[Webkit-unassigned] [Bug 193730] Web Inspector: CPU Usage Timeline

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Thu Jan 24 16:05:04 PST 2019


--- Comment #17 from Joseph Pecoraro <joepeck at webkit.org> ---
(In reply to Benjamin Poulain from comment #11)
> First, thanks for creating the new track. That is kick ass.
> Second, some popular sites really suck :(
> Some ideas:
> -Would it be possible to click on one of the rect to select exactly that
> zone? That way one can pick the highest rect, and see everything that
> happened (layout, js, etc).

Implementing it should be easy, so I'll look into it.

This exact situation is something we're debating given the placement of the bars right now. So when approaching this we may want to move the bars.

Currently we place the bars around when the sample happened. In this case selection would include some time before the same and some time after the sample:

    |   |
    |   |
    | + |

Devin suggested has moving the entire bar preceding the sample timestamp. In this case selection would include only time before the sample (which is arguably what caused the CPU usage to spike that high).

    |   |
    |   |
    |   +

Do you have any thoughts about which approach might be better?

> -The main view when clicking on the track looks a bit small on iMac where
> the display is very large. Maybe it should resize with the window?

This sounds like:
<https://webkit.org/b/153758> Web Inspector: Memory / CPU Timeline Views should be responsive / resizable

> -There is some space below the graph, we could eventually add more
> information (# timers, etc). What do you think?

Yep. Feel free to throw suggestions my way. I definitely want to add more content to the main view.

> -I get artifacts from time to time. See radar.

Will look.

You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20190125/1c50e0e7/attachment-0001.html>

More information about the webkit-unassigned mailing list