[Bug target/63171] New: Segfault in rs6000.c:rs6000_emit_move for multiple tls tests starting with r214658

[Bug target/63171] New: Segfault in rs6000.c:rs6000_emit_move for multiple tls tests starting with r214658

[pthaugen at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63171

            Bug ID: 63171
           Summary: Segfault in rs6000.c:rs6000_emit_move for multiple tls
                    tests starting with r214658
           Product: gcc
           Version: 5.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: pthaugen at gcc dot gnu.org
                CC: bergner at gcc dot gnu.org, dje.gcc at gmail dot com,
                    rsandifo at gcc dot gnu.org
              Host: powerpc64-unknown-linux-gnu
            Target: powerpc64-unknown-linux-gnu
             Build: powerpc64-unknown-linux-gnu

Several tls tests from the testsuite are now failing for powerpc64
(https://gcc.gnu.org/ml/gcc-testresults/2014-09/msg00363.html). Bisected to
revision 214658. Following is example failure.

$ ~/install/gcc/trunk/bin/gcc -c -m64
~/src/gcc/trunk/gcc/gcc/testsuite/gcc.dg/tls/opt-11.c
/home/pthaugen/src/gcc/trunk/gcc/gcc/testsuite/gcc.dg/tls/opt-11.c: In function
‘foo’:
/home/pthaugen/src/gcc/trunk/gcc/gcc/testsuite/gcc.dg/tls/opt-11.c:23:10:
internal compiler error: Segmentation fault
   return &a.i;
          ^
0x106e3d67 crash_signal
    /home/pthaugen/src/gcc/trunk/gcc/gcc/toplev.c:339
0x10a70b34 rs6000_emit_move(rtx_def*, rtx_def*, machine_mode)
    /home/pthaugen/src/gcc/trunk/gcc/gcc/config/rs6000/rs6000.c:8659
0x10b6d39f gen_movdi(rtx_def*, rtx_def*)
    /home/pthaugen/src/gcc/trunk/gcc/gcc/config/rs6000/rs6000.md:7748
0x10396f3b insn_gen_fn::operator()(rtx_def*, rtx_def*) const
    /home/pthaugen/src/gcc/trunk/gcc/gcc/recog.h:308
0x10396f3b emit_move_ccmode
    /home/pthaugen/src/gcc/trunk/gcc/gcc/expr.c:3407
0x10396f3b emit_move_insn_1(rtx_def*, rtx_def*)
    /home/pthaugen/src/gcc/trunk/gcc/gcc/expr.c:3564
0x1039739f emit_move_insn(rtx_def*, rtx_def*)
    /home/pthaugen/src/gcc/trunk/gcc/gcc/expr.c:3640
0x1039e743 store_expr(tree_node*, rtx_def*, int, bool)
    /home/pthaugen/src/gcc/trunk/gcc/gcc/expr.c:5498
0x103a054b expand_assignment(tree_node*, tree_node*, bool)
    /home/pthaugen/src/gcc/trunk/gcc/gcc/expr.c:5125
0x10266067 expand_gimple_stmt_1
    /home/pthaugen/src/gcc/trunk/gcc/gcc/cfgexpand.c:3274
0x10266067 expand_gimple_stmt
    /home/pthaugen/src/gcc/trunk/gcc/gcc/cfgexpand.c:3376
0x10267b8b expand_gimple_basic_block
    /home/pthaugen/src/gcc/trunk/gcc/gcc/cfgexpand.c:5215
0x10269a97 execute
    /home/pthaugen/src/gcc/trunk/gcc/gcc/cfgexpand.c:5821
Please submit a full bug report,
>From gcc-bugs-return-461223-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Thu Sep 04 15:10:37 2014
Return-Path: <gcc-bugs-return-461223-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 23974 invoked by alias); 4 Sep 2014 15:10:36 -0000
Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc-bugs.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc-bugs/>
List-Post: <mailto:gcc-bugs@gcc.gnu.org>
List-Help: <mailto:gcc-bugs-help@gcc.gnu.org>
Sender: gcc-bugs-owner@gcc.gnu.org
Delivered-To: mailing list gcc-bugs@gcc.gnu.org
Received: (qmail 23936 invoked by uid 48); 4 Sep 2014 15:10:33 -0000
From: "manu at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug driver/61106] [4.8/4.9] impliedness of -Wunused-parameter depends on -W option ordering
Date: Thu, 04 Sep 2014 15:10:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: driver
X-Bugzilla-Version: 4.8.3
X-Bugzilla-Keywords: diagnostic
X-Bugzilla-Severity: normal
X-Bugzilla-Who: manu 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: 4.8.4
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields: short_desc cf_known_to_fail
Message-ID: <bug-61106-4-luAT31MK5p@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-61106-4@http.gcc.gnu.org/bugzilla/>
References: <bug-61106-4@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: 2014-09/txt/msg01057.txt.bz2
Content-length: 752

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61106

Manuel López-Ibáñez <manu at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[4.8/4.9/5 Regression]      |[4.8/4.9] impliedness of
                   |impliedness of              |-Wunused-parameter depends
                   |-Wunused-parameter depends  |on -W option ordering
                   |on -W option ordering       |
      Known to fail|4.10.0                      |

