On Tue, Dec 12, 2017 at 5:50PM, Gavin Lambert wrote:
>I ended up rolling my own code for asynchronous serial I/O at one point
>because ASIO was a little too mutex-happy, which was causing latency
>issues.
>[...]
>But most memory allocators are decently fast nowadays, to the point
>where having memory allocations on threads that are doing socket I/O
>will probably be dominated by the I/O rather than the allocation.

Thanks for warning me that it's somewhat mutex-happy.
Perhaps that can be fixed, but that'd be one more thing I'd have to do.

And my motivation for controlling memory-allocation wasn't speed, it was to make sure I don't exceed my budget, or cause stability problems (e.g. memory fragmentation), in an embedded environment.

-Steven