This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Using FDO to speed up compiler?
> On Wed, Feb 12, 2003 at 10:59:45AM -0800, Richard Henderson wrote:
> > At the moment I'm bootstrapping to find out how exactly how much
> > compile time I'm going to save by removing the cse code. That'll
> > give me some idea if this is worth persuing at all for 3.4.
>
> Before: 1168.34user 35.03system 5:05.02elapsed 394%CPU
> After: 1109.25user 35.07system 4:50.41elapsed 394%CPU
>
> That's a 5% decrease in user time.
This is not bad :)
The number reminds me not-so-related note. I've also hacked makefiles
in the past to add profiling into the bootstrap. I used "make check"
for train run and measured speedup in between 5-10% on compiling
on compiling binutils on early rtlopt-branch tree.
Do we want such a machinery in mainline? Using FDO for compiler seems to
make sense, but it makes bootstrap much longer. I set it up as
"profiledbootstrap" in makefile that would imply that for development we
will use different switches that for production that can cause problems
in future...
Also getting this right for crosscompilation may be dificult.
Honza