Excellent, thanks!
The ordering is guaranteed, because full synchronization is used
when threads migrate. (It goes something like, a capability with
a full run queue grabs all idle capabilities, distributes its
threads to those capabilities, and then releases them. The act
of acquiring and releasing a capability is a synchronization point.)
Cheers,
Edward
Excerpts from David Turner's message of 2015-05-28 09:24:50 -0700:
> Hi,
>
> If I make a sequence of FFI calls (on a single Haskell thread) but
> which end up being called from different OS threads, is there any kind
> of ordering guarantee given? More specifically, is there a full memory
> barrier at the point where a Haskell thread migrates to a new OS
> thread?
>
> Many thanks,
>
> David