[Bug target/64300] New: s390x, ICE, unable to generate reloads, in curr_insn_transform, at lra-constraints.c

[Bug target/64300] New: s390x, ICE, unable to generate reloads, in curr_insn_transform, at lra-constraints.c

[jan.kratochvil at redhat dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64300

            Bug ID: 64300
           Summary: s390x, ICE, unable to generate reloads, in
                    curr_insn_transform, at lra-constraints.c
           Product: gcc
           Version: 5.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: jan.kratochvil at redhat dot com
              Host: x86_64-linux-gnu
            Target: s390x-linux-gnu

Created attachment 34274
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=34274&action=edit
libgcc2.i

../gcchead/configure --prefix=/home/jkratoch/redhat/gccheads390x-root
--enable-languages=c --target=s390x-linux-gnu

I get the same ICE with:
CFLAGS=-g CXXFLAGS=-g CFLAGS_FOR_TARGET=-g CXXFLAGS_FOR_TARGET=-g
../gcchead/configure --prefix=/home/jkratoch/redhat/gccheads390x-root
--enable-languages=c --target=s390x-linux-gnu

r218702 | ian | 2014-12-13 02:58:46 +0100 (Sat, 13 Dec 2014) | 3 lines

/home/jkratoch/redhat/gccheads390x-build/./gcc/xgcc
-B/home/jkratoch/redhat/gccheads390x-build/./gcc/
-B/home/jkratoch/redhat/gccheads390x-root/s390x-linux-gnu/bin/
-B/home/jkratoch/redhat/gccheads390x-root/s390x-linux-gnu/lib/ -isystem
/home/jkratoch/redhat/gccheads390x-root/s390x-linux-gnu/include -isystem
/home/jkratoch/redhat/gccheads390x-root/s390x-linux-gnu/sys-include    -g -O2
-m31 -O2  -g -O2 -DIN_GCC  -DCROSS_DIRECTORY_STRUCTURE  -W -Wall -Wno-narrowing
-Wwrite-strings -Wcast-qual -Wstrict-prototypes -Wmissing-prototypes
-Wold-style-definition  -isystem ./include   -fPIC -mlong-double-128 -g
-DIN_LIBGCC2 -fbuilding-libgcc -fno-stack-protector -Dinhibit_libc  -fPIC
-mlong-double-128 -I. -I. -I../../.././gcc -I../../../../gcchead/libgcc
-I../../../../gcchead/libgcc/. -I../../../../gcchead/libgcc/../gcc
-I../../../../gcchead/libgcc/../include  -DHAVE_CC_TLS  -o _addvdi3.o -MT
_addvdi3.o -MD -MP -MF _addvdi3.dep -DL_addvdi3 -c
../../../../gcchead/libgcc/libgcc2.c -fvisibility=hidden -DHIDE_EXPORTS
-save-temps
../../../../gcchead/libgcc/libgcc2.c: In function ‘__addvdi3’:
../../../../gcchead/libgcc/libgcc2.c:109:1: error: unable to generate reloads
for:
 }
 ^
(insn 7 3 8 2 (parallel [
            (set (reg/v:DI 48 [ w ])
                (plus:DI (reg/v:DI 50 [ a ])
                    (reg/v:DI 51 [ b ])))
            (clobber (reg:CC 33 %cc))
        ]) ../../../../gcchead/libgcc/libgcc2.c:103 324 {*adddi3_31}
     (expr_list:REG_UNUSED (reg:CC 33 %cc)
        (nil)))
../../../../gcchead/libgcc/libgcc2.c:109:1: internal compiler error: in
curr_insn_transform, at lra-constraints.c:3383
0x9aa548 _fatal_insn(char const*, rtx_def const*, char const*, int, char
const*)
        ../../gcchead/gcc/rtl-error.c:110
0x8df546 curr_insn_transform
        ../../gcchead/gcc/lra-constraints.c:3383
0x8e026e lra_constraints(bool)
        ../../gcchead/gcc/lra-constraints.c:4324
0x8d0621 lra(_IO_FILE*)
        ../../gcchead/gcc/lra.c:2277
0x891909 do_reload
        ../../gcchead/gcc/ira.c:5391
0x891909 execute
        ../../gcchead/gcc/ira.c:5561
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.
>From gcc-bugs-return-470550-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Sat Dec 13 18:43:24 2014
Return-Path: <gcc-bugs-return-470550-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 28319 invoked by alias); 13 Dec 2014 18:43:24 -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 28285 invoked by uid 48); 13 Dec 2014 18:43:21 -0000
From: "ville.voutilainen at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug c++/61648] g++ accepts literal operator friends for non-template classes but not for *template* classes
Date: Sat, 13 Dec 2014 18:43:00 -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: 5.0
X-Bugzilla-Keywords: rejects-valid
X-Bugzilla-Severity: normal
X-Bugzilla-Who: ville.voutilainen 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: keywords bug_status cf_reconfirmed_on cc everconfirmed cf_known_to_fail
Message-ID: <bug-61648-4-ssdD8GGDSc@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-61648-4@http.gcc.gnu.org/bugzilla/>
References: <bug-61648-4@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: 2014-12/txt/msg01557.txt.bz2
Content-length: 628

