863978 - B2G: 2.6s MEM_PRESSURE GC during Browser application start-up.

Status: RESOLVED WONTFIX

(Core :: JavaScript Engine, defect)

Description

[Nicolas B. Pierron [:nbp]]

attachment 739929 [details] reports that when we switch to a new application, we see MEM_PRESSURE GCs which might take a lot of time, including one which is taking up to 2.6 seconds to complete.

Based on marionette log, this GC pause of 2.6 seconds should be at the end of the browser load, likely done in another process.  The log does not explicit which application is having this MEM_PRESSURE GC, but I guess this is likely the home screen.

We should probably delay the MEM_PRESSURE GC when we are running multiple applications, such as it does not impact the start-up time of the next application.

Comment 1

[Gregor Wagner [:gwagner]]

Whenever an app goes into the background, we send a memory pressure event to this app in order to minimize after a few sec.

Comment 2

[Justin Lebar (not reading bugmail)]

> We should probably delay the MEM_PRESSURE GC when we are running multiple applications, 
> such as it does not impact the start-up time of the next application.

It would be helpful if you could try to determine whether the GC we're running in this unknown process is impacting startup time.  Otherwise, everything is behaving as expected.

I don't expect a startup time impact from this, but it's possible given how broken setpriority is.

Comment 3

[Benjamin Bouvier [:bbouvier] (inactive)]

Mass-closing JS bugs for which the platform is Gonk (Firefox OS), since Firefox OS is gone. Feel free to re-open if still valid.