Implementation of concurrency primitives

[Alan Bateman]

On 18/11/2021 06:44, Andrey Lomakin wrote:
> Good morning, guys.
>
> I am considering creating an application that will use a
> thread-per-core approach. Data mostly will not be shared between cores and
> I am going to use a single thread executor to schedule fibres.
>
> My question is may I use my own concurrency primitives which do not inject
> memory barriers unlike the standard concurrency primitives to
> achieve maximum performance and if that is not possible could you explain
> why?
I don't see how this could be reliable in general as APIs and library 
code that you use could execute code that is pre-empted, it could happen 
anywhere.

Also just to say, for now anyway, there isn't an API exposed that allows 
you to set your own scheduler. That is something that the project will 
have to come back to in the future.

-Alan