https://gcc.gnu.org/bugzilla/show_bug.cgi?ida648

Ville Voutilainen <ville.voutilainen at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |rejects-valid
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2014-12-13
                 CC|                            |ville.voutilainen at gmail dot com
     Ever confirmed|0                           |1
      Known to fail|                            |4.8.2, 4.9.1, 5.0

[Bug target/64300] s390x, ICE, unable to generate reloads, in curr_insn_transform, at lra-constraints.c

[kkojima at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64300

Kazumoto Kojima <kkojima at gcc dot gnu.org> changed:

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

--- Comment #1 from Kazumoto Kojima <kkojima at gcc dot gnu.org> ---
I get the same ICE on my local sh-lra branch after r218688.  I've looked
at what is going on the test case in c#0 with the cross s390 compiler.
It seems that the change at r218688

+          if (GET_MODE (*curr_id->operand_loc[nop]) != VOIDmode
+          && ! hard_reg_set_empty_p (this_alternative_set)
+          && ! HARD_REGNO_MODE_OK (ira_class_hard_regs
+                       [this_alternative][0],
+                       GET_MODE (*curr_id->operand_loc[nop])))

doesn't work as intended for some targets.  These lines are to check
whether a requested mode value can be held by the registers in
this_alternative class or not.  In the problematic case, gdb shows
that GET_MODE (*curr_id->operand_loc[nop]) is DImode and this_alternative
is GENERAL_REGS.  ira_class_hard_regs[this_alternative] is
  {1, 2, 3, 4, 5, 0, 11, 10, 9, 8, 7, 6, 0 <repeats 26 times>}
which is a set of regno for GENERAL_REGS in the preferred allocation order.
Then ira_class_hard_regs[this_alternative][0] is 1 and HARD_REGNO_MODE_OK
(1, DImode) is false.  I guess that s390 uses a register pair for DImode
in this case and 1 is bad as the starting regno for DImode.  Is it right?
SH uses the similar allocation order of which the first regno isn't match
to the register pair.

[Bug rtl-optimization/64300] [5 Regression] s390x, ICE, unable to generate reloads, in curr_insn_transform, at lra-constraints.c

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64300

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |ra
                 CC|                            |vmakarov at gcc dot gnu.org
          Component|target                      |rtl-optimization
   Target Milestone|---                         |5.0
            Summary|s390x, ICE, unable to       |[5 Regression] s390x, ICE,
                   |generate reloads, in        |unable to generate reloads,
                   |curr_insn_transform, at     |in curr_insn_transform, at
                   |lra-constraints.c           |lra-constraints.c

[Bug rtl-optimization/64300] [5 Regression] s390x, ICE, unable to generate reloads, in curr_insn_transform, at lra-constraints.c

