[Bug bootstrap/63966] New: [5 regression] inconsistent operand constraints compiling build libcpp

[Bug bootstrap/63966] New: [5 regression] inconsistent operand constraints compiling build libcpp

[ro at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63966

            Bug ID: 63966
           Summary: [5 regression] inconsistent operand constraints
                    compiling build libcpp
           Product: gcc
           Version: 5.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: bootstrap
          Assignee: unassigned at gcc dot gnu.org
          Reporter: ro at gcc dot gnu.org
                CC: ubizjak at gmail dot com
              Host: i386-apple-darwin14.0.0
            Target: i386-apple-darwin14.0.0
             Build: i386-apple-darwin14.0.0

Bootstrapping Darwin/i386 (non-default 32-bit configuration) with gcc 4.9.1 
FAILs compiling the stage1 build libcpp:

g++ -m32  -I/vol/gcc/src/hg/trunk/local/libcpp -I.
-I/vol/gcc/src/hg/trunk/local/libcpp/../include
-I/vol/gcc/src/hg/trunk/local/libcpp/include  -g -O2 -W -Wall -Wno-narrowing
-Wwrite-strings -Wmissing-format-attribute -pedantic -Wno-long-long 
-fno-exceptions -fno-rtti -I/vol/gcc/src/hg/trunk/local/libcpp -I.
-I/vol/gcc/src/hg/trunk/local/libcpp/../include
-I/vol/gcc/src/hg/trunk/local/libcpp/include   -c -o lex.o -MT lex.o -MMD -MP
-MF .deps/lex.Tpo /vol/gcc/src/hg/trunk/local/libcpp/lex.c
In file included from /vol/gcc/src/hg/trunk/local/libcpp/lex.c:474:0:
/vol/gcc/src/hg/trunk/local/libcpp/../gcc/config/i386/cpuid.h: In function
'void _cpp_init_lexer()':
/vol/gcc/src/hg/trunk/local/libcpp/../gcc/config/i386/cpuid.h:209:46: error:
inconsistent operand constraints in an 'asm'
   __cpuid (__ext, __eax, __ebx, __ecx, __edx);
                                              ^
/vol/gcc/src/hg/trunk/local/libcpp/../gcc/config/i386/cpuid.h:232:52: error:
inconsistent operand constraints in an 'asm'
   __cpuid (__level, *__eax, *__ebx, *__ecx, *__edx);
                                                    ^
make: *** [lex.o] Error 1

This was caused by the following patch:

2014-10-17  Uros Bizjak  <ubizjak@gmail.com>

    * config/i386/cpuid.h (__cpuid): Remove definitions that handle %ebx
    register in a special way.
    (__cpuid_count): Ditto.
    * config/i386/driver-i386.h: Protect with
    "#if defined(__GNUC__) && (__GNUC__ >= 5 || !defined(__PIC__))".
    (host_detect_local_cpu): Mention that GCC with non-fixed %ebx
    is required to compile the function.

Darwin/x86_64 isn't affected.

  Rainer

[Bug bootstrap/63966] [5 regression] inconsistent operand constraints compiling build libcpp

[ro at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63966

Rainer Orth <ro at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |5.0

[Bug target/63966] [5 regression] inconsistent operand constraints compiling build libcpp

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63966

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|bootstrap                   |target

--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
Huh, I think we should use the host compiler cpuid.h, not the one from the
source
tree we build .... (why do we do that?  we check for using a GCC compiler
reasonably recent)

[Bug target/63966] [5 regression] inconsistent operand constraints compiling build libcpp

