Consider dispatching IMENotificationSender runnable differently (possibly off of the idle queue)

RESOLVED FIXED

Status

()

enhancement
P1
normal
RESOLVED FIXED
2 years ago
2 years ago

People

(Reporter: mstange, Assigned: masayuki)

Tracking

(Blocks 1 bug)

Trunk
Points:
---
Dependency tree / graph

Firefox Tracking Flags

(Not tracked)

Details

Reporter

Description

2 years ago
Here's a speedometer profile in which we're dispatching a IMENotificationSender runnable during an async test (so it gets measured and contributes to the score): http://bit.ly/2saFDqT
Flags: needinfo?(bugs)
This is very much related to bug 1375491.
Using idle dispatch sounds wrong.
Depends on: 1375491
Flags: needinfo?(bugs) → needinfo?(masayuki)
We should rather run the runnable around the time we flush layout anyhow.
I think so. If "idle" always occurred immediately after input event, my though would be wrong, though.
Flags: needinfo?(masayuki)
Idle may happen at any point or may not happen at all.
If this indeed negatively affects Speedometer then it should be P1.
Assignee: nobody → masayuki
Priority: -- → P1
bug 1375491 changed this quite a bit. This needs first some reprofiling.
mstange:

Could you check if we still need to do it after fixing bug 1375491?
Flags: needinfo?(mstange)
So this was slow because of the reflow.
bug 1375491 should have moved the flush to happen in better place.
Marking fixed, but please reopen if this is still an issue.
Status: NEW → RESOLVED
Closed: 2 years ago
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.