what are the "primary targets" for GCC?

[Mike Stump]

On Dec 2, 2004, at 5:36 AM, Giovanni Bajo wrote:
> I asked the SC a clarification on the release criteria for 4.x here:
> http://gcc.gnu.org/ml/gcc/2004-10/msg01229.html
>
> but got no answer.

I don't feel they have to answer these questions.  If they feel any 
specific one is compelling to them, they should of course state a 
direction for us to move in; otherwise, the answers to your questions 
can be a consensus view of the people doing the work in gcc combined 
with what work people sign up to actually do.  Someone can propose 
numbers and criteria and solicit feedback.  The RM has the latitude to 
push us in that direction, or this one and to publish any such list as 
he sees fit.


For example, I propose that we consider Darwin a primary platform.  
This isn't crucial or critical, but, I think reflects reality better.

I propose that 25% speed regressions at -O0 are beyond what we consider 
ok.  I think we can conservatively say that 4% might be more reasonable 
and reflective of the current climate.