[Bug target/61618] [4.10 Regression]: ICE in expand_vec_perm_pblendv, at config/i386/i386.c with -mavx

[Bug target/61618] [4.10 Regression]: ICE in expand_vec_perm_pblendv, at config/i386/i386.c with -mavx

[ubizjak at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61618

Uroš Bizjak <ubizjak at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Target|                            |x86
   Target Milestone|---                         |4.10.0
>From gcc-bugs-return-455019-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Thu Jun 26 12:34:42 2014
Return-Path: <gcc-bugs-return-455019-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 16317 invoked by alias); 26 Jun 2014 12:34:42 -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 16299 invoked by uid 48); 26 Jun 2014 12:34:38 -0000
From: "ubizjak at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/61618] New: [4.10 Regression]: ICE in expand_vec_perm_pblendv, at config/i386/i386.c with -mavx
Date: Thu, 26 Jun 2014 12:34:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: new
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: target
X-Bugzilla-Version: 4.10.0
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: ---
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-61618-4@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: 2014-06/txt/msg02101.txt.bz2
Content-length: 2239

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61618

            Bug ID: 61618
           Summary: [4.10 Regression]: ICE in expand_vec_perm_pblendv, at
                    config/i386/i386.c with -mavx
           Product: gcc
           Version: 4.10.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: ubizjak at gmail dot com

gcc.dg/torture/vshuf-v8sf.c currently fails with -O2 -mavx:

gcc.dg/torture/vshuf-main.inc: In function ‘test_9’:
gcc.dg/torture/vshuf-main.inc:15:5: internal compiler error: in
expand_vec_perm_pblendv, at config/i386/i386.c:43257
   d = __builtin_shuffle (a, b, mask);  \
     ^
gcc.dg/torture/vshuf-8.inc:12:1: note: in expansion of macro ‘T’
 T (9, 13, 11, 10, 15, 8, 5, 14, 8) \
 ^
gcc.dg/torture/vshuf-main.inc:28:1: note: in expansion of macro ‘TESTS’
 TESTS
 ^
0xc6bd88 expand_vec_perm_pblendv
        /home/uros/gcc-svn/trunk/gcc/config/i386/i386.c:43257
0xc6bd88 ix86_expand_vec_perm_const_1
        /home/uros/gcc-svn/trunk/gcc/config/i386/i386.c:44647
0xca4373 ix86_expand_vec_perm_const(rtx_def**)
        /home/uros/gcc-svn/trunk/gcc/config/i386/i386.c:44768
0xd67b25 gen_vec_perm_constv8sf(rtx_def*, rtx_def*, rtx_def*, rtx_def*)
        /home/uros/gcc-svn/trunk/gcc/config/i386/sse.md:8985
0x9003f8 maybe_expand_insn(insn_code, unsigned int, expand_operand*)
        /home/uros/gcc-svn/trunk/gcc/optabs.c:8352
0x907537 expand_vec_perm_1
        /home/uros/gcc-svn/trunk/gcc/optabs.c:6626
0x9077da expand_vec_perm(machine_mode, rtx_def*, rtx_def*, rtx_def*, rtx_def*)
        /home/uros/gcc-svn/trunk/gcc/optabs.c:6667

(gdb) up
#2  0x0000000000c6bd89 in expand_vec_perm_pblendv (d=0x7fffffffd410) at
/home/uros/gcc-svn/trunk/gcc/config/i386/i386.c:43257
43257     gcc_assert (ok);
(gdb) list
43252
43253     for (i = 0; i < nelt; ++i)
43254       dcopy.perm[i] = d->perm[i] & (nelt - 1);
43255
43256     ok = expand_vec_perm_1 (&dcopy);
43257     gcc_assert (ok);
43258
43259     /* Next we put permuted elements into their positions.  */
43260     dcopy1 = *d;
43261     if (which == 2)
(gdb)
>From gcc-bugs-return-455021-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Thu Jun 26 12:37:31 2014
Return-Path: <gcc-bugs-return-455021-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 19818 invoked by alias); 26 Jun 2014 12:37:31 -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 19773 invoked by uid 48); 26 Jun 2014 12:37:27 -0000
From: "pault at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug fortran/60898] model compile error with gfortran 4.7 and gcc 4.9 on Mac OS 10.9
Date: Thu, 26 Jun 2014 12:37:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: fortran
X-Bugzilla-Version: unknown
X-Bugzilla-Keywords:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: pault at gcc dot gnu.org
X-Bugzilla-Status: WAITING
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-60898-4-SlNrpfK6Y8@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-60898-4@http.gcc.gnu.org/bugzilla/>
References: <bug-60898-4@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: 2014-06/txt/msg02103.txt.bz2
Content-length: 743

https://gcc.gnu.org/bugzilla/show_bug.cgi?id`898

--- Comment #6 from Paul Thomas <pault at gcc dot gnu.org> ---
(In reply to Paul Thomas from comment #5)
> (In reply to Dominique d'Humieres from comment #4)
> > > After providing all the missing 'USE' items:
> >
> > Where did you get them?
>
> Dear Jerry,
>
> Dominique's question is sort of... well, necessary since there are no
> namelist declarations in the source provided.
>
> Could you either reduce the source to a testcase or post the missing bits,
> please?
>
> I presume that, since the problem is with a declaration, the enclosing
> program unit by itself will be enough to generate the ICE.
>
> Cheers
>
> Paul

Duuuuh! From the comgsi website, of course!

