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: [patch] Avoid ssa update in cfg cleanup


On Thu, 2006-02-23 at 17:40 +0100, Zdenek Dvorak wrote:
> Hello,
> 
> > On Wed, 2006-02-15 at 10:56 -0500, Diego Novillo wrote:
> > > Zdenek Dvorak wrote:
> > > 
> > I also prefer not to unite the maydef and mustdef structures. They are
> > conceptually different and should remain separate I think.
> 
> I do not see them as that much different -- in fact, I would find it
> much more natural to have only single type of VDEF operands, with a
> single bit telling whether the definition is killing or not.

And I see it the other way. I prefer that this information be contextual
rather than a bit in the structure.  From a design point of view they
represent different conceptual things and if one ever needs to change
one of them to a different representation, it is much easier.


> > Its simple
> > enough to provide a function to convert a maydef to a mustdef while
> > maintaining use links if that is what is needed.
> 
> I have considered this; however, it is fairly difficult to get this work
> with the way tree-ssa-opfinalize.h.

I don't understand why. Perhaps I'm a bit dense right now, but exactly
How/when do you know you need to change a maydef to a mustdef?  and why
does it need to be in the opfinalize hunk?

Andrew


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