The event emission in eo is slow. We know that. Evas for example works arround that by tracking internally if someone subscribed to a event before emitting it. To get a number of how often we are emitting a event without anyone listening to it, i have created a hack in eo, you can find the branch at: https://git.enlightenment.org/core/efl.git/log/?h=devs/bu5hm4n/dead_events (I am calling these situations dead events)
The output looks like:
Emitting dead event del 0 2
The first number (the 0) indicates if the is legacy or not. (0 not legacy, 1 legacy).
del is the name of the event
The second number (the 2) is the amount of callbacks we have walked.
The more callbacks we have walked, the more time we have spent in there. So maybe the one with a high second number should be prioritized.
After a quick look over a example run with elm test:
- legacy move events (@zmike)
- render,flush,pre events (@zmike)
- focus_geometry (@bu5hm4n)
- invalidate (@cedric)
(Lets reevalulate what we are doing with this once we have fixed the biggest ones)