[Bug rtl-optimization/57422] [4.9 Regression] ICE: SIGSEGV in dominated_by_p with custom flags

["jakub at gcc dot gnu.org" <gcc-bugzilla at gcc dot gnu dot org>]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57422

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |abel at gcc dot gnu.org,
                   |                            |jakub at gcc dot gnu.org,
                   |                            |vmakarov at gcc dot gnu.org

--- Comment #2 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Started with r198695, but guess it just uncovered a latent issue in the
selective scheduler.

The reason why FENCE_INSN doesn't have BLOCK_FOR_INSN is that it has been
earlier removed:
#0  delete_insn (insn=0x7ffff19e8750) at ../../gcc/cfgrtl.c:175
#1  0x0000000000a276e1 in sel_remove_insn (insn=0x7ffff19e8750,
only_disconnect=false, full_tidying=false) at ../../gcc/sel-sched-ir.c:3938
#2  0x0000000000a3e035 in remove_insn_from_stream (insn=0x7ffff19e8750,
only_disconnect=false) at ../../gcc/sel-sched.c:6042
#3  0x0000000000a3e0ea in move_op_orig_expr_found (insn=0x7ffff19e8750,
expr=0x19b6848, lparams=0x7fffffffdc60, static_params=0x7fffffffdc30)
    at ../../gcc/sel-sched.c:6065
#4  0x0000000000a3ecdb in code_motion_path_driver (insn=0x7ffff19e8750,
orig_ops=0x19b6840, path=0x19b7f38, local_params_in=0x7fffffffdc60, 
    static_params=0x7fffffffdc30) at ../../gcc/sel-sched.c:6603
#5  0x0000000000a3f000 in move_op (insn=0x7ffff19e8750, orig_ops=0x19b5058,
expr_vliw=0x19b7e50, dest=0x7ffff19fc9e0, c_expr=0x7fffffffdd40, 
    should_move=0x7fffffffdd1a) at ../../gcc/sel-sched.c:6758
#6  0x0000000000a3c651 in move_exprs_to_boundary (bnd=0x19b8af8,
expr_vliw=0x19b7e50, expr_seq=0x19b5058, c_expr=0x7fffffffdd40)
    at ../../gcc/sel-sched.c:5292
#7  0x0000000000a3d122 in schedule_expr_on_boundary (bnd=0x19b8af8,
expr_vliw=0x19b7e50, seqno=-1) at ../../gcc/sel-sched.c:5504
#8  0x0000000000a3d55c in fill_insns (fence=0x19b7748, seqno=-1,
scheduled_insns_tailpp=0x7fffffffdf20) at ../../gcc/sel-sched.c:5646
#9  0x0000000000a406ae in schedule_on_fences (fences=0x19b7740, max_seqno=30,
scheduled_insns_tailpp=0x7fffffffdf20) at ../../gcc/sel-sched.c:7410
#10 0x0000000000a40b16 in sel_sched_region_2 (orig_max_seqno=85) at
../../gcc/sel-sched.c:7544
#11 0x0000000000a40c83 in sel_sched_region_1 () at ../../gcc/sel-sched.c:7583
#12 0x0000000000a410cf in sel_sched_region (rgn=0) at
../../gcc/sel-sched.c:7684
#13 0x0000000000a411e9 in run_selective_scheduling () at
../../gcc/sel-sched.c:7760

and within the same fill_insns call (so the same fence) in the next cycle we
ICE.