Steven Bosscher wrote:
On 4/20/07, Vladimir N. Makarov <[EMAIL PROTECTED]> wrote:
> Yes, you have complained that you believe the data structure of DF is
> too fat. I guess that is a valid complaint. I don't see the "rtl info
> duplication" though. You've only complained about the current data
> structures, but I have not really seen you propose anything leaner.
>
I did proposed to attach the info to rtl reg. It is more difficult way
than the current approach because of code sharing before the reload.
But may be it is more rewarding. To be honest I don't know.
Changing the dataflow solvers is one thing. Changing a fundamental
assumption in the compiler, namely that pseudoregs are shared, is
another. We rely on that property in so many places. It would not
require rewriting parts of the compiler, but rewriting the entire
backend.
I would not say that is rewriting the entire backend. Yes, it is hard
but it is doable. May be it could improve RTL code locality too.
Although I guess it is a very big job with uncertain result. That is the
scarry part.
I've just remembered we talked for years of how many efforts we should
work to implement OPENMP until Richard Henderson was stuck for a day in
an airport and started the work there and it was mostly done by him,
Diego and Jakub in a few weeks.