[ubizjak at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63966

--- Comment #2 from Uroš Bizjak <ubizjak at gmail dot com> ---
What happens if you remove check for __PIC__ from the #if guard?
>From gcc-bugs-return-467509-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Wed Nov 19 15:03:02 2014
Return-Path: <gcc-bugs-return-467509-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 2223 invoked by alias); 19 Nov 2014 15:03:02 -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 2116 invoked by uid 55); 19 Nov 2014 15:02:53 -0000
From: "ro at CeBiTec dot Uni-Bielefeld.DE" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/63966] [5 regression] inconsistent operand constraints compiling build libcpp
Date: Wed, 19 Nov 2014 15:03:00 -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:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: ro at CeBiTec dot Uni-Bielefeld.DE
X-Bugzilla-Status: UNCONFIRMED
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-63966-4-u9vZv8gcxA@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-63966-4@http.gcc.gnu.org/bugzilla/>
References: <bug-63966-4@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/msg01981.txt.bz2
Content-length: 408

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63966

--- Comment #3 from ro at CeBiTec dot Uni-Bielefeld.DE <ro at CeBiTec dot Uni-Bielefeld.DE> ---
> --- Comment #2 from Uroš Bizjak <ubizjak at gmail dot com> ---
> What happens if you remove check for __PIC__ from the #if guard?

Which guard do you mean?  There's no such guard in either libcpp/lex.c
or cpuid.h, only in driver-i386.c.

    Rainer
>From gcc-bugs-return-467510-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Wed Nov 19 15:07:44 2014
Return-Path: <gcc-bugs-return-467510-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 5156 invoked by alias); 19 Nov 2014 15:07:44 -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 5133 invoked by uid 48); 19 Nov 2014 15:07:40 -0000
From: "ubizjak at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/63966] [5 regression] inconsistent operand constraints compiling build libcpp
Date: Wed, 19 Nov 2014 15:07:00 -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:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: ubizjak at gmail dot com
X-Bugzilla-Status: UNCONFIRMED
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-63966-4-lmY2gCxB6S@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-63966-4@http.gcc.gnu.org/bugzilla/>
References: <bug-63966-4@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/msg01982.txt.bz2
Content-length: 494

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63966

--- Comment #4 from Uroš Bizjak <ubizjak at gmail dot com> ---
(In reply to ro@CeBiTec.Uni-Bielefeld.DE from comment #3)
> > --- Comment #2 from Uroš Bizjak <ubizjak at gmail dot com> ---
> > What happens if you remove check for __PIC__ from the #if guard?
> 
> Which guard do you mean?  There's no such guard in either libcpp/lex.c
> or cpuid.h, only in driver-i386.c.

Oh, I was looking to driver-i386.c, this one is in lex.c
>From gcc-bugs-return-467511-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Wed Nov 19 15:08:16 2014
Return-Path: <gcc-bugs-return-467511-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 5914 invoked by alias); 19 Nov 2014 15:08:16 -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 5866 invoked by uid 48); 19 Nov 2014 15:08:13 -0000
From: "jakub at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/61384] [5 Regression] ICE on valid code at -O3 on x86_64-linux-gnu
Date: Wed, 19 Nov 2014 15:08:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: tree-optimization
X-Bugzilla-Version: 5.0
X-Bugzilla-Keywords:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: jakub at gcc dot gnu.org
X-Bugzilla-Status: RESOLVED
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-61384-4-luFVEA103P@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-61384-4@http.gcc.gnu.org/bugzilla/>
References: <bug-61384-4@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/msg01983.txt.bz2
Content-length: 166

https://gcc.gnu.org/bugzilla/show_bug.cgi?ida384

--- Comment #5 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Seems this got fixed or made latent by r217303.

[Bug target/63966] [5 regression] inconsistent operand constraints compiling build libcpp

[ubizjak at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63966

--- Comment #5 from Uroš Bizjak <ubizjak at gmail dot com> ---
Can you please test this patch:

--cut here--
Index: lex.c
===================================================================
--- lex.c       (revision 217789)
+++ lex.c       (working copy)
@@ -471,7 +471,11 @@

