On Friday, May 31, 2019 10:36:10 PM MSK Konstantin Osipov wrote:
> * Georgy Kirichenko <georgy@tarantool.org> [19/05/23 11:48]:
> 
> I appreciate you don't use a fiber_call() or fiber_wakeup()
> right from the callback,
> 
> In other places we actually use fiber_wakeup(). Please see gc.c,
> memtx_engine.c and others.
It isn't possible because this fiber could process some yielding activity 
(ddl).
> 
> So please let's not use has_message/cond, simply wakeup the fiber
> in question.
> 
> More comments as to why you need to run these in a fiber would not
> harm.