GCC 4.0 vs. 3.5

[Paul Schlie]

Where then possibly the lno branch could then be merged into 3.5 (current
mainline) which could then also possibly serve as the basis for the ssa
merged 4.0?

> Paul Schlie wrote:
> Unless I misunderstood, the 3.5 performance in this context referred to
> mainline not ssa, so wonder if it makes sense to enable the current mainline
> to remain accessible as 3.5 as an evolution of 3.4 prior to it's merge with
> ssa, where mainline+ssa becomes 4.0?
> 
>> Roger Sayle wrote:
>>> On Fri, 16 Apr 2004, Mark Mitchell wrote:
>>> So, if we're switching to 4.0 it should be because the code we generate
>>> is far better.  Not that some code is slightly better, or that we have
>>> the infrastructure we need to generate far better code -- but that much
>>> actual code is actually far better.
>> 
>> I guess the recent thread "100% speed improvement 3.4 -> 3.5 for ia64!"
>> http://gcc.gnu.org/ml/gcc/2004-04/msg00646.html might support the 4.0
>> argument?  :>