 /* Check the CPU capabilities.  */

+#if (GCC_VERSION >= 5000)
 #include "../gcc/config/i386/cpuid.h"
+#else
+#include <cpuid.h>
+#endif

 typedef const uchar * (*search_line_fast_type) (const uchar *, const uchar *);
 static search_line_fast_type search_line_fast;
--cut here--
>From gcc-bugs-return-467575-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Wed Nov 19 19:29:09 2014
Return-Path: <gcc-bugs-return-467575-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 15497 invoked by alias); 19 Nov 2014 19:29:08 -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 15473 invoked by uid 55); 19 Nov 2014 19:29:05 -0000
From: "dmalcolm at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug jit/63854] Fix memory leaks seen in JIT
Date: Wed, 19 Nov 2014 19:29:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: jit
X-Bugzilla-Version: 5.0
X-Bugzilla-Keywords:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: dmalcolm at gcc dot gnu.org
X-Bugzilla-Status: ASSIGNED
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: dmalcolm at gcc dot gnu.org
X-Bugzilla-Target-Milestone: ---
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields:
Message-ID: <bug-63854-4-GD4eAamukX@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-63854-4@http.gcc.gnu.org/bugzilla/>
References: <bug-63854-4@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/msg02047.txt.bz2
Content-length: 532

https://gcc.gnu.org/bugzilla/show_bug.cgi?idc854

--- Comment #6 from dmalcolm at gcc dot gnu.org ---
Author: dmalcolm
Date: Wed Nov 19 19:28:33 2014
New Revision: 217794

URL: https://gcc.gnu.org/viewcvs?rev!7794&root=gcc&view=rev
Log:
PR jit/63854: Fix memory leak within bb-reorder.c

gcc/ChangeLog:
    PR jit/63854
    * bb-reorder.c
    (find_rarely_executed_basic_blocks_and_crossing_edges): Convert
    local bbs_in_hot_partition from vec<> to auto_vec<>.


Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/bb-reorder.c

[Bug target/63966] [5 regression] inconsistent operand constraints compiling build libcpp

[ro at CeBiTec dot Uni-Bielefeld.DE]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63966

--- Comment #6 from ro at CeBiTec dot Uni-Bielefeld.DE <ro at CeBiTec dot Uni-Bielefeld.DE> ---
> --- Comment #5 from Uroš Bizjak <ubizjak at gmail dot com> ---
> Can you please test this patch:

i386-apple-darwin14.0.0 bootstrap into stage2 now, so seems ok.

Thanks.
        Rainer
>From gcc-bugs-return-467609-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Wed Nov 19 21:14:59 2014
Return-Path: <gcc-bugs-return-467609-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 22264 invoked by alias); 19 Nov 2014 21:14:58 -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 22210 invoked by uid 48); 19 Nov 2014 21:14:54 -0000
From: "davem at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug sanitizer/63958] [5 Regression] bootstrap failure in the sanitizer libs on sparc-linux-gnu
Date: Wed, 19 Nov 2014 21:14:00 -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: 5.0
X-Bugzilla-Keywords:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: davem 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: 5.0
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields: cc
Message-ID: <bug-63958-4-Nw7pwukoyu@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-63958-4@http.gcc.gnu.org/bugzilla/>
References: <bug-63958-4@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/msg02081.txt.bz2
Content-length: 1127

https://gcc.gnu.org/bugzilla/show_bug.cgi?idc958

davem at gcc dot gnu.org changed:

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

--- Comment #2 from davem at gcc dot gnu.org ---
Can you contribute something other than a knee-jerk reactionary posting of that
submission rules URL?

Those requirements are entirely impractical.  I checked things in directly to
the gcc repo because a sparc gcc-4.9 build failure persisted for 9 months,
which is proof that this proceedure for sanitizer development is completely not
tenable.

A sanitizer developer was CC:'d on the gcc bugzilla in question, and I even
posted a working version of that developers patch and attached it to that bug.
Nothing was done for 4 months after that.

After consulting with Richard Henderson I just checked it in.

And I will continue to do the same in the future, I apologize in advance if
fixing the build in gcc after waiting for upstream to do something about it for
months upsets you.

