[Bug other/53961] New: internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[Bug other/53961] New: internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[wbrana at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

             Bug #: 53961
           Summary: internal compiler error: in memory_address_length, at
                    config/i386/i386.c:23341
    Classification: Unclassified
           Product: gcc
           Version: 4.7.1
            Status: UNCONFIRMED
          Severity: major
          Priority: P3
         Component: other
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: wbrana@gmail.com


see https://bugs.gentoo.org/show_bug.cgi?id=426572

[Bug other/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[steven at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

Steven Bosscher <steven at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |WAITING
   Last reconfirmed|                            |2012-07-14
                 CC|                            |steven at gcc dot gnu.org
     Ever Confirmed|0                           |1

--- Comment #1 from Steven Bosscher <steven at gcc dot gnu.org> 2012-07-14 14:46:40 UTC ---
Please provide a test case, not a link.

[Bug other/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[wbrana at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

--- Comment #2 from wbrana <wbrana at gmail dot com> 2012-07-14 15:14:18 UTC ---
Created attachment 27788
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=27788
source code

/usr/bin/gcc:     file format elf32-x86-64

 gcc -m64 -O2 -c ./host_ppc_isel.i -o test.o
In file included from priv/host_ppc_isel.c:37:0:
../VEX/pub/libvex_basictypes.h: In function 'Ptr_to_ULong':
../VEX/pub/libvex_basictypes.h:195:16: warning: cast from pointer to integer of
different size [-Wpointer-to-int-cast]
../VEX/pub/libvex_basictypes.h: In function 'ULong_to_Ptr':
../VEX/pub/libvex_basictypes.h:200:14: warning: cast to pointer from integer of
different size [-Wint-to-pointer-cast]
priv/host_ppc_isel.c: In function 'iselInt64Expr':
priv/host_ppc_isel.c:2834:1: internal compiler error: in memory_address_length,
at config/i386/i386.c:23341

[Bug target/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[ubizjak at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

--- Comment #3 from Uros Bizjak <ubizjak at gmail dot com> 2012-07-14 19:13:57 UTC ---
The testcase works for me with:

GNU C (GCC) version 4.7.2 20120713 (prerelease) [gcc-4_7-branch revision
189471] (x86_64-unknown-linux-gnu)

and 4.8.0.

[Bug target/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[wbrana at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

--- Comment #4 from wbrana <wbrana at gmail dot com> 2012-07-14 19:22:16 UTC ---
Are you using X32 GCC binary?

[Bug target/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[pinskia at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

--- Comment #5 from Andrew Pinski <pinskia at gcc dot gnu.org> 2012-07-14 20:13:18 UTC ---
(In reply to comment #4)
> Are you using X32 GCC binary?

So -mx32 makes it fail or rather GCC defaulting to x32 makes it fail?

[Bug target/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[wbrana at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

--- Comment #6 from wbrana <wbrana at gmail dot com> 2012-07-14 20:17:23 UTC ---
if gcc is compiled as X32 
/usr/bin/gcc:     file format elf32-x86-64

and command 
gcc -m64 -O2 -c ./host_ppc_isel.i -o test.o

[Bug target/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[hjl.tools at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

H.J. Lu <hjl.tools at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|WAITING                     |NEW

--- Comment #7 from H.J. Lu <hjl.tools at gmail dot com> 2012-07-14 21:10:28 UTC ---
It is fixed on trunk.  It is on 4.7 branch as of revision 189343.

[Bug target/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[hjl.tools at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

H.J. Lu <hjl.tools at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |hjl.tools at gmail dot com

--- Comment #8 from H.J. Lu <hjl.tools at gmail dot com> 2012-07-14 21:19:01 UTC ---
This address:

(gdb) call debug_rtx (addr)
(and:DI (symbol_ref:DI ("h_DPBtoBCD") [flags 0x41] <function_decl
0x7ffff1b8b400 h_DPBtoBCD>)
    (const_int 4294967295 [0xffffffff]))
(gdb) 

triggers ICE.

