6 ms·

Steps to reproduce: 1) Pick a rMBP. Any one. 2) Set display options - "more space" under System Preferences. 3) Run Firefox. Load any website. No extensions;

Steps to reproduce: 1) Pick a rMBP. Any one. 2) Set display options - "more space" under System Preferences. 3) Run Firefox. Load any website. No extensions; safe mode doesn't matter. 4) 100%+ CPU, fans kick up and run high, system gets hot and slows to a crawl. Chrome and Safari run totally fine under the same conditions. It's documented pretty thoroughly in the first Bugzilla link I posted; the second one is an offshoot where they're trying to solve this specific issue. If you do happen to work at Mozilla, would really appreciate your help escalating. EDIT: I now have a perf file showing significant issues when clicking on a YouTube video. It gives an error when trying to upload it to the site, but if you contact me through email (in my HN profile), I can send it over.

exikyut 8y ago
I don't have a rMBP, and others who may be curious to help may not either. Uploading performance profiles from various websites to the mentioned URL will give everyone something to poke at.
mycelium 8y ago
I spent a few minutes trying to replicate on a MacBook Pro (Retina, 13-inch, Late 2013) with Firefox 62.0.2 and could not replicate, browsing a number of resource intense sites.
- ericabiz 8y ago
  Interesting, as this is the same MBP I have, and same Firefox version. I now have a perf file showing the issue. Can't upload it (it gives an error), but would be happy to send it to anyone who thinks they can help.
  - muizelaar 8y ago
    You can send it to me at jrmuizel@mozilla.com
  - mycelium 8y ago
    I'm on Mac OS 10.11.6 (15G22010). The regression might be due to some interaction between Firefox and changes in the OS, if you're on a more recent version.
Groxx 8y ago
I've been doing this since the first-gen rMBP, and I've never had this issue. No matter my scaling, I idle at about 5% CPU with about 80 tabs and 10 windows actively open.
JonathonW 8y ago
Can't reproduce that here (on either of two different machines)-- something else may be going on on your machine.
inetknght 8y ago
I'm not sure what it is with your MBP but my Firefox does the same thing anytime I open a phys.org page -- one Firefox at work using Ubuntu 16.04 and the other at home using Windows 7 both exhibit it. Edit: Actually I just checked and the Windows 7 machine doesn't do it any more. I'm not sure if maybe a Firefox update fixed that here. I'll double check at work tomorrow too
muizelaar 8y ago
1) I'm running a MacBook Pro (Retina, 15-inch, Mid 2015) 2) My display options are set to "more space" 3) I loaded this hackernews thread. 4) Firefox and it's child processes are using <1% cpu. If I aggressively scroll this page I can push the cpu usage up. We have a variety of fixes in the pipeline that will help with this (https://bugzilla.mozilla.org/show_bug.cgi?id=1429522 https://bugzilla.mozilla.org/show_bug.cgi?id=1429522, https://bugzilla.mozilla.org/show_bug.cgi?id=1265824 https://bugzilla.mozilla.org/show_bug.cgi?id=1265824). That being said, only 4.7% of our users are on MacOS so it's more difficult to justify prioritizing work specific to that platform.
- gpm 8y ago
  4.7% of your users, but what percentage of web developers? If that second number is substantially higher (and I personally suspect it is, but I don't have data backing that up) I think it would be very reasonable for Mozilla to prioritize MacOS. Getting web developers to use firefox makes firefox work better for everyone because it makes more sites work well with firefox.
  - bholley 8y ago
    Yeah, we do prioritize Mac disproportionately to its market share for that reason. It does sometimes happen that we have to make hard calls though, and that's what Jeff is referring to. On that note, one of the key motivators behind WebRender (our new graphics backend) is that it provides hardware acceleration across all platforms, including mac (whereas our current Direct2D acceleration works only on Windows). So Mac graphics performance should generally improve once we get that shipped.
- dwaltrip 8y ago
  First off, thank you for all of your hardwork. I'm grateful for the existence of Mozilla. However, I would like to ask... 4.7% translates to how many millions of people?
- sjwright 8y ago
  4.7% of your users... and how much of that is because you are finding it difficult to prioritise work specific to that platform? It sounds like a chicken-and-egg problem to me.

exikyut 8y ago

I don't have a rMBP, and others who may be curious to help may not either. Uploading performance profiles from various websites to the mentioned URL will give everyone something to poke at.

mycelium 8y ago

I spent a few minutes trying to replicate on a MacBook Pro (Retina, 13-inch, Late 2013) with Firefox 62.0.2 and could not replicate, browsing a number of resource intense sites.

ericabiz 8y ago
Interesting, as this is the same MBP I have, and same Firefox version. I now have a perf file showing the issue. Can't upload it (it gives an error), but would be happy to send it to anyone who thinks they can help.
- muizelaar 8y ago
  You can send it to me at jrmuizel@mozilla.com
- mycelium 8y ago
  I'm on Mac OS 10.11.6 (15G22010). The regression might be due to some interaction between Firefox and changes in the OS, if you're on a more recent version.

Groxx 8y ago

I've been doing this since the first-gen rMBP, and I've never had this issue. No matter my scaling, I idle at about 5% CPU with about 80 tabs and 10 windows actively open.

JonathonW 8y ago

Can't reproduce that here (on either of two different machines)-- something else may be going on on your machine.

inetknght 8y ago

I'm not sure what it is with your MBP but my Firefox does the same thing anytime I open a phys.org page -- one Firefox at work using Ubuntu 16.04 and the other at home using Windows 7 both exhibit it. Edit: Actually I just checked and the Windows 7 machine doesn't do it any more. I'm not sure if maybe a Firefox update fixed that here. I'll double check at work tomorrow too

muizelaar 8y ago

1) I'm running a MacBook Pro (Retina, 15-inch, Mid 2015) 2) My display options are set to "more space" 3) I loaded this hackernews thread. 4) Firefox and it's child processes are using <1% cpu. If I aggressively scroll this page I can push the cpu usage up. We have a variety of fixes in the pipeline that will help with this (https://bugzilla.mozilla.org/show_bug.cgi?id=1429522 https://bugzilla.mozilla.org/show_bug.cgi?id=1429522, https://bugzilla.mozilla.org/show_bug.cgi?id=1265824 https://bugzilla.mozilla.org/show_bug.cgi?id=1265824). That being said, only 4.7% of our users are on MacOS so it's more difficult to justify prioritizing work specific to that platform.

gpm 8y ago
4.7% of your users, but what percentage of web developers? If that second number is substantially higher (and I personally suspect it is, but I don't have data backing that up) I think it would be very reasonable for Mozilla to prioritize MacOS. Getting web developers to use firefox makes firefox work better for everyone because it makes more sites work well with firefox.
- bholley 8y ago
  Yeah, we do prioritize Mac disproportionately to its market share for that reason. It does sometimes happen that we have to make hard calls though, and that's what Jeff is referring to. On that note, one of the key motivators behind WebRender (our new graphics backend) is that it provides hardware acceleration across all platforms, including mac (whereas our current Direct2D acceleration works only on Windows). So Mac graphics performance should generally improve once we get that shipped.
dwaltrip 8y ago
First off, thank you for all of your hardwork. I'm grateful for the existence of Mozilla. However, I would like to ask... 4.7% translates to how many millions of people?
sjwright 8y ago
4.7% of your users... and how much of that is because you are finding it difficult to prioritise work specific to that platform? It sounds like a chicken-and-egg problem to me.