[Bug rtl-optimization/49114] [x32] 454.calculix in SPEC CPU 2006 failed

[hjl.tools at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49114

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |vmakarov at redhat dot com
          Component|target                      |rtl-optimization

--- Comment #3 from H.J. Lu <hjl.tools at gmail dot com> 2011-05-24 23:14:55 
UTC ---
gen_reload has

      if (CONSTANT_P (op1) || MEM_P (op1) || GET_CODE (op1) == SUBREG
          || (REG_P (op1)
              && REGNO (op1) >= FIRST_PSEUDO_REGISTER)
          || (code != CODE_FOR_nothing
              && !insn_operand_matches (code, 2, op1)))
        tem = op0, op0 = op1, op1 = tem; 

When IN is

(plus:SI (subreg:SI (reg/v/f:DI 182 [ b ]) 0)
    (const_int 8 [0x8]))

OP0 is SUBREG.  Swapping seems odd since we prefer SUBREG as OP0.
I am testing this patch

@@ -8528,7 +8528,9 @@ gen_reload (rtx out, rtx in, int opnum, enum reload_type
type)

       code = optab_handler (add_optab, GET_MODE (out));

-      if (CONSTANT_P (op1) || MEM_P (op1) || GET_CODE (op1) == SUBREG
+      if (((CONSTANT_P (op1) || MEM_P (op1))
+       && GET_CODE (op0) != SUBREG)
+      || GET_CODE (op1) == SUBREG
       || (REG_P (op1)
           && REGNO (op1) >= FIRST_PSEUDO_REGISTER)
       || (code != CODE_FOR_nothing