--- Comment #19 from Manuel López-Ibáñez <manu at gcc dot gnu.org> ---
The patch was never reverted on trunk, so this is only a regression on the
branches.
>From gcc-bugs-return-461225-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Thu Sep 04 15:38:43 2014
Return-Path: <gcc-bugs-return-461225-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 9814 invoked by alias); 4 Sep 2014 15:38:43 -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 9773 invoked by uid 48); 4 Sep 2014 15:38:39 -0000
From: "boger at us dot ibm.com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug go/63172] New: gccgo testcase cplx2.go execution provides incorrect answers on trunk for powerpc64, powerpc64le
Date: Thu, 04 Sep 2014 15:38:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: new
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: go
X-Bugzilla-Version: 5.0
X-Bugzilla-Keywords:
X-Bugzilla-Severity: normal
X-Bugzilla-Who: boger at us dot ibm.com
X-Bugzilla-Status: UNCONFIRMED
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: ian at airs dot com
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 cc
Message-ID: <bug-63172-4@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: 2014-09/txt/msg01059.txt.bz2
Content-length: 3084

https://gcc.gnu.org/bugzilla/show_bug.cgi?idc172

            Bug ID: 63172
           Summary: gccgo testcase cplx2.go execution provides incorrect
                    answers on trunk for powerpc64, powerpc64le
           Product: gcc
           Version: 5.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: go
          Assignee: ian at airs dot com
          Reporter: boger at us dot ibm.com
                CC: bergner at vnet dot ibm.com, cmang at google dot com

Built gccgo from trunk.

Testcase cplx2.go in the go testsuite fails with incorrect answers.

>From the go.log when running 'make check-gcc-go'

opcode x (+7.218935e-001-5.325444e-001i) (+7.218935e-001-5.325444e-001i)
panic: fail

goroutine 16 [running]:
main.main

/home/boger/gccgo.work/orig/trunk-src/gcc/testsuite/go.test/test/ken/cplx2.go:102
created by main
        /home/boger/gccgo.work/orig/trunk-src/libgo/runtime/go-main.c:42
FAIL: go.test/test/ken/cplx2.go execution,  -O2 -g

Looking at an objdump of the executable, the compiler has determined at compile
time that the values are not correct, omits the code to do the conversion and
compare and generates the printlns and panic directly.

        r64 := real(complex128(ce))
        if r64 != real(Ce) {
                println("real(complex128(ce))", r64, real(Ce))
    10001318:   30 00 9f e8     ld      r4,48(r31)
    1000131c:   28 00 7f e8     ld      r3,40(r31)
    10001320:   41 fc ff 4b     bl      10000f60
<0000005f.plt_call.__go_print_string>
    10001324:   18 00 41 e8     ld      r2,24(r1)
    10001328:   09 fc ff 4b     bl      10000f30
<0000005f.plt_call.__go_print_space>
    1000132c:   18 00 41 e8     ld      r2,24(r1)
    10001330:   90 f8 20 fc     fmr     f1,f31
    10001334:   5d fc ff 4b     bl      10000f90
<0000005f.plt_call.__go_print_double>
    10001338:   18 00 41 e8     ld      r2,24(r1)
    1000133c:   f5 fb ff 4b     bl      10000f30
<0000005f.plt_call.__go_print_space>
    10001340:   18 00 41 e8     ld      r2,24(r1)
    10001344:   90 e8 20 fc     fmr     f1,f29
    10001348:   49 fc ff 4b     bl      10000f90
<0000005f.plt_call.__go_print_double>
    1000134c:   18 00 41 e8     ld      r2,24(r1)
    10001350:   d1 fb ff 4b     bl      10000f20
<0000005f.plt_call.__go_print_nl>
    10001354:   18 00 41 e8     ld      r2,24(r1)
/home/boger/gccgo.work/orig/trunk-src/gcc/testsuite/go.test/test/ken/cplx2.go:120
                panic("fail")
    10001358:   10 00 60 38     li      r3,16
    1000135c:   f5 fb ff 4b     bl      10000f50 <0000005f.plt_call.__go_new>
    10001360:   18 00 41 e8     ld      r2,24(r1)
    10001364:   04 00 20 39     li      r9,4
    10001368:   78 1b 64 7c     mr      r4,r3
    1000136c:   00 00 e3 fb     std     r31,0(r3)
    10001370:   08 00 23 f9     std     r9,8(r3)
    10001374:   ff ff 62 3c     addis   r3,r2,-1
    10001378:   98 7c 63 38     addi    r3,r3,31896
    1000137c:   c5 fb ff 4b     bl      10000f40 <0000005f.plt_call.__go_panic>
    10001380:   18 00 41 e8     ld      r2,24(r1)

[Bug target/63171] Segfault in rs6000.c:rs6000_emit_move for multiple tls tests starting with r214658

[rsandifo at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63171

rsandifo at gcc dot gnu.org <rsandifo at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |ASSIGNED
   Last reconfirmed|                            |2014-09-04
           Assignee|unassigned at gcc dot gnu.org      |rsandifo at gcc dot gnu.org
     Ever confirmed|0                           |1

--- Comment #1 from rsandifo at gcc dot gnu.org <rsandifo at gcc dot gnu.org> ---
Testing a fix.

[Bug target/63171] Segfault in rs6000.c:rs6000_emit_move for multiple tls tests starting with r214658

[rsandifo at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63171

--- Comment #2 from rsandifo at gcc dot gnu.org <rsandifo at gcc dot gnu.org> ---
Author: rsandifo
Date: Sat Sep  6 10:04:44 2014
New Revision: 214989

URL: https://gcc.gnu.org/viewcvs?rev=214989&root=gcc&view=rev
Log:
gcc/
    PR middle-end/63171
    * rtlanal.c (tls_referenced_p): Don't skip constant subrtxes.

Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/rtlanal.c

[Bug target/63171] Segfault in rs6000.c:rs6000_emit_move for multiple tls tests starting with r214658

[rsandifo at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63171

rsandifo at gcc dot gnu.org <rsandifo at gcc dot gnu.org> changed:

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

--- Comment #3 from rsandifo at gcc dot gnu.org <rsandifo at gcc dot gnu.org> ---
Patch applied.  Sorry for the breakage.