[Bug target/58421] [4.9 regression] FAIL: gcc.c-torture/compile/20051216-1.c -O3 -fomit-frame-pointer (internal compiler error)

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58421

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |ice-on-valid-code
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2013-09-16
                 CC|rguenther at suse dot de           |rguenth at gcc dot 
gnu.org
     Ever confirmed|0                           |1

--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
Confirmed with a cross from x86_64 and -O3.

#1  0x00000000007c857e in add_insn_before_nobb (insn=0x7ffff6ed9b40, 
    before=0x7ffff6e8c708)
    at /space/rguenther/src/svn/trunk/gcc/emit-rtl.c:3836
3836      gcc_assert (!optimize || !INSN_DELETED_P (before));
(gdb) call debug_rtx (before)
(insn/v 361 1077 638 69 (set (reg/f:DI 37 r38 [orig:463 new_vp.2 ] [463])
        (unspec:DI [
                (mem/f/c:DI (reg/f:DI 47 r48 [760]) [6 new_vp+0 S8 A64])
            ] UNSPEC_LDS)) 20051216-1.c:103 23 {movdi_speculative}
     (nil))

#7  0x0000000000adaba2 in get_nop_from_pool (insn=0x7ffff6e8c708)
    at /space/rguenther/src/svn/trunk/gcc/sel-sched-ir.c:1048
1048      nop = emit_insn_before (nop, insn);

some unsharing is maybe missing?  Looks like a backend or scheduling bug
anyway.