Bug 167717 – Support Performance API (performance.now(), UserTiming) in Workers

• Status: RESOLVED
• Resolution: FIXED
• Component: WebCore Misc.

[Joseph Pecoraro]

Summary:
Support Performance API (performance.now(), UserTiming) in Workers

Level 2 of these specifications brings support to workers.

[Joseph Pecoraro]

Created attachment 300374 [details]
[PATCH] Proposed Fix

[Sam Weinig]

Comment on attachment 300374 [details]
[PATCH] Proposed Fix

Can we use WTF's Seconds class instead of double for all this timeOrigin passing?

[Joseph Pecoraro]

Created attachment 300377 [details]
[PATCH] Proposed Fix

[Joseph Pecoraro]

(In reply to comment #2)
> Comment on attachment 300374 [details]
> [PATCH] Proposed Fix
> 
> Can we use WTF's Seconds class instead of double for all this timeOrigin
> passing?

Sure! I'll look into that and put up another patch.

[Ryosuke Niwa]

Comment on attachment 300377 [details]
[PATCH] Proposed Fix

View in context: https://bugs.webkit.org/attachment.cgi?id=300377&action=review

restrictedMarkFunction in PerformanceUserTiming.cpp is not thread safe.
Please fix that before landing this patch.

> Source/WebCore/page/GlobalPerformance.idl:2
> - * Copyright (C) 2012 Intel Inc. All rights reserved.
> + * Copyright (C) 2017 Apple Inc. All rights reserved.

I don't think we can get rid of Intel's copyright.

[Joseph Pecoraro]

Comment on attachment 300377 [details]
[PATCH] Proposed Fix

View in context: https://bugs.webkit.org/attachment.cgi?id=300377&action=review

>> Source/WebCore/page/GlobalPerformance.idl:2
>> + * Copyright (C) 2017 Apple Inc. All rights reserved.
> 
> I don't think we can get rid of Intel's copyright.

That is because my git diff thinks we are copying this file from PerformanceMeasure.idl, but we aren't. This is a new file with WebIDL just copied from the spec (extracted from Performance.idl).

[Joseph Pecoraro]

Created attachment 300426 [details]
[PATCH] Additional Changes

These additional changes (mostly tests and a minor tweak to User Timing for Workers) came up because of some comments rniwa raised in person.

I haven't forgotten about moving to MonotonicTime / Seconds classes. I have that patch right now but I'll put it up for review as a follow-up since it reaches into more and more places not relevant for this bug. Also, converting everything related to Performance APIs at once will be nice.

[Joseph Pecoraro]

Created attachment 300428 [details]
[PATCH] For Bots

[Build Bot]

Comment on attachment 300428 [details]
[PATCH] For Bots

Attachment 300428 [details] did not pass mac-ews (mac):
Output: http://webkit-queues.webkit.org/results/2993036

New failing tests:
imported/w3c/web-platform-tests/user-timing/test_user_timing_mark_and_measure_exception_when_invoke_with_timing_attributes.html

[Build Bot]

Created attachment 300438 [details]
Archive of layout-test-results from ews103 for mac-elcapitan

The attached test failures were seen while running run-webkit-tests on the mac-ews.
Bot: ews103  Port: mac-elcapitan  Platform: Mac OS X 10.11.6

[Build Bot]

Comment on attachment 300428 [details]
[PATCH] For Bots

Attachment 300428 [details] did not pass mac-debug-ews (mac):
Output: http://webkit-queues.webkit.org/results/2993051

New failing tests:
imported/w3c/web-platform-tests/user-timing/test_user_timing_mark_and_measure_exception_when_invoke_with_timing_attributes.html

[Build Bot]

Created attachment 300441 [details]
Archive of layout-test-results from ews112 for mac-elcapitan

The attached test failures were seen while running run-webkit-tests on the mac-debug-ews.
Bot: ews112  Port: mac-elcapitan  Platform: Mac OS X 10.11.6

[Joseph Pecoraro]

Created attachment 300442 [details]
[PATCH] For Bots

[Ryosuke Niwa]

Comment on attachment 300426 [details]
[PATCH] Additional Changes

Sure, this change looks good to me.

[Joseph Pecoraro]

<https://trac.webkit.org/changeset/211594>