When trying to run the test accessibility/canvas-fallback-content-2.html in content_shell like this: echo file://$(pwd)/LayoutTests/accessibility/canvas-fallback-content-2.html | out/Debug/content_shell --no-sandbox --dump-render-tree I get the following crash: [20467:20467:1001/103914:9811384406:ERROR:process_util_posix.cc(144)] Received signal 11 base::debug::StackTrace::StackTrace() [0x50cf7e] base::(anonymous namespace)::StackDumpSignalHandler() [0x524df4] 0x7f602d4eb4c0 WTF::StringImpl::rawHash() [0xc7520a] WTF::StringImpl::hasHash() [0xc751e5] WTF::StringImpl::existingHash() [0xc75169] WTF::AtomicStringHash::hash() [0xc7513d] WTF::HashMapTranslator<>::hash<>() [0x3042bf5] WTF::HashTable<>::add<>() [0x30427cd] WTF::HashMap<>::inlineAdd() [0x30426cc] WTF::HashMap<>::add() [0x30424a4] WebCore::Document::getCachedLocalizer() [0x303b928] WebCore::Element::localizer() [0x308d50d] WebCore::NumberInputType::localizeValue() [0x2ddc975] WebCore::NumberInputType::visibleValue() [0x2ddca31] WebCore::TextFieldInputType::updateInnerTextValue() [0x2de7e46] WebCore::HTMLInputElement::updateType() [0x2d75e1c] WebCore::HTMLInputElement::parseAttribute() [0x2d7710f] WebCore::Element::attributeChanged() [0x30888e9] WebCore::StyledElement::attributeChanged() [0x311a349] WebCore::Element::parserSetAttributes() [0x308943a] WebCore::HTMLConstructionSite::createHTMLElement() [0x2eb6856] WebCore::HTMLConstructionSite::insertSelfClosingHTMLElement() [0x2eb6dcf] WebCore::HTMLTreeBuilder::processStartTagForInBody() [0x2e4c85f] WebCore::HTMLTreeBuilder::processStartTag() [0x2e46426] WebCore::HTMLTreeBuilder::processToken() [0x2e45ae1] WebCore::HTMLTreeBuilder::constructTreeFromAtomicToken() [0x2e44e1b] WebCore::HTMLTreeBuilder::constructTreeFromToken() [0x2e44d1a] WebCore::HTMLDocumentParser::pumpTokenizer() [0x2e2ab66] WebCore::HTMLDocumentParser::pumpTokenizerIfPossible() [0x2e2a5f5] WebCore::HTMLDocumentParser::resumeParsingAfterScriptExecution() [0x2e2ba21] WebCore::HTMLDocumentParser::notifyFinished() [0x2e2bdc7] WebCore::HTMLDocumentParser::notifyFinished() [0x2e2be2f] WebCore::CachedResource::checkNotify() [0x28e97dd] WebCore::CachedScript::data() [0x2904118] WebCore::SubresourceLoader::didFinishLoading() [0x28cb94a] WebCore::ResourceLoader::didFinishLoading() [0x28c5705] WebCore::ResourceHandleInternal::didFinishLoading() [0x39a2624] webkit_glue::WebURLLoaderImpl::Context::OnCompletedRequest() [0x1d40e54] content::ResourceDispatcher::OnRequestComplete() [0xbeb4bd] DispatchToMethod<>() [0xbecde8] ResourceMsg_RequestComplete::Dispatch<>() [0xbec846] content::ResourceDispatcher::DispatchMessage() [0xbea227] content::ResourceDispatcher::OnMessageReceived() [0xbe962a] ChildThread::OnMessageReceived() [0xacca3b] IPC::ChannelProxy::Context::OnDispatchMessage() [0x2091b59] base::internal::RunnableAdapter<>::Run() [0x2095c7a] base::internal::InvokeHelper<>::MakeItSo() [0x2095be1] base::internal::Invoker<>::Run() [0x2095b8c] base::Callback<>::Run() [0x42be1e] MessageLoop::RunTask() [0x49bcbc] MessageLoop::DeferOrRunPendingTask() [0x49c0ab] MessageLoop::DoWork() [0x49c255] base::MessagePumpDefault::Run() [0x4a6168] MessageLoop::RunInternal() [0x49b6e6] MessageLoop::RunHandler() [0x49b595] base::RunLoop::Run() [0x4bd5c2] MessageLoop::Run() [0x49ae31] RendererMain() [0xd16610] content::RunZygote() [0xca6297] content::RunNamedProcessTypeMain() [0xca6594] content::ContentMainRunnerImpl::Run() [0xca738d] Also tracked in https://code.google.com/p/chromium/issues/detail?id=153248
Created attachment 166471 [details] Patch
Comment on attachment 166471 [details] Patch Clearing flags on attachment: 166471 Committed r130062: <http://trac.webkit.org/changeset/130062>
All reviewed patches have been landed. Closing bug.
Reverted r130062 for reason: Causes a ton of gtest failures on the chromium bots Committed r130101: <http://trac.webkit.org/changeset/130101>
See http://build.chromium.org/p/chromium.webkit/builders/Win%20%28dbg%29/builds/7893 for the failures.
It wasn't clear to me whether this assert was showing places where the code is wrong or whether the assert is wrong. Either way, it would be good to fix the gtest failures first.
(In reply to comment #6) > It wasn't clear to me whether this assert was showing places where the code is wrong or whether the assert is wrong. Either way, it would be good to fix the gtest failures first. sorry for the mess. It's indeed a failure in the tests. I'm fixing this.
Landed again http://trac.webkit.org/changeset/130152