Adding Java and C++ to the MQTT benchmarks or: How I Learned to Stop Worrying and Love the Garbage Collector

[Benjamin Thaut]

Am 09.01.2014 08:07, schrieb Walter Bright:
>
> The point is, no matter how slow the GC is relative to malloc, not
> allocating is faster than allocating, and a GC can greatly reduce the
> amount of alloc/copy going on.
>

The points should be, if D is going to stay with a GC, and if so, when 
we will actually get propper GC support so a state of the art GC can be 
implemented. Or if we are going to replace the GC with ARC.

This is a really important topic which shouldn't wait until the language 
is 20 years old. I'm already using D since almost 3 years, and the more 
I learn about Garbage Collectors and about D, the more obvious becomes 
that D does not properly support garbage collection and it will require 
quite some effort and spec changes to do so. And in all the time I used 
D nothing changed about the garbage collector. The only thing that 
happend was the RtInfo template in object.d. But it still isn't used and 
only solves a small portion of the percise scanning problem. In my 
opinion D was designed with language features in mind that need a GC, 
but D was not designed to actually support a GC. And this needs to change.

If requested I can make a list with all language features / decisions so 
far that prevent the implementation of a state of the art GC.

-- 
Kind Regards
Benjamin Thaut