Cheers

Paul

[Bug target/61618] [4.10 Regression]: ICE in expand_vec_perm_pblendv, at config/i386/i386.c with -mavx

[ubizjak at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61618

--- Comment #1 from Uroš Bizjak <ubizjak at gmail dot com> ---
Testcase (compile with -O2 -mavx):

--cut here--
typedef float V __attribute__ ((vector_size (32)));
typedef unsigned int VI __attribute__ ((vector_size (32)));

extern V a, b, c, d;

void test_9 (void)
{
  VI mask = { 13, 11, 10, 15, 8, 5, 14, 8 };
  d = __builtin_shuffle (a, b, mask);
}
--cut here--
>From gcc-bugs-return-455026-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Thu Jun 26 12:47:13 2014
Return-Path: <gcc-bugs-return-455026-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 355 invoked by alias); 26 Jun 2014 12:47:13 -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 302 invoked by uid 48); 26 Jun 2014 12:47:09 -0000
From: "rguenth at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug fortran/61459] [4.8/4.9/4.10 Regression] segfault when assigning to allocatable function result from matmul result
Date: Thu, 26 Jun 2014 12:47:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: fortran
X-Bugzilla-Version: 4.9.0
X-Bugzilla-Keywords:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: rguenth at gcc dot gnu.org
X-Bugzilla-Status: NEW
X-Bugzilla-Priority: P4
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: 4.8.4
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields: priority
Message-ID: <bug-61459-4-DnzyTBPzsK@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-61459-4@http.gcc.gnu.org/bugzilla/>
References: <bug-61459-4@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: 2014-06/txt/msg02108.txt.bz2
Content-length: 291

https://gcc.gnu.org/bugzilla/show_bug.cgi?ida459

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P4

[Bug target/61618] [4.10 Regression]: ICE in expand_vec_perm_pblendv, at config/i386/i386.c with -mavx

[evstupac at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61618

Stupachenko Evgeny <evstupac at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |evstupac at gmail dot com

--- Comment #2 from Stupachenko Evgeny <evstupac at gmail dot com> ---
Confirm.
Caused by my changes:

commit 5c0d660cf1fc22a4366730c60707b4d34483ab24
Author: kyukhin <kyukhin@138bc75d-0d04-0410-961f-82ee72b054a4>
Date:   Tue Jun 10 12:40:16 2014 +0000

    gcc/
            * config/i386/i386.c (expand_vec_perm_pblendv): New.
            * config/i386/i386.c (ix86_expand_vec_perm_const_1): Use
            expand_vec_perm_pblendv.



    git-svn-id: svn+ssh://gcc.gnu.org/svn/gcc/trunk@211407
138bc75d-0d04-0410-961f-82ee72b054a4

For 32 bytes AVX there are 1 operand permutations requiring more than 1
instruction. 
So we should restrict the case here.
I'll make a patch for this.

[Bug target/61618] [4.10 Regression]: ICE in expand_vec_perm_pblendv, at config/i386/i386.c with -mavx

[evstupac at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61618

--- Comment #3 from Stupachenko Evgeny <evstupac at gmail dot com> ---
The patch fix the fail:

diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c
index 8046c67..2cffcef 100644
--- a/gcc/config/i386/i386.c
+++ b/gcc/config/i386/i386.c
@@ -43211,12 +43211,10 @@ expand_vec_perm_pblendv (struct expand_vec_perm_d *d)
   bool ok;

   /* Use the same checks as in expand_vec_perm_blend, but skipping
-     AVX2 as it requires more than 2 instructions for general case.  */
+     AVX float case and AVX2 as they require more than 2 instructions.  */
   if (d->one_operand_p)
     return false;
-  if (TARGET_AVX && (vmode == V4DFmode || vmode == V8SFmode))
-    ;
-  else if (TARGET_SSE4_1 && GET_MODE_SIZE (vmode) == 16)
+  if (TARGET_SSE4_1 && GET_MODE_SIZE (vmode) == 16)
     ;
   else
     return false;

Submitted for review:
https://gcc.gnu.org/ml/gcc-patches/2014-07/msg00182.html

[Bug target/61618] [5 Regression]: ICE in expand_vec_perm_pblendv, at config/i386/i386.c with -mavx

[ubizjak at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61618

Uroš Bizjak <ubizjak at gmail dot com> changed:

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

--- Comment #4 from Uroš Bizjak <ubizjak at gmail dot com> ---
Fixed in recent builds.
>From gcc-bugs-return-466837-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Sat Nov 15 08:24:23 2014
Return-Path: <gcc-bugs-return-466837-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 4651 invoked by alias); 15 Nov 2014 08:24:22 -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 4617 invoked by uid 48); 15 Nov 2014 08:24:19 -0000
From: "mpolacek at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug middle-end/63884] ICE: SIGSEGV in is_sec_implicit_index_fn with -fcilkplus and __builtin_sadd_overflow()
Date: Sat, 15 Nov 2014 08:24:00 -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: ice-on-valid-code
X-Bugzilla-Severity: normal
X-Bugzilla-Who: mpolacek at gcc dot gnu.org
X-Bugzilla-Status: NEW
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: 5.0
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields: bug_status cf_reconfirmed_on cc target_milestone everconfirmed
Message-ID: <bug-63884-4-dDMZfOtN5A@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-63884-4@http.gcc.gnu.org/bugzilla/>
References: <bug-63884-4@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/msg01309.txt.bz2
Content-length: 613

https://gcc.gnu.org/bugzilla/show_bug.cgi?idc884

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2014-11-15
                 CC|                            |mpolacek at gcc dot gnu.org
   Target Milestone|---                         |5.0
     Ever confirmed|0                           |1

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