[Bug target/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[ubizjak at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

--- Comment #9 from Uros Bizjak <ubizjak at gmail dot com> 2012-07-14 21:48:32 UTC ---
Can you test following patch that prevents non-register AND address operands:

--cut here--
Index: i386.c
===================================================================
--- i386.c      (revision 189483)
+++ i386.c      (working copy)
@@ -11591,7 +11591,8 @@ ix86_decompose_address (rtx addr, struct ix86_addr
          if (GET_CODE (addr) == SUBREG
              && GET_MODE (SUBREG_REG (addr)) == SImode)
            addr = SUBREG_REG (addr);
-         else if (GET_MODE (addr) == DImode)
+         else if (REG_P (addr)
+                  && GET_MODE (addr) == DImode)
            addr = gen_rtx_SUBREG (SImode, addr, 0);
          else if (GET_MODE (addr) != VOIDmode)
            return 0;
--cut here--

[Bug target/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[hjl.tools at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

--- Comment #10 from H.J. Lu <hjl.tools at gmail dot com> 2012-07-14 22:36:57 UTC ---
This only happens on hjl/x32/gcc-4_7-branch branch.

[Bug target/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[hjl.tools at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

--- Comment #11 from H.J. Lu <hjl.tools at gmail dot com> 2012-07-14 22:46:29 UTC ---
(In reply to comment #9)
> Can you test following patch that prevents non-register AND address operands:
> 
> --cut here--
> Index: i386.c
> ===================================================================
> --- i386.c      (revision 189483)
> +++ i386.c      (working copy)
> @@ -11591,7 +11591,8 @@ ix86_decompose_address (rtx addr, struct ix86_addr
>           if (GET_CODE (addr) == SUBREG
>               && GET_MODE (SUBREG_REG (addr)) == SImode)
>             addr = SUBREG_REG (addr);
> -         else if (GET_MODE (addr) == DImode)
> +         else if (REG_P (addr)
> +                  && GET_MODE (addr) == DImode)
>             addr = gen_rtx_SUBREG (SImode, addr, 0);
>           else if (GET_MODE (addr) != VOIDmode)
>             return 0;
> --cut here--

It doesn't work.  I will find which checkin on trunk fixed this.

[Bug target/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[hjl.tools at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

H.J. Lu <hjl.tools at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |ubizjak at gmail dot com
   Target Milestone|---                         |4.7.2

--- Comment #12 from H.J. Lu <hjl.tools at gmail dot com> 2012-07-14 23:53:22 UTC ---
It is fixed by revision 186839:

http://gcc.gnu.org/ml/gcc-cvs/2012-04/msg00791.html

on trunk.

[Bug target/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[uros at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

--- Comment #13 from uros at gcc dot gnu.org 2012-07-15 08:13:57 UTC ---
Author: uros
Date: Sun Jul 15 08:13:47 2012
New Revision: 189491

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=189491
Log:
    PR target/53961

    Backport from mainline
    2012-04-25  Jakub Jelinek  <jakub@redhat.com>

    PR target/53110
    * config/i386/i386.md (and<mode>3): For andq $0xffffffff, reg
    instead expand it as zero extension.


Modified:
    branches/gcc-4_7-branch/gcc/ChangeLog
    branches/gcc-4_7-branch/gcc/config/i386/i386.md

[Bug target/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[ubizjak at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

Uros Bizjak <ubizjak at gmail dot com> changed:

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

--- Comment #14 from Uros Bizjak <ubizjak at gmail dot com> 2012-07-15 08:16:49 UTC ---
Fixed.

[Bug target/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[hjl.tools at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

H.J. Lu <hjl.tools at gmail dot com> changed:

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

--- Comment #15 from H.J. Lu <hjl.tools at gmail dot com> 2012-07-15 13:48:18 UTC ---
I synced hjl/x32/gcc-4_7-branch with gcc-4_7-branch.

[Bug target/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[ubizjak at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

Uros Bizjak <ubizjak at gmail dot com> changed:

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

--- Comment #16 from Uros Bizjak <ubizjak at gmail dot com> 2012-07-16 08:44:29 UTC ---
.

[Bug target/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[uros at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

--- Comment #17 from uros at gcc dot gnu.org 2012-07-22 15:44:55 UTC ---
Author: uros
Date: Sun Jul 22 15:44:51 2012
New Revision: 189756

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=189756
Log:
    PR target/53961
    * config/i386/i386.md (*lea): New insn pattern.
    (*lea_1): Remove.
    (*lea<mode>_2): Ditto.
    (*lea_{3,4,5,6}_zext): Ditto.
    * config/i386/constraints.md (j): Remove address constraint.
    * config/i386/i386.c (ix86_decompose_address): Allow SImode subreg
    of an address.
    (ix86_print_operand_address): Handle SImode subreg of an address.
    (ix86_avoid_lea_for_add): Reject zero-extended addresses for now.


Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/config/i386/constraints.md
    trunk/gcc/config/i386/i386.c
    trunk/gcc/config/i386/i386.md
    trunk/gcc/config/i386/predicates.md

[Bug target/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[uros at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

--- Comment #18 from uros at gcc dot gnu.org 2012-07-23 16:04:30 UTC ---
Author: uros
Date: Mon Jul 23 16:04:23 2012
New Revision: 189787

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=189787
Log:
    PR target/53961
    * config/i386/i386.md (*lea): Add asserts to detect invalid addresses.
    * config/i386/i386.c (ix86_print_operand_address): Ditto.
    (ix86_decompose_address): Allow (zero_extend:DI (subreg:SI (...)))
    addresses.  Prevent zero extensions of CONST_INT operands.


Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/config/i386/i386.c
    trunk/gcc/config/i386/i386.md

[Bug target/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[uros at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

--- Comment #19 from uros at gcc dot gnu.org 2012-07-24 11:37:24 UTC ---
Author: uros
Date: Tue Jul 24 11:37:20 2012
New Revision: 189806

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=189806
Log:
    PR target/53961
    * config/i386/i386.c (ix86_legitimate_address_p): Move check for
    negative constant address for TARET_X32 ...
    (ix86_decompose_address): ... here.  Reject constant addresses
    that don't satisfy x86_64_immediate_operand predicate.


Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/config/i386/i386.c
    trunk/gcc/config/i386/i386.md

[Bug target/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[ubizjak at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

Uros Bizjak <ubizjak at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |ASSIGNED
         Resolution|FIXED                       |
         AssignedTo|unassigned at gcc dot       |ubizjak at gmail dot com
                   |gnu.org                     |

--- Comment #20 from Uros Bizjak <ubizjak at gmail dot com> 2012-07-24 16:18:33 UTC ---
These patches, and the patch for PR52530, should be backported to 4.7.2

[Bug target/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[uros at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

--- Comment #21 from uros at gcc dot gnu.org 2012-08-02 16:24:35 UTC ---
Author: uros
Date: Thu Aug  2 16:24:25 2012
New Revision: 190089

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=190089
Log:
    Backport from mainline
    2012-07-24  Uros Bizjak  <ubizjak@gmail.com>

    PR target/53961
    * config/i386/i386.c (ix86_legitimate_address_p): Move check for
    negative constant address for TARGET_X32 ...
    (ix86_decompose_address): ... here.  Reject constant addresses
    that don't satisfy x86_64_immediate_operand predicate.

    2012-07-23  Uros Bizjak  <ubizjak@gmail.com>

    PR target/53961
    * config/i386/i386.md (*lea): Add asserts to detect invalid addresses.
    * config/i386/i386.c (ix86_print_operand_address): Ditto.
    (ix86_decompose_address): Allow (zero_extend:DI (subreg:SI (...)))
    addresses.  Prevent zero extensions of CONST_INT operands.

    2012-07-22  Uros Bizjak  <ubizjak@gmail.com>

    PR target/53961
    * config/i386/i386.md (*lea): New insn pattern.
    (*lea_1): Remove.
    (*lea<mode>_2): Ditto.
    (*lea_{3,4,5,6}_zext): Ditto.
    * config/i386/predicates.md (lea_address_operand): Do not reject
    zero-extended address operands.
    * config/i386/constraints.md (j): Remove address constraint.
    * config/i386/i386.c (ix86_decompose_address): Allow SImode subreg
    of an address.
    (ix86_print_operand_address): Handle SImode subreg of an address.
    (ix86_avoid_lea_for_addr): Reject zero-extended addresses for now.


Modified:
    branches/gcc-4_7-branch/gcc/ChangeLog
    branches/gcc-4_7-branch/gcc/config/i386/constraints.md
    branches/gcc-4_7-branch/gcc/config/i386/i386.c
    branches/gcc-4_7-branch/gcc/config/i386/i386.md
    branches/gcc-4_7-branch/gcc/config/i386/predicates.md

[Bug target/53961] internal compiler error: in memory_address_length, at config/i386/i386.c:23341

[ubizjak at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53961

Uros Bizjak <ubizjak at gmail dot com> changed:

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

--- Comment #22 from Uros Bizjak <ubizjak at gmail dot com> 2012-08-02 16:30:39 UTC ---
Fixed everywhere.