How to express the fact that arguments don't escape?

[Tobias Schlüter]

Richard Guenther wrote:
> On Sat, Aug 15, 2009 at 1:11 PM, Thomas Koenig<tkoenig@gcc.gnu.org> wrote:
>> On Sat, 2009-08-15 at 12:11 +0200, Richard Guenther wrote:
>>>> To stay within the current, C-centric framework of gcc, we'll need
>>> to do
>>>> even more copying.
>>> Yes, of course.
>> What about the following pseudo-code:  For an scalar argument integer i,
>> to the subroutine foo with the corresponding dummy argument i_dummy, we
>> can the do (pseudo code)
>>
>> if (optimize || ! is_target(i_dummy)) {
>>   int i_temp;
>>   if (intent_inout(i_dummy) || intent_in(i_dummy) || intent_unpecified(i_dummy))
> 
> for this you can include intent_in(i_dummy) as well, thus do it unconditionally.
> 
>>      i_temp = i;
>>   call foo(&i_temp);
>>   if (is_no_active_do_loop_variable(i)
>>        && (intent_inout(i_dummy) || intent_out(i_dummy) || intent_unspecified(i_dummy)))
>>     i = i_temp;
> 
> just this copy is not needed for intent_in.
> 
> Note that this would be indeed profitable for all types that are cheap to copy
> (is_gimple_reg_type in gimple speak).

This is very valuable information!

I'll attack this next.

Cheers,
- Tobi