Trace Scheduling in GCC

Jan Hubicka jh@suse.cz
Thu Feb 17 16:19:00 GMT 2005


> Honza,
> 
> But what will happen if I try to bypass the tail duplication stage? In
> theory that should generate traces. However, the question is will the
> data dependence graph construction and scheduling stages still work? (for

I am not too faimilar with scheduler itself, but I believe that it
builds full dependence graph for region scheduling, so you ought to be
safe.

Honza
> my purposes, the scheduler does not have to produce correct schedules,
> since all what I need is importing the data dependence graphs to
> feed them into my standalone scheduler. So, it is all about the data
> dependence graphs for my purpose).
> 
> Thanks
> -Ghassan
> 
> 
> On Mon, 14 Feb 2005, Jan Hubicka wrote:
> 
> > > >
> > > >I am right about the functionalities of these options?
> > > >It seems to me that none of these options meets my objective of studying
> > > >traces that have mltiple entrances and multiple exits. Please correct me
> > > >if I am wrong. And if I am right, is there another way for me to get what
> > > >I want, may be by doing some quick hack that bypasses tail duplication
> > > >under the -fsched2-use-tarces option -does this make sense?
> > > >
> > > >
> > > >
> > > Sorry, I don't know this code well.  I am in the same position as you.
> > > I should research the code to answer your questions.  I think Jan is
> > > the right person to do it.  Because he wrote this code.
> >
> > The -fsched2-use-traces is simple superblock scheduler that is preceeded
> > by trace discovery and tail duplication pass.
> > So if you want traces directly, you might steal the code from tracer.c
> > but except for that I don't think it suits your needs very well,
> > unforutnately.
> >
> > Honza
> > >
> > > Vlad
> > >
> >



More information about the Gcc-help mailing list