-testresults/2014-11/msg01533.html