Bug 148533 – REGRESSION (r188820): fast/dom/HTMLObjectElement/object-as-frame.html is flaky

• Status: RESOLVED
• Resolution: FIXED
• Component: Page Loading

[Alexey Proskuryakov]

fast/dom/HTMLObjectElement/object-as-frame.html become flaky after the patch for decoding data URLs in WebContent process.

This is hard to reproduce locally, but it happens frequently enough on bots 

I could sometimes reproduce on El Capitan by running tests like this:

run-webkit-tests fast/dom/HTMLObjectElement -f --repeat 100 --no-retry

https://webkit-test-results.webkit.org/dashboards/flakiness_dashboard.html#showAllRuns=true&showExpectations=true&revision=188926&tests=fast%2Fdom%2FHTMLObjectElement%2Fobject-as-frame.html

@@ -4,5 +4,5 @@
 Loading 'data:text/html,Updated contents of iframe' in iframe...
 Loading 'data:text/html,Updated contents of object' in object...
 PASS window.frames[0].location.toString() is "data:text/html,Updated contents of iframe"
-PASS window.frames[1].location.toString() is "data:text/html,Updated contents of object"
+FAIL window.frames[1].location.toString() should be data:text/html,Updated contents of object. Was data:text/html,New contents of frame.

[Radar WebKit Bug Importer]

<rdar://problem/22459237>

[Alexey Proskuryakov]

Marked as flaky in r189041.

[Antti Koivisto]

This is an existing bug where <object> may cause document load event to be fire too early. Faster data URL decoding just changed timings and made it more likely here.

[Antti Koivisto]

Created attachment 260147 [details]
patch

[Build Bot]

Comment on attachment 260147 [details]
patch

Attachment 260147 [details] did not pass mac-wk2-ews (mac-wk2):
Output: http://webkit-queues.webkit.org/results/116618

Number of test failures exceeded the failure limit.

[Build Bot]

Created attachment 260148 [details]
Archive of layout-test-results from ews106 for mac-mavericks-wk2

The attached test failures were seen while running run-webkit-tests on the mac-wk2-ews.
Bot: ews106  Port: mac-mavericks-wk2  Platform: Mac OS X 10.9.5

[Build Bot]

Comment on attachment 260147 [details]
patch

Attachment 260147 [details] did not pass mac-ews (mac):
Output: http://webkit-queues.webkit.org/results/116648

Number of test failures exceeded the failure limit.

[Build Bot]

Created attachment 260149 [details]
Archive of layout-test-results from ews102 for mac-mavericks

The attached test failures were seen while running run-webkit-tests on the mac-ews.
Bot: ews102  Port: mac-mavericks  Platform: Mac OS X 10.9.5

[Antti Koivisto]

Created attachment 260150 [details]
patch

[Build Bot]

Comment on attachment 260150 [details]
patch

Attachment 260150 [details] did not pass mac-ews (mac):
Output: http://webkit-queues.webkit.org/results/116738

New failing tests:
webarchive/loading/object.html
http/tests/loading/basic-auth-resend-wrong-credentials.html
http/tests/loading/basic-credentials-sent-automatically.html

[Build Bot]

Created attachment 260151 [details]
Archive of layout-test-results from ews102 for mac-mavericks

The attached test failures were seen while running run-webkit-tests on the mac-ews.
Bot: ews102  Port: mac-mavericks  Platform: Mac OS X 10.9.5

[Build Bot]

Comment on attachment 260150 [details]
patch

Attachment 260150 [details] did not pass mac-wk2-ews (mac-wk2):
Output: http://webkit-queues.webkit.org/results/116745

New failing tests:
webarchive/loading/object.html
http/tests/loading/basic-auth-resend-wrong-credentials.html
http/tests/loading/basic-credentials-sent-automatically.html

[Build Bot]

Created attachment 260152 [details]
Archive of layout-test-results from ews104 for mac-mavericks-wk2

The attached test failures were seen while running run-webkit-tests on the mac-wk2-ews.
Bot: ews104  Port: mac-mavericks-wk2  Platform: Mac OS X 10.9.5

[Antti Koivisto]

Created attachment 260153 [details]
patch

[Chris Dumez]

Comment on attachment 260153 [details]
patch

View in context: https://bugs.webkit.org/attachment.cgi?id=260153&action=review

r=me

> Source/WebCore/ChangeLog:8
> +        <object> element may cause document load even fire too early. By making data url

"event", "data URI".

[Antti Koivisto]

"URL" is the correct term. "URI" should not be used, see https://url.spec.whatwg.org

[Antti Koivisto]

https://trac.webkit.org/r189164