Re: Web Smoothness

In my POC I'm leaving the 'duration' field of PerformanceEntry un-set, but
that could also be set to the time spent performing that action, and then
you could see the breakdown.

If there is a separate event generated on the main thread versus the
render/layout/impl thread then it would be pretty easy to compare the
durations of the two. If layout happens on the main thread, you'd see a
large chunk of time there, and a relatively small amount of time on other
events.

(I think it's also worth noting that some events can be handled entirely
off-main-thread.)


On Fri, Sep 19, 2014 at 12:52 PM, Boris Zbarsky <bzbarsky@mit.edu> wrote:

> On 9/19/14, 3:47 PM, Ben Maurer wrote:
>
>> This would be fantastic. One quick comment -- it'd be great if there was
>> some kind of breakdown of the time spent on the main thread (JS
>> execution vs layout
>>
>
> You just made the assumption that layout happens on the main thread.
>
> This is not a good assumption, especially to bake into standards.
>
> -Boris
>
>