RESOLVED FIXED 204901
Web Inspector: http/tests/inspector/target/pause-on-inline-debugger-statement.html is crashing in debug
https://bugs.webkit.org/show_bug.cgi?id=204901
Summary Web Inspector: http/tests/inspector/target/pause-on-inline-debugger-statement...
Yury Semikhatsky
Reported 2019-12-05 10:52:05 PST
Triggers same assertion on all platforms: https://results.webkit.org/?suite=layout-tests&test=http%2Ftests%2Finspector%2Ftarget%2Fpause-on-inline-debugger-statement.html stderr: ASSERTION FAILED: m_reasonForSuspendingActiveDOMObjects == ReasonForSuspension::PageWillBeSuspended ./dom/ScriptExecutionContext.cpp(257) : virtual void WebCore::ScriptExecutionContext::suspendActiveDOMObjects(WebCore::ReasonForSuspension) 1 0x132210619 WTFCrash 2 0x11828d88b WTFCrashWithInfo(int, char const*, char const*, int) 3 0x11ac6f979 WebCore::ScriptExecutionContext::suspendActiveDOMObjects(WebCore::ReasonForSuspension) 4 0x11aa88ed4 WebCore::Document::suspendActiveDOMObjects(WebCore::ReasonForSuspension) 5 0x11aa869a9 WebCore::Document::suspendScheduledTasks(WebCore::ReasonForSuspension) 6 0x11b85d1aa WebCore::PageGroupLoadDeferrer::PageGroupLoadDeferrer(WebCore::Page&, bool) 7 0x11b85d41c WebCore::PageGroupLoadDeferrer::PageGroupLoadDeferrer(WebCore::Page&, bool) 8 0x11b723649 WebCore::Chrome::runJavaScriptAlert(WebCore::Frame&, WTF::String const&) 9 0x11b74228f WebCore::DOMWindow::alert(WTF::String const&) 10 0x118b0987d WebCore::jsDOMWindowInstanceFunctionAlert2Body(JSC::JSGlobalObject*, JSC::CallFrame*, WebCore::JSDOMWindow*, JSC::ThrowScope&) 11 0x118b096a7 WebCore::jsDOMWindowInstanceFunctionAlertOverloadDispatcher(JSC::JSGlobalObject*, JSC::CallFrame*, WebCore::JSDOMWindow*, JSC::ThrowScope&) 12 0x118a13d4f long long WebCore::IDLOperation<WebCore::JSDOMWindow>::call<&(WebCore::jsDOMWindowInstanceFunctionAlertOverloadDispatcher(JSC::JSGlobalObject*, JSC::CallFrame*, WebCore::JSDOMWindow*, JSC::ThrowScope&)), (WebCore::CastedThisErrorBehavior)0>(JSC::JSGlobalObject&, JSC::CallFrame&, char const*) 13 0x118a13a34 WebCore::jsDOMWindowInstanceFunctionAlert(JSC::JSGlobalObject*, JSC::CallFrame*) 14 0x388079c0116b 15 0x13271ffbc llint_entry 16 0x13271ffbc llint_entry 17 0x13271ffbc llint_entry 18 0x132703133 vmEntryToJavaScript 19 0x1334c9947 JSC::JITCode::execute(JSC::VM*, JSC::ProtoCallFrame*) 20 0x1334c8f59 JSC::Interpreter::executeProgram(JSC::SourceCode const&, JSC::JSGlobalObject*, JSC::JSObject*) 21 0x133830a3c JSC::evaluate(JSC::JSGlobalObject*, JSC::SourceCode const&, JSC::JSValue, WTF::NakedPtr<JSC::Exception>&) 22 0x133830ca1 JSC::evaluateWithScopeExtension(JSC::JSGlobalObject*, JSC::SourceCode const&, JSC::JSObject*, WTF::NakedPtr<JSC::Exception>&) 23 0x133445734 Inspector::JSInjectedScriptHost::evaluateWithScopeExtension(JSC::JSGlobalObject*, JSC::CallFrame*) 24 0x13344ebd3 Inspector::jsInjectedScriptHostPrototypeFunctionEvaluateWithScopeExtension(JSC::JSGlobalObject*, JSC::CallFrame*) 25 0x388079c0116b 26 0x13271ffbc llint_entry 27 0x13271ffbc llint_entry 28 0x132703133 vmEntryToJavaScript 29 0x1334c9947 JSC::JITCode::execute(JSC::VM*, JSC::ProtoCallFrame*) 30 0x1334c9fa4 JSC::Interpreter::executeCall(JSC::JSGlobalObject*, JSC::JSObject*, JSC::CallType, JSC::CallData const&, JSC::JSValue, JSC::ArgList const&) 31 0x1337d2e2c JSC::call(JSC::JSGlobalObject*, JSC::JSValue, JSC::CallType, JSC::CallData const&, JSC::JSValue, JSC::ArgList const&) LEAK: 2 WebPageProxy
Attachments
Patch (3.79 KB, patch)
2019-12-05 13:10 PST, Yury Semikhatsky
no flags
Yury Semikhatsky
Comment 1 2019-12-05 13:10:20 PST
Devin Rousso
Comment 2 2019-12-05 14:38:11 PST
Comment on attachment 384937 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=384937&action=review > LayoutTests/ChangeLog:8 > + Restructured the test to avoid inadvertent alert() when navigating to a new Can you explain what "inadvertent alert()" means? It sounds like the issue was that logging during a page navigation triggered an `ASSERT` somewhere? Is this test just masking the problem, in that we need to change non-test logic somewhere, or is it purely that the test is doing something normally impossible? > LayoutTests/http/tests/inspector/target/pause-on-inline-debugger-statement.html:16 > + const url = "http://localhost:8000/inspector/target/resources/inline-debugger-statement.html"; > + WI.mainTarget.PageAgent.navigate(url); NIT: you could inline this
Yury Semikhatsky
Comment 3 2019-12-05 14:53:49 PST
Comment on attachment 384937 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=384937&action=review >> LayoutTests/ChangeLog:8 >> + Restructured the test to avoid inadvertent alert() when navigating to a new > > Can you explain what "inadvertent alert()" means? It sounds like the issue was that logging during a page navigation triggered an `ASSERT` somewhere? Is this test just masking the problem, in that we need to change non-test logic somewhere, or is it purely that the test is doing something normally impossible? The test revealed existing problem: when when execution is paused on debugger statement in inline script evaluating 'alert(1)' in console would trigger the assert. Since this is unrelated to PSON I suggest we address it separately, filed https://bugs.webkit.org/show_bug.cgi?id=204923. This test should not trigger assert while printing output in the first place. >> LayoutTests/http/tests/inspector/target/pause-on-inline-debugger-statement.html:16 >> + WI.mainTarget.PageAgent.navigate(url); > > NIT: you could inline this I'd rather leave it as named variable for documentation purposes.
Devin Rousso
Comment 4 2019-12-05 14:58:42 PST
Comment on attachment 384937 [details] Patch r=me, please wait for all-green EWS
WebKit Commit Bot
Comment 5 2019-12-05 15:48:48 PST
Comment on attachment 384937 [details] Patch Clearing flags on attachment: 384937 Committed r253179: <https://trac.webkit.org/changeset/253179>
WebKit Commit Bot
Comment 6 2019-12-05 15:48:49 PST
All reviewed patches have been landed. Closing bug.
Radar WebKit Bug Importer
Comment 7 2019-12-05 15:49:22 PST
Note You need to log in before you can comment on or make changes to this bug.