lang/gcc46

[Sunpoet Po-Chuan Hsieh]

On Mon, Dec 12, 2011 at 10:03:14AM -0800, Kevin Oberman wrote:
> On Mon, Dec 12, 2011 at 9:36 AM, Jason Hellenthal <jhell at dataix.net> wrote:
> >
> > Hi Gerald,
> >
> > As a request once again similiar to one I have made in the past... Would it be possible yet to slow down the update process for the gcc46 port ?
> >
> > This is turning out to be quite the pain in the U-Know-What with version flapping and rebuilding because a port depends on it. If I am correct it is updated weekly. I caught the tail end of the previous update and the day after it was bumped to the next snapshot version & by the time both of those were finished the port had once again been bumped to _1.
> >
> > Is there anything that could be done to stabalize this ... ?
> >
> > At this point I am left for the manual intervention of using +IGNOREME files or excluding by whatever means neccesary as weekly updates seem completely unneccesary now that alot of ports are shifting to depend on gcc46.
> >
> > Can a gcc46-devel port be branched for those that absolutely need the weekly updates ?
> +1
> 
> gcc46 is used by so many ports that I am continually re-building it
> and on slow machines, this takes a while. How about a gcc46-devel port
> that gets the regular updates and let gcc46 stay stable when there are
> not major fixes?

We have lang/gcc already. This port is created for perferred gcc releases
(4.6.2 currently). What we're waiting for is a bsd.gcc.mk update to allow
users build ports with lang/gcc instead of lang/gcc46.