July 28, 2011
1:42 a.m.
On Wed, 2011-07-27 at 18:10 -0400, Glyph Lefkowitz wrote:
Mostly though, an additional queue for non-reentrancy would be an ad-hoc special-case solution to one tiny part of the more general problem that reactor event ordering is a complete accident with unforseen consequences. If there were a more general scheduling mechanism, we could schedule events in a better order overall.
We already have a queue, for reactor.callFromThread. We could add, say, reactor.runInABit, and make reactor.callFromThread that just adds a waker notification after calling that.