[vmakarov at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64300

--- Comment #2 from Vladimir Makarov <vmakarov at gcc dot gnu.org> ---
(In reply to Kazumoto Kojima from comment #1)
> I get the same ICE on my local sh-lra branch after r218688.  I've looked
> at what is going on the test case in c#0 with the cross s390 compiler.
> It seems that the change at r218688
> 
> +	      if (GET_MODE (*curr_id->operand_loc[nop]) != VOIDmode
> +		  && ! hard_reg_set_empty_p (this_alternative_set)
> +		  && ! HARD_REGNO_MODE_OK (ira_class_hard_regs
> +					   [this_alternative][0],
> +					   GET_MODE (*curr_id->operand_loc[nop])))
> 
> doesn't work as intended for some targets.  These lines are to check
> whether a requested mode value can be held by the registers in
> this_alternative class or not.  In the problematic case, gdb shows
> that GET_MODE (*curr_id->operand_loc[nop]) is DImode and this_alternative
> is GENERAL_REGS.  ira_class_hard_regs[this_alternative] is
>   {1, 2, 3, 4, 5, 0, 11, 10, 9, 8, 7, 6, 0 <repeats 26 times>}
> which is a set of regno for GENERAL_REGS in the preferred allocation order.
> Then ira_class_hard_regs[this_alternative][0] is 1 and HARD_REGNO_MODE_OK
> (1, DImode) is false.  I guess that s390 uses a register pair for DImode
> in this case and 1 is bad as the starting regno for DImode.  Is it right?
> SH uses the similar allocation order of which the first regno isn't match
> to the register pair.

Thanks for reporting.  I've just committed a patch focusing on the same
problem.  Could you check that the patch solves the problem.
>From gcc-bugs-return-470811-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org Mon Dec 15 20:25:47 2014
Return-Path: <gcc-bugs-return-470811-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Delivered-To: listarch-gcc-bugs@gcc.gnu.org
Received: (qmail 12250 invoked by alias); 15 Dec 2014 20:25:47 -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 12216 invoked by uid 48); 15 Dec 2014 20:25:43 -0000
From: "jakub at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug rtl-optimization/64316] [5 Regression] ICE in simplify_const_unary_operation after r218503
Date: Mon, 15 Dec 2014 20:25:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: rtl-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: jakub at gcc dot gnu.org
X-Bugzilla-Target-Milestone: 5.0
X-Bugzilla-Flags:
X-Bugzilla-Changed-Fields: bug_status resolution
Message-ID: <bug-64316-4-J1Ivo7TxA6@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-64316-4@http.gcc.gnu.org/bugzilla/>
References: <bug-64316-4@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: 2014-12/txt/msg01818.txt.bz2
Content-length: 422

https://gcc.gnu.org/bugzilla/show_bug.cgi?idd316

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

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

--- Comment #4 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Fixed.

[Bug rtl-optimization/64300] [5 Regression] s390x, ICE, unable to generate reloads, in curr_insn_transform, at lra-constraints.c

[kkojima at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64300

--- Comment #3 from Kazumoto Kojima <kkojima at gcc dot gnu.org> ---
(In reply to Vladimir Makarov from comment #2)
> Thanks for reporting.  I've just committed a patch focusing on the same
> problem.  Could you check that the patch solves the problem.

It fixes the ICE for the given test case on s390 and the similar
ICEs on my local sh-lra branch.  Thanks!

[Bug rtl-optimization/64300] [5 Regression] s390x, ICE, unable to generate reloads, in curr_insn_transform, at lra-constraints.c

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64300

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

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

[Bug rtl-optimization/64300] [5 Regression] s390x, ICE, unable to generate reloads, in curr_insn_transform, at lra-constraints.c

[kkojima at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64300

--- Comment #4 from Kazumoto Kojima <kkojima at gcc dot gnu.org> ---
I think this has been addressed with r218760 by Makarov.  Can we
close this as resolved?

[Bug rtl-optimization/64300] [5 Regression] s390x, ICE, unable to generate reloads, in curr_insn_transform, at lra-constraints.c

[jan.kratochvil at redhat dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64300

Jan Kratochvil <jan.kratochvil at redhat dot com> changed:

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

--- Comment #5 from Jan Kratochvil <jan.kratochvil at redhat dot com> ---
libgcc2.i from Comment 0 builds for me now, thanks.