[Bug rtl-optimization/32557] [4.3 Regression] internal compiler error: RTL check: expected code 'reg', have 'subreg' in rhs_regno, at rtl.h:956

[rask at gcc dot gnu dot org]

------- Comment #7 from rask at gcc dot gnu dot org  2007-08-20 16:06 -------
Created an attachment (id=14082)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=14082&action=view)
Preprocessed testcase; compile with -O2 -g -m2a -fno-builtin

It appears to fail the same way as earlier:

(gdb) frame 3
#3  0x000000000050a7f5 in df_note_bb_compute (bb_index=28, live=0xf335c0,
do_not_gen=0xf332e0, artificial_uses=0xf33580)
    at /n/12/rask/src/all/gcc/df-problems.c:3490
3490              if ((mws->type != DF_REF_REG_DEF)
(gdb) list
3485          /* Process the uses.  */
3486          mws_rec = DF_INSN_UID_MWS (uid);
3487          while (*mws_rec)
3488            {
3489              struct df_mw_hardreg *mws = *mws_rec;
3490              if ((mws->type != DF_REF_REG_DEF)
3491                  && !df_ignore_stack_reg (REGNO (mws->mw_reg)))
3492                old_dead_notes
3493                  = df_set_dead_notes_for_mw (insn, old_dead_notes,
3494                                              mws, live, do_not_gen,
(gdb) call debug_rtx(mws->mw_reg)
(subreg:DF (reg/v:DI 3 r3 [orig:327 d2 ] [327]) 0)
(gdb) print *mws
$1 = {mw_reg = 0x2b27c6a994e0, type = DF_REF_REG_USE, flags = 100, start_regno
= 3, end_regno = 4, mw_order = 207550}


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32557