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] Fix behavior of TER on unrolled loops


On Sunday 21 August 2005 01:25, Daniel Berlin wrote:
> > I am somewhat pessimistic about ever getting register allocator that is
> > able to mysteriously fix all the problems with register pressure.
>
> It's the job of the RA and scheduler, of course, to do so, and it seems
> to work well for everyone else.
> Why do i never see these kinds of hacks in other compiler middle ends?

Maybe everyone else manages to get good code without something like
TER.  Or maybe other compilers for x86 do scheduling to shorten live
ranges before register allocation.  More likely, other compilers you
look at do not have take register pressure into account because they
don't have to target x86 as one of their most important platforms ;-)

Zdenek, maybe you could you try and see if enabling pre-regalloc
scheduling would also make this particular problem you're seeing go
away, just as an experiment...

Gr.
Steven


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