[Webkit-unassigned] [Bug 155255] MemoryPressureHandler doesn't work if cgroups aren't present in Linux

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=155255

Sergio Villar Senin <svillar at igalia.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |svillar at igalia.com

--- Comment #20 from Sergio Villar Senin <svillar at igalia.com> ---
(In reply to comment #18)
> (In reply to comment #15)
> > (In reply to comment #13)
> > 
> > > > Source/WebCore/platform/linux/MemoryPressureHandlerLinux.cpp:149
> > > > +        sleep(s_pollingIntervalInSeconds);
> > > > +    } while (true);
> > > 
> > > I'm not happy with this either, but I have no idea how to do this if meminfo
> > > is not pollable. Note that there's one memory pressure handler per process .
> > 
> > We can use a larger polling interval (5 sec?) and wider memory threshold
> > margins so that we trigger the pressure condition earlier without needing
> > such a high monitoring frequency. Actually, I thing that the current
> > thresholds for remaining memory of 300MB (non critical) and 100MB (critical)
> > are wide enough.
> > 
> > About battery consumption, take into account the relative expense of
> > "opening a text file, parsing its first lines and closing it". What can it
> > be, some milliseconds? I can't believe it's so expensive compared to the
> > huge amount of time (in terms of CPU time scale) that the thread is going to
> > be idle.
> 
> The problem is precisely that the thread is never going to be idle, it's
> going to wake up every second. It's not only a matter of what you do when
> you wake up, it's just that you are waking up every second. I've seen
> patches in embedded devices to fix battery drain that simply removed loops
> like this one. If there's no other way to do this, we should figure out if
> it's possible to disable this while there's no activity, or pages are all
> hidden, etc.

Yes please, let's not land something like this unless we want embedded devices not to use webkitgtk

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.webkit.org/pipermail/webkit-unassigned/attachments/20160314/b4197735/attachment.html>