This is the mail archive of the gcc@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: 64-bit integers - long long (only on x86?)


> Jan Hubicka wrote:
> >>In my own experience, gcc does poorer than other good compilers when 
> >>dealing with long long 64-bit integers on x86. I have also talked to 
> >>others experiencing the same, and just the other day, Jack Lloyd posted 
> >>a link to some results which indicated the same. Is there a wellknown 
> >>reason for this, and what is it? If it's not wellknown, I will produce 
> >
> >
> >It is well known.  The GCC register allocator assumes that the both parts
> >of long long needs to be stored in consetuctive registers limiting the
> >amount of choices it have resulting in  ugly code.
> >Sadly changing this design decision is nontrivial, but some steps are being
> >worked on.
> >
> >Honza
> >
> 
> OK. I am glad it is at least a known problem. Are there any concrete 
> goals for, say, 3.2 that addresses this? Or will it take longer than that?

There is new register allocator in progress, that has potential to solve
the problem (it does not at present AFAIK).  It is dificult to tell, whether
3.2 will solve it or not at the moment...

Honza
> 
> /David
> 


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