Alias Analysis Improvement Patch

[tm_gccmail@mail.kloo.net]

On Fri, 13 Jun 2003, Vladimir N. Makarov wrote:

> In average gcc with the patch is worse especially for twolf.  I think I
> should search for the reason of degradation in counteraction of scheduler
> and alias analysis.

I suspect it's similar to the SH problem.

On the SH arch, when we naively enable the first scheduling pass, it
creates worse code because it lengthens register lifetimes and creates
more register spills.

On the IA64 it could be a similar problem where the first instruction
scheduling pass has fairly limited 
Toshi