Bug 183218 – Add a second tier of prevalence to facilitate telemetry on very prevalent domains

• Component: WebKit Misc.
• Status: RESOLVED FIXED

[John Wilander]

We should have a second tier for very prevalent resources. This allows for more detailed telemetry but also for potential UI with "top trackers."

[Radar WebKit Bug Importer]

<rdar://problem/37992389>

[Radar WebKit Bug Importer]

<rdar://problem/37992388>

[John Wilander]

I will use <rdar://problem/37992388> as the radar equivalent.

[John Wilander]

Created attachment 334755 [details]
Patch

[Brent Fulgham]

Comment on attachment 334755 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=334755&action=review

r=me

> Source/WebKit/Platform/classifier/ResourceLoadStatisticsClassifier.cpp:37
> +    return sqrt(a * a + b * b + c * c);

This could probably be "std::hypot(a, b, c)", though I think this is known to be faster to execute.

> Source/WebKit/UIProcess/WebResourceLoadStatisticsStore.cpp:532
> +{

ASSERT(isMainThread());

> Source/WebKit/UIProcess/WebResourceLoadStatisticsStore.cpp:547
> +        resourceStatistic.isVeryPrevalentResource = true;

resourceStatistic.isVeryPrevalentResource = newPrevlance == ResourceLoadPrevalence::VeryHigh;

> Source/WebKit/UIProcess/WebResourceLoadStatisticsStore.cpp:576
> +{

ASSERT(isMainThread());

[Brent Fulgham]

Comment on attachment 334755 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=334755&action=review

>> Source/WebKit/Platform/classifier/ResourceLoadStatisticsClassifier.cpp:37
>> +    return sqrt(a * a + b * b + c * c);
> 
> This could probably be "std::hypot(a, b, c)", though I think this is known to be faster to execute.

I mean that YOUR code is known to be more efficient (if not as able to handle over/underflow).

[EWS Watchlist]

Comment on attachment 334755 [details]
Patch

Attachment 334755 [details] did not pass win-ews (win):
Output: http://webkit-queues.webkit.org/results/6708809

New failing tests:
http/tests/preload/download_resources.html

[EWS Watchlist]

Created attachment 334763 [details]
Archive of layout-test-results from ews206 for win-future

The attached test failures were seen while running run-webkit-tests on the win-ews.
Bot: ews206  Port: win-future  Platform: CYGWIN_NT-6.1-2.9.0-0.318-5-3-x86_64-64bit

[John Wilander]

Created attachment 334766 [details]
Patch for landing

[John Wilander]

Thanks, Brent! Fixed all the things you commented on.

[WebKit Commit Bot]

Comment on attachment 334766 [details]
Patch for landing

Clearing flags on attachment: 334766

Committed r229106: <https://trac.webkit.org/changeset/229106>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.

[Chris Dumez]

Comment on attachment 334766 [details]
Patch for landing

View in context: https://bugs.webkit.org/attachment.cgi?id=334766&action=review

> Source/WebKit/Platform/classifier/ResourceLoadStatisticsClassifier.cpp:55
> +        ASSERT(currentPrevalence == Low);

I am hitting this assertion in debug wk2.

[John Wilander]

(In reply to Chris Dumez from comment #13)
> Comment on attachment 334766 [details]
> Patch for landing
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=334766&action=review
> 
> > Source/WebKit/Platform/classifier/ResourceLoadStatisticsClassifier.cpp:55
> > +        ASSERT(currentPrevalence == Low);
> 
> I am hitting this assertion in debug wk2.

Thanks for the report, Chris. I removed the assertion in https://bugs.webkit.org/show_bug.cgi?id=183233.