[Bug target/63966] [5 regression] inconsistent operand constraints compiling build libcpp

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63966

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

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

--- Comment #7 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
(In reply to Uroš Bizjak from comment #5)
> Can you please test this patch:
> 
> --cut here--
> Index: lex.c
> ===================================================================
> --- lex.c       (revision 217789)
> +++ lex.c       (working copy)
> @@ -471,7 +471,11 @@
>  
>  /* Check the CPU capabilities.  */
>  
> +#if (GCC_VERSION >= 5000)
>  #include "../gcc/config/i386/cpuid.h"
> +#else
> +#include <cpuid.h>
> +#endif
>  
>  typedef const uchar * (*search_line_fast_type) (const uchar *, const uchar
> *);
>  static search_line_fast_type search_line_fast;
> --cut here--

That assumes there is <cpuid.h> provided by the system compiler.
The code is guarded with
#if (GCC_VERSION >= 4005) && (defined(__i386__) || defined(__x86_64__)) &&
!(defined(__sun__) && defined(__svr4__))
but doesn't clang and icc pretend to be some versions of GCC?
I think better would be to include #include "../gcc/config/i386/cpuid.h"
only if GCC_VERSION >= 5000 (or perhaps that or __PIC__ not defined or
__x86_64__ defined), and otherwise not use __get_cpuid at all, just statically
decide based on minimum value.
>From gcc-bugs-return-467693-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Thu Nov 20 10:21:27 2014
Return-Path: <gcc-bugs-return-467693-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 11603 invoked by alias); 20 Nov 2014 10:21:26 -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 11568 invoked by uid 48); 20 Nov 2014 10:21:23 -0000
From: "rguenth at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/63986] [5 Regression][SH] gcc.target/sh/pr51244-15.c failures
Date: Thu, 20 Nov 2014 10:21:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: tree-optimization
X-Bugzilla-Version: 5.0
X-Bugzilla-Keywords: 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: 5.0
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields: bug_status cc assigned_to
Message-ID: <bug-63986-4-UYKg3vfyzE@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-63986-4@http.gcc.gnu.org/bugzilla/>
References: <bug-63986-4@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/msg02165.txt.bz2
Content-length: 2042

https://gcc.gnu.org/bugzilla/show_bug.cgi?idc986

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |NEW
                 CC|                            |ppalka at gcc dot gnu.org,
                   |                            |rguenth at gcc dot gnu.org
           Assignee|rguenth at gcc dot gnu.org         |unassigned at gcc dot gnu.org

--- Comment #3 from Richard Biener <rguenth at gcc dot gnu.org> ---
Ok, now already existing forwprop code gets fed with

  <bb 2>:
  _3 = a_2(D) == 0;
  x_4 = (char) _3;
  _7 = ~_3;
  _8 = (int) _7;
  MEM[(int *)d_5(D) + 8B] = _8;
  if (x_4 != 0)

where we now in the first forwprop pass identified the opportunity
to use ~_3 instead of x_4 == 0 thus x_4 is now no longer multi-use.
This makes us optimize if (x_4 != 0) to if (_3 != 0) which we
re-optimize in fold_gimple_cond now to '_3' and then of course
if (_3 != 0) (err, and we return "changed"....) which means we
now propagate _again_ via forward_propagate_into_gimple_cond
which now specifically allows aggressive forwarding of compares,
bypassing single-use restrictions.  See

2014-11-16  Patrick Palka  <ppalka@gcc.gnu.org>

        PR middle-end/63790
        * tree-ssa-forwprop.c (forward_propagate_into_comparison_1):
        Always combine comparisons or conversions from booleans.

thus me fixing my "mistake" does not help anymore.

I suppose RTL CSE cannot CSE flag register sets...?

Btw, my previous comment was incorrect - the code is what is now produced
on trunk while on the 4.9 branch we create

test_0:
.LFB0:
        .cfi_startproc
        testl   %edi, %edi
        movl    %edx, %eax
        sete    %r8b
        movl    %r8d, %edi
        xorl    $1, %edi
        testb   %r8b, %r8b
        movzbl  %dil, %edi
        cmovne  %esi, %eax
        movl    %edi, 8(%rcx)
        ret

which means code generation improved for x86...

