Experimenting with tree inliner parameter settings

[Jakub Jelinek]

On Sun, May 04, 2003 at 10:55:22PM +0200, Steven Bosscher wrote:
> Op zo 04-05-2003, om 22:44 schreef Richard Guenther:
> > [Summary / analysis of inline params snipped]
> > 
> > The SPEC results show, that at least defaults for C and Fortran do not
> > necessarily apply to C++ code (as the testcases from the PRs show).
> 
> Hmm, where do you see that indicated in the SPEC results?  I have to
> admit I did not look at that.  I'm not sure how C++ and C would be
> different -- what works for C++ should work for C as well, no?

No, typical C and C++ code differs. C++ has lots of stuff in the headers,
while C usually has just prototypes in headers and when an inline function
is in C headers, it is usually beneficial to inline it.
E.g. for glibc and linux kernel, current inline limits (whether in 3.2.x, or
3.3.x) are way too low, recent linux kernels even define inline to
inline __attribute__((always_inline))...

	Jakub