Re: 3.5 integration branch created

[Eric Christopher <echristo at redhat dot com>]

> > 1. This branch is for fully-tested, approved patches.  The rules for 
> > committing to it are the same as the rules that apply during Stage 1 
> > of GCC development.  Experimental or incomplete work should not be put 
> > on the branch.
> 
> Well then you should not be accepting patches that usually go in for 
> Stage1 then.
> 

Stage 1 is not for experimental or incomplete either.


> What happens if no one uses this as it having branches is a pain in the 
> neck.

obviously nothing will need to be merged then will it?


> 
> Well if that is the case no one is going to use the branch as they 
> should be more focused
> towards 3.4.  Also are you going to merges where the source has changed 
> so much that you cannot do a merge within one week?
> 

I'd be surprised if this happens.

> > 4. Anyone who commits to the branch is still responsible for 
> > maintaining the patch on the branch: fixing regressions that it 
> > causes, and sometimes updating or reintegrating it after merges.  I 
> > expect that for most patches, this will be much less work than 
> > maintaining the patch on their own.
> 
> Well this just makes this branch to hard to use and no one is going to 
> use it.
> 

This is a standard mainline rule.

> > 5. I may back a patch out of the branch if it (a) causes bootstrap 
> > failure or significant regressions on any platform and the author 
> > doesn't appear to be able to fix it quickly, or (b) don't appear to 
> > have followed Rule 1 or Rule 3.
> 
> What about compile time regressions also, are you going to test speed 
> of the compiler too?
> 

Why bother? no one else does in any of the other regression testers.

-eric

-- 
Eric Christopher <echristo@redhat.com>