Potential GSoC project - GC improvements

[Joakim via Digitalmars-d]

On Wednesday, 9 March 2016 at 22:49:28 UTC, Jeremy DeHaan wrote:
> Hey all,
>
> I'm trying to think of good project ideas for this years GSoC, 
> and one in particular I thought would be a great was working on 
> and improving the GC. I'm not sure what the scope of this 
> project would be like, but at the moment I am thinking writing 
> a generational collector would be a good place to start.
>
> I have more ideas, but I don't have a proposal yet. I just 
> wanted some initial feedback on the idea, perhaps some advice 
> for scope with the time frame in mind, and hopefully someone on 
> the mentor list willing to take this on if it were to be 
> accepted.

I'd think working on the GC would make a great GSoC project, as 
it's needed and fairly self-contained while having large impact.  
Perhaps someone could build off of Sociomantic's concurrent GC 
(https://www.sociomantic.com/blog/2013/06/porting-cdgc-to-d2/), 
which I assume has been ported to D2, porting it to Windows or 
whatever else remains to be done and then adding to it.

There were some good ideas in this thread from a couple years ago 
too:

http://forum.dlang.org/thread/ucsamkqvrihfuspnxfsk@forum.dlang.org