Re: Concurrency, again

[Eli Zaretskii]

> From: Christopher Allan Webber <address@hidden>
> Cc: Eli Zaretskii <address@hidden>, Philipp Stephani <address@hidden>, 
> address@hidden, address@hidden
> Date: Thu, 27 Oct 2016 15:55:02 -0500
> 
> Python's GIL + threading is probably not the best model to work off
> of... while the GIL is largely misunderstood, there have been a lot of
> challenges with getting this model to work nice.  At one point, it
> turned out that the GIL + threads would result in constant CPU
> thrashing.  This isn't the case any more, but it was once, and was a
> challenging thing to fix:
> 
>   http://dabeaz.com/python/UnderstandingGIL.pdf

It sounds like you are talking about problems inherent to Python
threads, and not about what the concurrency branch in Emacs attempts
to do.

The problematic issues described in the above presentation are not
relevant to the Emacs concurrency branch, because (1) it doesn't have
the thread preemption for CPU-bound threads, and (2) it doesn't switch
threads on _any_ I/O, just on I/O with an external process or user
input, i.e. when Emacs calls pselect or its equivalent to see whether
any input has arrived.

So the performance penalties discussed in that presentation have
nothing to do with what the Emacs concurrency branch implements.