Comment #0 by andrej.mitrovich — 2023-06-21T10:00:16Z
As Steven mentions here: https://forum.dlang.org/post/[email protected]
> The D GC has a single global lock to allocate memory -- even memory that might be on a free list. So the threads are all bottlenecked on waiting their turn for the lock.
This behavior isn't documented on https://dlang.org/spec/garbage.html
But it is mentioned in passing on this page: https://dlang.org/articles/d-array-article.html#caching
Specifically "Doing this means an O(lg(n)) lookup in the GC's memory pool for every append (not to mention acquiring the global GC lock)."
Comment #1 by robert.schadek — 2024-12-15T15:27:51Z