This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [tree-ssa] Removal of gotos from cfg based ir


On Thu, 2003-11-27 at 17:05, Zdenek Dvorak wrote:
> Hello,
> 
> > > > Its abosultely related, and central to the issue. Thats not for us to
> > > > decide, thats for the people writing those passes to decide. If they
> > > > dont want or need a CFG fopr whatever they want to do to the IL, we
> > > > ought not be forcing them to use one because we think its a good idea in
> > > > some other part of the compiler.
> > > 
> > > and I have never proposed anything like that, did I? I simply state that
> > > for usage during the cfg-based optimizations the no-goto form is more
> > > convenient (which IHMO is obvious, and I and Honza have based this
> > > claim on several examples) and that I consider it a reason 
> > > good enough for using it.
> > > 
> > > Frankly, I don't like the way you try to argument:
> > > 
> > > "I don't want to have no-gotos cfg form, because there are optimizations
> > > that don't work over cfg."
> > > 
> > > I just don't see the connection between these issues.
> > > 
> > I understood that the intention was to translate to CFG form, and keep
> > the CFG straight through the expanders and into RTL.
> > 
> > Is that not the case?
> 
> yes, this is Honza's plan.  But it is not directly related to the
> discussed patch (and this is why interpreting your arguments as related
> to the patch was quite confusing :-)

Sorry for any confusion, but I thought they were related, and that this
was a step along the way.

I would prefer to not have mutiple forms of the IL whenever it is 
possible. So we need compelling reason why we cant do something in the
current form. This seems to be driven entirely by the creation of
forwarder blocks, so let see if we can find another solution that isnt a
horrible hack before we make a change like this.

Lets follow on with:
http://gcc.gnu.org/ml/gcc-patches/2003-11/msg02199.html

Are there other problems which are significant?

Andrew




Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]