Bug 102971 – REGRESSION(r135399): Wrong assertion causing an assertion failure: m_rawAttributes.initialScale > 0

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

[Thiago Marcos P. Santos]

crash log for WebKitTestRunner (pid 9248):
STDOUT: <empty>
STDERR: ASSERTION FAILED: m_rawAttributes.initialScale > 0
STDERR: /home/buildslave-1/webkit-buildslave/efl-linux-64-debug-wk2/build/Source/WebKit2/UIProcess/PageViewportController.cpp(152) : void WebKit::PageViewportController::pageTransitionViewportReady()
STDERR: 1   0x7f0bd508cd5a WebKit::PageViewportController::pageTransitionViewportReady()
STDERR: 2   0x7f0bd52b0579 WebKit::PageClientDefaultImpl::pageTransitionViewportReady()
STDERR: 3   0x7f0bd50e2535 WebKit::WebPageProxy::pageTransitionViewportReady()
STDERR: 4   0x7f0bd52e889e void CoreIPC::callMemberFunction<WebKit::WebPageProxy, void (WebKit::WebPageProxy::*)()>(CoreIPC::Arguments0 const&, WebKit::WebPageProxy*, void (WebKit::WebPageProxy::*)())
STDERR: 5   0x7f0bd52e4eaa void CoreIPC::handleMessage<Messages::WebPageProxy::PageTransitionViewportReady, WebKit::WebPageProxy, void (WebKit::WebPageProxy::*)()>(CoreIPC::MessageDecoder&, WebKit::WebPageProxy*, void (WebKit::WebPageProxy::*)())
STDERR: 6   0x7f0bd52de2bd WebKit::WebPageProxy::didReceiveWebPageProxyMessage(CoreIPC::Connection*, CoreIPC::MessageID, CoreIPC::MessageDecoder&)
STDERR: 7   0x7f0bd50dd8d9 WebKit::WebPageProxy::didReceiveMessage(CoreIPC::Connection*, CoreIPC::MessageID, CoreIPC::MessageDecoder&)
STDERR: 8   0x7f0bd511e9f5 WebKit::WebProcessProxy::didReceiveMessage(CoreIPC::Connection*, CoreIPC::MessageID, CoreIPC::MessageDecoder&)
STDERR: 9   0x7f0bd5006e64 CoreIPC::Connection::dispatchMessage(CoreIPC::MessageID, CoreIPC::MessageDecoder&)
STDERR: 10  0x7f0bd5006fd0 CoreIPC::Connection::dispatchMessage(CoreIPC::Connection::Message<CoreIPC::MessageDecoder>&)
STDERR: 11  0x7f0bd500721b CoreIPC::Connection::dispatchOneMessage()
STDERR: 12  0x7f0bd50113bc WTF::FunctionWrapper<void (CoreIPC::Connection::*)()>::operator()(CoreIPC::Connection*)
STDERR: 13  0x7f0bd50111c2 WTF::BoundFunctionImpl<WTF::FunctionWrapper<void (CoreIPC::Connection::*)()>, void (CoreIPC::Connection*)>::operator()()
STDERR: 14  0x7f0bdc1b8dca WTF::Function<void ()>::operator()() const
STDERR: 15  0x7f0bd867fc5e WebCore::RunLoop::performWork()
STDERR: 16  0x7f0bd909d69b WebCore::RunLoop::wakeUpEvent(void*, void*, unsigned int)
STDERR: 17  0x7f0bd441a621
STDERR: 18  0x7f0bd4419571
STDERR: 19  0x7f0bd4419ab7 ecore_main_loop_begin
STDERR: 20  0x43362b WTR::TestController::platformRunUntil(bool&, double)
STDERR: 21  0x41e94e WTR::TestController::runUntil(bool&, WTR::TestController::TimeoutDuration)
STDERR: 22  0x425a06 WTR::TestInvocation::invoke()
STDERR: 23  0x41e686 WTR::TestController::runTest(char const*)
STDERR: 24  0x41e7bf WTR::TestController::runTestingServerLoop()
STDERR: 25  0x41e859 WTR::TestController::run()
STDERR: 26  0x41c291 WTR::TestController::TestController(int, char const**)
STDERR: 27  0x4337c6 main
STDERR: 28  0x7f0bd2f7a76d __libc_start_main
STDERR: 29  0x41ac09
STDERR: LEAK: 2 CachedResource

[Andras Becsi]

Ouch, the assertion is not actually wrong, but my patch was not complete.
I think I know what's going on, and can fix it.

Mikhail, I see you assigned it to yourself, do you mind if I clean up my mess instead?

[Mikhail Pozdnyakov]

Created attachment 175632 [details]
patch

[Mikhail Pozdnyakov]

(In reply to comment #1)
> Ouch, the assertion is not actually wrong, but my patch was not complete.
> I think I know what's going on, and can fix it.
> 
> Mikhail, I see you assigned it to yourself, do you mind if I clean up my mess instead?

sorry, did not see your message. Uploaded already a patch, feel free to take the bug if you see the patch does not suffice or wrong.

[Andras Becsi]

Comment on attachment 175632 [details]
patch

View in context: https://bugs.webkit.org/attachment.cgi?id=175632&action=review

I think this is the right fix. The patch looks good to me.

> Source/WebKit2/ChangeLog:9
> +        m_rawAttributes.initialScale has to be initialized even if m_minimumScaleToFit hasn't changed
> +        it's value.

Indeed, the basic problem was that m_minimumScaleToFit is initialized to 1 and if the calculated minimum scale to fit turned out to be 1 we wrongly returned early and did not initialize the initialScale attribute.

[Raphael Kubo da Costa (:rakuco)]

Comment on attachment 175632 [details]
patch

Clearing flags on attachment: 175632

Committed r135504: <http://trac.webkit.org/changeset/135504>

[Raphael Kubo da Costa (:rakuco)]

All reviewed patches have been landed.  Closing bug.

[Thiago Marcos P. Santos]

*** Bug 102922 has been marked as a duplicate of this bug. ***