Issue metadata
Sign in to add a comment
|
High CPU consumption on background tab |
||||||||||||||||||||||
Issue descriptionUserAgent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_6) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/11.1 Safari/605.1.15 Example URL: https://flytothesky.ru/5-fraz-kotorye-govoryat-manipulyatory-chtoby-svesti-vas-s-uma/?utm_referrer=https%3A%2F%2Fzen.yandex.com Steps to reproduce the problem: 1. Open the URL provided, wait for page load (it may take up to 15 seconds) 2. Put mouse cursor into middle of the page content 3. Press Cmd+T to open new tab 4. In the new tab, open about:tracing and collect "Javascript and rendering" traces for several seconds What is the expected behavior? There is none to low work on opened page. What went wrong? Some mouse event is repeatedly processed causing re-layouts and high power consumption. Does it occur on multiple sites: N/A Is it a problem with a plugin? N/A Did this work before? N/A Does this work in other browsers? N/A Chrome version: 66.0.3359.0 Channel: dev OS Version: OS X 10.12.6 Flash Version: Not installed Everything works as expected when new tab is open with mouse/touchpad.
,
Apr 20 2018
I've noticed a difference between Mac & Windows. On Windows, even if the tab is switched using the keyboard, MouseLeave event is passed to renderer. On Mac, there is no such event in the same circumstances.
,
Apr 20 2018
Mac triage: Per #2, marking for Blink>Input triage.
,
Apr 26 2018
I'm unable to try this on a Mac but everything appears as expected on Win/Linux. This seems like a scheduling/tab backgrounding issue so over to scheduling.
,
Apr 26 2018
Nice catch! I got a repro on Mac (see attached trace). We're firing a lot of mouse move events: MouseEventManager::fakeMouseMoveEventTimerFired. It seems that the right option is to stop firing them when we are in the background. So back to input team.
,
Apr 26 2018
,
Apr 26 2018
Thanks for finding that, Navid - could someone familiar with mouse events take a look?
,
Aug 15
Navid, could you take a look?
,
Aug 15
,
Aug 27
The NextAction date has arrived: 2018-08-27
,
Aug 28
Hello, could you please provide an access for me to https://bugs.chromium.org/p/chromium/issues/detail?id=855576?
,
Aug 28
You might want to look at issue 877132 which is going to make issue 855576 obsolete as it is going remove the timer altogether.
,
Aug 29
Thank you, I'll take a look at that issue. |
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 by vamshi.kommuri@chromium.org
, Apr 19 2018