On 1/06/2023 06:42, Christian Mazakas wrote:
Unordered's concurrent flat map
Which is where, exactly? It does not appear to currently exist in the library as far as I can tell, not even in develop...
contains a hand-rolled spinlock (graciously provided by Peter Dimov) but it contains a dependency on SmartPtr for the following two headers:
#include
#include I advocate that we move these to Core. They're small, self-contained and are useful in broader contexts.
Is there some reason why the spinlock itself shouldn't also be in Core, in that case? Or perhaps Boost.Compat should get a std::shared_mutex equivalent for this purpose instead? I dislike libraries bypassing the standard library for this sort of thing, as it makes it harder to deal with platform quirks or reason about blocking in latency-sensitive code.