[Bug optimization/10679] [3.3/3.4] --param min-inline-insns not honoured

["rguenth at tat dot physik dot uni-tuebingen dot de" <gcc-bugzilla at gcc dot gnu dot org>]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=10679



------- Additional Comments From rguenth at tat dot physik dot uni-tuebingen dot de  2003-07-16 07:25 -------
Subject: Re:  [3.3/3.4] --param min-inline-insns not
 honoured

On 15 Jul 2003, pinskia at physics dot uc dot edu wrote:

>  think this is fixed in 3.4 (20030715) with the new inlining counting mechanism so that
> since bar is empty, gcc will inline it always. I just tried one with 30000 function calls to bar
> and it always inlined bar into foo at -O2.  It took over 90 seconds though it is a little too
> much time spent on it though.

Are you sure this is true in all circumstances? In gcc/tree-inline.c I
still see the offending check at line 1020. So if not inlinable_function_p
is called with nolimit set always, we'll still hit this bug (I suppose
this happens at least with -fno-unit-at-a-time). If it is, this function
could be cleaned up a lot.

Richard.

--
Richard Guenther <richard dot guenther at uni-tuebingen dot de>
WWW: http://www.tat.physik.uni-tuebingen.de/~rguenth/