Joining aside, the basic point ("But when size = 7279, the data submitted reaches 64k, so the writting thread blocks on the write syscall.") is not clear from the docs, right?

IMO, it would be nice if I could ask my queue, "Just what is your capacity (in bytes, not entries) anyways? I want to know how much I can put in here without worrying about whether the remote side is dequeueing." I guess I'd settle for explicit documentation that the bound exists. But how should I expect my code to be portable? Are there platforms which provide less than 64k? Less than 1k? Less than 256 bytes?