debugging internal compiler errors ...

[Ian Lance Taylor]

kevin diggs <diggskevin38@gmail.com> writes:

> On Sun, Apr 17, 2011 at 11:38 PM, Ian Lance Taylor <iant@google.com> wrote:
>> kevin diggs <diggskevin38@gmail.com> writes:
>>
>>> Based on this:
>>>
>>> #12 0x0813c8b9 in expand_expr_real_1 (exp=0xb73c76ac, target=0x0,
>>>     tmode=VOIDmode, modifier=EXPAND_NORMAL, alt_rtl=0x0)
>>>     at ../../gcc-4.3.5/gcc/expr.c:9158
>>> 9158          expand_assignment (lhs, rhs, MOVE_NONTEMPORAL (exp));
>>> (gdb) call debug_rtx(lhs)
>>> (const/j (nil))
>>> (gdb) call debug_rtx(rhs)
>>> (lshiftrt (nil)
>>>     (nil))
>>
>> That certainly looks wrong.  There shouldn't be any "(nil)" there.
>> Either the debugger is confused or the compiler is confused.
>>
>>
>> Ian
>>
>
> Sorry, Ian. Both lhs and rhs are trees. So I am guessing the
> debug_rtx() output is meaningless?

Well, yes, that is true.  Good point.

Ian