Re: Some performance notes

[Owen Taylor <otaylor redhat com>]

Tim Janik <timj gtk org> writes:

> On 5 Aug 2001, Owen Taylor wrote:
> 
> > So, without including hard data (I don't have much), here are
> > my observations:
> > 
> >  * The GTK_IS_FOO() and GTK_FOO() debugging checks are hitting
> >    us pretty hard right now. The overhead of this is somewhere
> >    around 30%, which is a lot more than than it was in 
> >    GTK+-1.2.
> 
> erk, i noticed that i didn't- as i thought i had -commit the
> gcc-specific instance and class checks recently. so you might want
> to check these 30% again with applying the appended patch first
> (i'm currently hacking interfaces in gtype.[hc] so i'll take a couple
> more days untill this goes into CVS).

I don't expect it to make a big effect since earlier tests
indicated that a-is-a checks only were 10-20% for a typical
GTK+ program, but I can certainly try it at some point.

> > I would like to get the overhead of debugging down to the point where
> > we can ship with --enable-debug=minimum as we did for GTK+-1.2, but if
> > necessary we can go with --disable-debug for production builds and
> > just encourage developers to use --enable-debug versions.
> 
> did you get any hard numbers that would indicate we should really ship
> with --disable-debug instead of --enable-debug=minimum? i wonder what
> extra noticable impact minimum over none should have.