D GC theory

[via Digitalmars-d]

On Tuesday, 24 February 2015 at 17:07:03 UTC, Ola Fosheim Grøstad 
wrote:
> But if you can control where the code is running when you run 
> the GC scan then you might as well have a separate code path 
> for concurrent GC too (i.e. two versions of the code). One with 
> fences and one without...

I'd say this is impractical. You could only reasonably expect 
this to work at certain checkpoints, say, an event loop with 
short calls into the "actual" application code. You cannot simply 
switch right in the middle of a deep call.

And with Sociomantic's GC, you don't have any more "control" over 
the GC than you have with the current one in druntime.