[Bug target/63966] [5 regression] inconsistent operand constraints compiling build libcpp

[uros at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63966

--- Comment #8 from uros at gcc dot gnu.org ---
Author: uros
Date: Thu Nov 20 10:43:45 2014
New Revision: 217830

URL: https://gcc.gnu.org/viewcvs?rev=217830&root=gcc&view=rev
Log:
    PR target/63966
    * lex.c [__i386__ || __x86_64__]: Include system <cpuid.h> for
    GCC_VERSION < 5000.


Modified:
    trunk/libcpp/ChangeLog
    trunk/libcpp/lex.c

[Bug target/63966] [5 regression] inconsistent operand constraints compiling build libcpp

[ubizjak at gmail dot com]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="UTF-8", Size: 3380 bytes --]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63966

--- Comment #9 from Uroš Bizjak <ubizjak at gmail dot com> ---
(In reply to Jakub Jelinek from comment #7)
> That assumes there is <cpuid.h> provided by the system compiler.
> The code is guarded with
> #if (GCC_VERSION >= 4005) && (defined(__i386__) || defined(__x86_64__)) &&
> !(defined(__sun__) && defined(__svr4__))
> but doesn't clang and icc pretend to be some versions of GCC?
> I think better would be to include #include "../gcc/config/i386/cpuid.h"
> only if GCC_VERSION >= 5000 (or perhaps that or __PIC__ not defined or
> __x86_64__ defined), and otherwise not use __get_cpuid at all, just
> statically decide based on minimum value.

Whops, I didn't see this message. Yes, we can follow the example in
driver-i386.h and just disable specialization for gcc < 5.0
>From gcc-bugs-return-467703-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Thu Nov 20 10:59:31 2014
Return-Path: <gcc-bugs-return-467703-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 17958 invoked by alias); 20 Nov 2014 10:59:31 -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 17920 invoked by uid 48); 20 Nov 2014 10:59:28 -0000
From: "jakub at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug testsuite/61137] [5 regression] FAIL: gcc.target/ia64/small-addr-1.c (test for excess errors)
Date: Thu, 20 Nov 2014 10:59:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: testsuite
X-Bugzilla-Version: 5.0
X-Bugzilla-Keywords:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: jakub at gcc dot gnu.org
X-Bugzilla-Status: NEW
X-Bugzilla-Priority: P1
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: 5.0
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields: cc attachments.created
Message-ID: <bug-61137-4-tno0dk7REp@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-61137-4@http.gcc.gnu.org/bugzilla/>
References: <bug-61137-4@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/msg02175.txt.bz2
Content-length: 711

https://gcc.gnu.org/bugzilla/show_bug.cgi?ida137

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> ---
Created attachment 34051
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id4051&actioníit
gcc5-pr61137.patch

Untested fix (well, tested just on the testcase).  Don't have access to
contemporary ia64 nor cycles for an almost dead architecture, so if Andreas or
anybody else is willing to bootstrap/regtest it, the patch is here.

[Bug target/63966] [5 regression] inconsistent operand constraints compiling build libcpp

[ubizjak at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63966

--- Comment #10 from Uroš Bizjak <ubizjak at gmail dot com> ---
Alternative patch:

--cut here--
Index: lex.c
===================================================================
--- lex.c    (revision 217830)
+++ lex.c    (working copy)
@@ -270,7 +270,7 @@ search_line_acc_char (const uchar *s, const uchar
    extensions used, so SSE4.2 executables cannot run on machines that
    don't support that extension.  */

-#if (GCC_VERSION >= 4005) && (defined(__i386__) || defined(__x86_64__)) &&
!(defined(__sun__) && defined(__svr4__))
+#if (GCC_VERSION >= 4005) && (__GNUC__ >= 5 || !defined(__PIC__)) &&
(defined(__i386__) || defined(__x86_64__)) && !(defined(__sun__) &&
defined(__svr4__))

