Bug 227256 – REGRESSION: [iOS] ASSERTION FAILED: pixelBuffer->data().length() >= 4 under WebCore::sampleColor

• Status: RESOLVED
• Resolution: FIXED
• Component: New Bugs

[Devin Rousso]

failing on internal iOS Simulator EWS

ASSERTION FAILED: pixelBuffer->data().length() >= 4
./page/PageColorSampler.cpp(131) : std::optional<Lab<float>> WebCore::sampleColor(WebCore::Document &, WebCore::IntPoint &&)
1   0x7bab979b9 WTFCrash
2   0x7db40babb WTFCrashWithInfo(int, char const*, char const*, int)
3   0x7dee9ae00 WebCore::sampleColor(WebCore::Document&, WebCore::IntPoint&&)
4   0x7dee91e8b WebCore::PageColorSampler::sampleTop(WebCore::Page&)
5   0x7dee91404 WebCore::Page::doAfterUpdateRendering()
6   0x7dee90c3e WebCore::Page::updateRendering()
7   0x7a4083941 WebKit::WebPage::updateRendering()
8   0x7a27b5cc9 WebKit::RemoteLayerTreeDrawingArea::updateRendering()
9   0x7a27bffee decltype(*(std::__1::forward<WebKit::RemoteLayerTreeDrawingArea*&>(fp0)).*fp()) std::__1::__invoke<void (WebKit::RemoteLayerTreeDrawingArea::*&)(), WebKit::RemoteLayerTreeDrawingArea*&, void>(void (WebKit::RemoteLayerTreeDrawingArea::*&)(), WebKit::RemoteLayerTreeDrawingArea*&)
10  0x7a27bff6d std::__1::__bind_return<void (WebKit::RemoteLayerTreeDrawingArea::*)(), std::__1::tuple<WebKit::RemoteLayerTreeDrawingArea*>, std::__1::tuple<>, __is_valid_bind_return<void (WebKit::RemoteLayerTreeDrawingArea::*)(), std::__1::tuple<WebKit::RemoteLayerTreeDrawingArea*>, std::__1::tuple<> >::value>::type std::__1::__apply_functor<void (WebKit::RemoteLayerTreeDrawingArea::*)(), std::__1::tuple<WebKit::RemoteLayerTreeDrawingArea*>, 0ul, std::__1::tuple<> >(void (WebKit::RemoteLayerTreeDrawingArea::*&)(), std::__1::tuple<WebKit::RemoteLayerTreeDrawingArea*>&, std::__1::__tuple_indices<0ul>, std::__1::tuple<>&&)
11  0x7a27bff20 std::__1::__bind_return<void (WebKit::RemoteLayerTreeDrawingArea::*)(), std::__1::tuple<WebKit::RemoteLayerTreeDrawingArea*>, std::__1::tuple<>, __is_valid_bind_return<void (WebKit::RemoteLayerTreeDrawingArea::*)(), std::__1::tuple<WebKit::RemoteLayerTreeDrawingArea*>, std::__1::tuple<> >::value>::type std::__1::__bind<void (WebKit::RemoteLayerTreeDrawingArea::*&)(), WebKit::RemoteLayerTreeDrawingArea*>::operator()<>()
12  0x7a27bfeb9 WTF::Detail::CallableWrapper<std::__1::__bind<void (WebKit::RemoteLayerTreeDrawingArea::*&)(), WebKit::RemoteLayerTreeDrawingArea*>, void>::call()
13  0x7a20d0532 WTF::Function<void ()>::operator()() const
14  0x7a27baed9 WebCore::Timer::fired()
15  0x7df08785f WebCore::ThreadTimers::sharedTimerFiredInternal()
16  0x7df091801 WebCore::ThreadTimers::setSharedTimer(WebCore::SharedTimer*)::$_0::operator()() const
17  0x7df091799 WTF::Detail::CallableWrapper<WebCore::ThreadTimers::setSharedTimer(WebCore::SharedTimer*)::$_0, void>::call()
18  0x7db422942 WTF::Function<void ()>::operator()() const
19  0x7df045501 WebCore::MainThreadSharedTimer::fired()
20  0x7df0fbbe6 WebCore::timerFired(__CFRunLoopTimer*, void*)
21  0x79051a20d __CFRUNLOOP_IS_CALLING_OUT_TO_A_TIMER_CALLBACK_FUNCTION__
22  0x790519d09 __CFRunLoopDoTimer
23  0x7905192c8 __CFRunLoopDoTimers
24  0x7905138f6 __CFRunLoopRun
25  0x790512c39 CFRunLoopRunSpecific
26  0x790e47111 -[NSRunLoop(NSRunLoop) runMode:beforeDate:]
27  0x790e4732e -[NSRunLoop(NSRunLoop) run]
28  0x791af90f7 _xpc_objc_main
29  0x791afb0b9 xpc_main
30  0x7a2dc82c5 WebKit::XPCServiceMain(int, char const**)
31  0x7a448fe2b WKXPCServiceMain

[Devin Rousso]

<rdar://problem/79584286>

[Devin Rousso]

Created attachment 431969 [details]
Patch

[Alexey Proskuryakov]

Comment on attachment 431969 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=431969&action=review

> Source/WebCore/ChangeLog:7
> +        Unreviewed, internal text fix.

"text"? Also, is this something that can be tested for in open source?

[Devin Rousso]

Comment on attachment 431969 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=431969&action=review

>> Source/WebCore/ChangeLog:7
>> +        Unreviewed, internal text fix.
> 
> "text"? Also, is this something that can be tested for in open source?

oops will fix typo :P

I'm still trying to figure out why exactly this is happening, as it only reproduces when the entire test suite is run.  Not really sure what's going on.  I put this patch up (and am planning on landing it) so that the internal EWS doesn't have to deal with this anymore.

[Devin Rousso]

Created attachment 431985 [details]
Patch

[EWS]

Committed r279148 (239056@main): <https://commits.webkit.org/239056@main>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 431985 [details].