Chrome eating battery on Mac |
||||||
Issue descriptionVersion: 52.0.2743.33 OS: Mac OS X 10.11.5 The last two days, Chrome seems to be absolutely killing the battery on my Mac. Whereas I could normally browse for hours without having to worry about charging the battery, the fans on my Macbook Air (13-inch, Early 2014) are constantly running. This is true even with few tabs open. It meant I was unable to get much work done on a recent flight, because Chrome ran down my battery in a short period of time. Screenshot attached of the Mac battery manager UI. Chrome is the only application I'm running that's using any energy. I want to emphasize that this is a really bad regression as I have not noticed this problem before. Happy to provide more details if needed.
,
Jun 14 2016
(The OS>Kernel>Power component is just for Chrome OS power management. I'm not sure what's used for power-related issues on other platforms.)
,
Jun 14 2016
[mac triage] I think that kind of energy impact is typical for Chrome - it's sampled over 8 hours. Spinning fans is usually a rogue process at high CPU, and the screenshots don't show Chrome with a high CPU. Is there anything in the `CPU` tab? Make sure you have `View -> All Processes` ticked, and sort by CPU%. And does this go away on a reboot?
,
Jun 14 2016
Just bumping this and CCing a few more folks. Easily reproducible for me and happy to get debug logs if you like. Moving to canary (53.0.2766.0) fixes this for me.
,
Jun 14 2016
Re #3: Does not go away after a reboot. I can tell you that this battery usage is *very* atypical as I use Chrome heavily on this machine, and it has never been running that hot before.
,
Jun 14 2016
I agree with #3 -- sounds like a rogue process, but the screenshots show nothing. In Activity Monitor, please sort by CPU usage. The "Energy Impact" is only vaguely related to actual power usage.
,
Jun 15 2016
mdw@ does anything show up sorting by CPU, with View -> All Processes?
,
Jun 16 2016
I switched back to M52 yesterday and could not reproduce. I'm closing this for now and will reopen if it happens again. Thanks!
,
Jun 17 2016
+shrike I'm confused, you filed this against M52 but you say going back to M52 fixes it?
,
Jun 21 2016
Re #9: Sorry, what I meant was that: - I started on M52 and saw the problem. - I switched to M53 which did not have the problem. - I switched back to M52 and the problem did not (yet) re-occur. So it must have been a runaway process. |
||||||
►
Sign in to add a comment |
||||||
Comment 1 by mdw@chromium.org
, Jun 12 2016120 KB
120 KB View Download