 /* Replicated character data to be shared between implementations.
    Recall that outside of a context with vector support we can't
--cut here--
>From gcc-bugs-return-467706-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Thu Nov 20 11:03:40 2014
Return-Path: <gcc-bugs-return-467706-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 24360 invoked by alias); 20 Nov 2014 11:03:40 -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 24291 invoked by uid 48); 20 Nov 2014 11:03:36 -0000
From: "fxcoudert at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug sanitizer/63939] [5 Regression] Massive asan failures (356) on darwin
Date: Thu, 20 Nov 2014 11:03:00 -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: 5.0
X-Bugzilla-Keywords:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: fxcoudert at gcc dot gnu.org
X-Bugzilla-Status: RESOLVED
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-63939-4-9FAsTGfF6m@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-63939-4@http.gcc.gnu.org/bugzilla/>
References: <bug-63939-4@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/msg02178.txt.bz2
Content-length: 847

https://gcc.gnu.org/bugzilla/show_bug.cgi?idc939

--- Comment #25 from Francois-Xavier Coudert <fxcoudert at gcc dot gnu.org> ---
(In reply to howarth from comment #22)
> At r217817 on x86_64-apple-darwin14, without llvm-symbolizer in the path but
> with https://gcc.gnu.org/bugzilla/show_bug.cgi?idc534#c50 applied, I still
> see failures for...
> FAIL: c-c++-common/asan/null-deref-1.c   -O1  output pattern test, is
> ASAN:SIGSEGV

I don't see that one (even at -m32, without llvm-symbolizer, and with the patch
you mention), but then my tree is not uniform anymore (revision range
217694:217817M). So it probably appeared recently.

It appears related to optimization, as apparently you don't see it at -O0. If
so, please file a separate PR, as this appears to be a regression and may be a
real issue (and not only a pattern adjustment).

[Bug target/63966] [5 regression] inconsistent operand constraints compiling build libcpp

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63966

--- Comment #11 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Looks ok to me.  Or even add
|| (defined(__x86_64__) && !defined(__code_model_medium__) &&
!defined(__code_model_large__))
to that and to driver-i386.c (right after !defined(__PIC__) ).

[Bug target/63966] [5 regression] inconsistent operand constraints compiling build libcpp

[ubizjak at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63966

--- Comment #12 from Uroš Bizjak <ubizjak at gmail dot com> ---
(In reply to Jakub Jelinek from comment #11)
> Looks ok to me.  Or even add
> || (defined(__x86_64__) && !defined(__code_model_medium__) &&
> !defined(__code_model_large__))
> to that and to driver-i386.c (right after !defined(__PIC__) ).

I don't think we want to complicate things that much... Let's keep the simple
definition for now.
>From gcc-bugs-return-467713-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Thu Nov 20 11:15:10 2014
Return-Path: <gcc-bugs-return-467713-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 15004 invoked by alias); 20 Nov 2014 11:15:10 -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 14458 invoked by uid 48); 20 Nov 2014 11:15:04 -0000
From: "thomas.petazzoni@free-electrons.com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug middle-end/61207] Win64 gcc 4.9.0: ICE at -Os compiling some C++ code
Date: Thu, 20 Nov 2014 11:15:00 -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: 4.9.0
X-Bugzilla-Keywords:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: thomas.petazzoni@free-electrons.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: cc
Message-ID: <bug-61207-4-54lnd6Lyul@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-61207-4@http.gcc.gnu.org/bugzilla/>
References: <bug-61207-4@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/msg02185.txt.bz2
Content-length: 1236

https://gcc.gnu.org/bugzilla/show_bug.cgi?ida207

Thomas Petazzoni <thomas.petazzoni@free-electrons.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |thomas.petazzoni@free-elect
                   |                            |rons.com

--- Comment #2 from Thomas Petazzoni <thomas.petazzoni@free-electrons.com> ---
Interestingly, I have the exact same error while compiling some C++ code in Qt5
on an ARM platform:

svg/SVGPathElement.h: In member function 'virtual WebCore::FloatRect
WebCore::SVGPathElement::_ZTv0_n24_N7WebCore14SVGPathElement7getBBoxENS_12SVGLocatable19StyleUpdateStrategyE(WebCore::SVGLocatable::StyleUpdateStrategy)':
svg/SVGPathElement.h:95:23: internal compiler error: in
expand_expr_addr_expr_1, at expr.c:7669
     virtual FloatRect getBBox(StyleUpdateStrategy = AllowStyleUpdate);
                       ^
Please submit a full bug report,
with preprocessed source if appropriate.

The file is also built with -Os. This is with gcc 4.9.1.

More build output at
http://autobuild.buildroot.org/results/9fd/9fdbe9bb3bd32276f9793fa60d802756811c1abe/build-end.log.

[Bug target/63966] [5 regression] inconsistent operand constraints compiling build libcpp

[uros at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63966

--- Comment #13 from uros at gcc dot gnu.org ---
Author: uros
Date: Thu Nov 20 12:10:12 2014
New Revision: 217851

URL: https://gcc.gnu.org/viewcvs?rev=217851&root=gcc&view=rev
Log:
    PR target/63966
    * lex.c [__i386__ || __x86_64__]: Compile special SSE functions
    only for (__GNUC__ >= 5 || !defined(__PIC__)).


Modified:
    trunk/libcpp/ChangeLog
    trunk/libcpp/lex.c

[Bug target/63966] [5 regression] inconsistent operand constraints compiling build libcpp

[ubizjak at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63966

Uroš Bizjak <ubizjak at gmail dot com> changed:

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

--- Comment #14 from Uroš Bizjak <ubizjak at gmail dot com> ---
Fixed.
>From gcc-bugs-return-467736-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Thu Nov 20 12:20:22 2014
Return-Path: <gcc-bugs-return-467736-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 2821 invoked by alias); 20 Nov 2014 12:20:22 -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 2791 invoked by uid 48); 20 Nov 2014 12:20:19 -0000
From: "rguenth at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug middle-end/63994] New: Ada bootstrap fails with -fcheck-pointer-bounds -mmpx
Date: Thu, 20 Nov 2014 12:20:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: new
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: middle-end
X-Bugzilla-Version: 5.0
X-Bugzilla-Keywords: ice-on-valid-code
X-Bugzilla-Severity: normal
X-Bugzilla-Who: rguenth 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 keywords bug_severity priority component assigned_to reporter cc
Message-ID: <bug-63994-4@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/msg02208.txt.bz2
Content-length: 1459

https://gcc.gnu.org/bugzilla/show_bug.cgi?idc994

            Bug ID: 63994
           Summary: Ada bootstrap fails with -fcheck-pointer-bounds -mmpx
           Product: gcc
           Version: 5.0
            Status: UNCONFIRMED
          Keywords: ice-on-valid-code
          Severity: normal
          Priority: P3
         Component: middle-end
          Assignee: unassigned at gcc dot gnu.org
          Reporter: rguenth at gcc dot gnu.org
                CC: enkovich.gnu at gmail dot com

+===========================GNAT BUG DETECTED==============================+
| 5.0.0 20141119 (experimental) (x86_64-unknown-linux-gnu) GCC error:      |
| chkp_make_addressed_object_bounds: Unexpected tree code                  |
|     array_range_ref                                                      |
| Error detected around
/space/rguenther/src/svn/trunk/gcc/ada/a-elchha.adb:140:4|
| Please submit a bug report; see http://gcc.gnu.org/bugs.html.            |
| Use a subject line meaningful to you and us to track the bug.            |
| Include the entire contents of this bug box in the report.               |
| Include the exact gcc or gnatmake command that you entered.              |
| Also include sources listed below in gnatchop format                     |
| (concatenated together with no headers between files).                   |
+==========================================================================+

and very many similar ICEs.