test_bug618176.xul fails on Windows (7, XP & 8) with runByDir enabled. Here's a link to try : https://treeherder.mozilla.org/#/jobs?repo=try&revision=313ffe0c5721&filter-resultStatus=testfailed&filter-resultStatus=busted&filter-resultStatus=exception&filter-classifiedState=unclassified&filter-searchStr=win It can be reproduced by : ./mach mochitest js/xpconnect/tests/chrome 21:25:05 INFO - 2798 INFO TEST-START | js/xpconnect/tests/chrome/test_bug618176.xul 21:25:05 INFO - ++DOMWINDOW == 108 (0D243F80) [pid = 1260] [serial = 108] [outer = 1C35CD00] 21:25:05 INFO - ++DOCSHELL 123EFD00 == 18 [pid = 1260] [id = 30] 21:25:05 INFO - ++DOMWINDOW == 109 (0E901F00) [pid = 1260] [serial = 109] [outer = 00000000] 21:25:05 INFO - ++DOMWINDOW == 110 (0E902680) [pid = 1260] [serial = 110] [outer = 0E901F00] 21:25:05 INFO - ++DOCSHELL 125FDA00 == 19 [pid = 1260] [id = 31] 21:25:05 INFO - ++DOMWINDOW == 111 (0E905880) [pid = 1260] [serial = 111] [outer = 00000000] 21:25:05 INFO - ++DOMWINDOW == 112 (0EAC3F00) [pid = 1260] [serial = 112] [outer = 0E905880] 21:25:05 INFO - ++DOMWINDOW == 113 (0EAC4400) [pid = 1260] [serial = 113] [outer = 0E905880] 21:25:05 INFO - [1260] ###!!! ASSERTION: Compositor backend changed.: 'mCompositorBackend == LayersBackend::LAYERS_NONE', file c:/builds/moz2_slave/try-w32-d-00000000000000000000/build/src/gfx/thebes/gfxPlatform.cpp, line 2211 21:25:05 INFO - #01: nsBaseWidget::CreateCompositor(int,int) [widget/nsBaseWidget.cpp:1136] 21:25:05 INFO - #02: nsBaseWidget::CreateCompositor() [widget/nsBaseWidget.cpp:867] 21:25:05 INFO - #03: nsWindow::GetLayerManager(mozilla::layers::PLayerTransactionChild *,mozilla::layers::LayersBackend,nsIWidget::LayerManagerPersistence,bool *) [widget/windows/nsWindow.cpp:3547] 21:25:05 INFO - #04: nsWindow::OnPaint(HDC__ *,unsigned int) [widget/windows/nsWindowGfx.cpp:223] 21:25:05 INFO - #05: nsWindow::ProcessMessage(unsigned int,unsigned int &,long &,long *) [widget/windows/nsWindow.cpp:5073] 21:25:05 INFO - #06: nsWindow::WindowProcInternal(HWND__ *,unsigned int,unsigned int,long) [widget/windows/nsWindow.cpp:4612] 21:25:05 INFO - #07: CallWindowProcCrashProtected [xpcom/base/nsCrashOnException.cpp:35] 21:25:05 INFO - #08: nsWindow::WindowProc(HWND__ *,unsigned int,unsigned int,long) [widget/windows/nsWindow.cpp:4564] 21:25:05 INFO - #09: USER32 + 0x1c4e7 21:25:05 INFO - #10: USER32 + 0x15f9f 21:25:05 INFO - #11: USER32 + 0x14f0e 21:25:05 INFO - #12: USER32 + 0x14f7d 21:25:05 INFO - #13: ntdll + 0x46fee 21:25:05 INFO - #14: nsWindow::Show(bool) [widget/windows/nsWindow.cpp:1266] 21:25:05 INFO - #15: nsXULWindow::SetVisibility(bool) [xpfe/appshell/nsXULWindow.cpp:820] 21:25:05 INFO - #16: nsXULWindow::OnChromeLoaded() [xpfe/appshell/nsXULWindow.cpp:1025] 21:25:05 INFO - #17: nsWebShellWindow::OnStateChange(nsIWebProgress *,nsIRequest *,unsigned int,nsresult) [xpfe/appshell/nsWebShellWindow.cpp:584] 21:25:05 INFO - #18: nsDocLoader::DoFireOnStateChange(nsIWebProgress * const,nsIRequest * const,int &,nsresult) [uriloader/base/nsDocLoader.cpp:1250] 21:25:05 INFO - #19: nsDocLoader::doStopDocumentLoad(nsIRequest *,nsresult) [uriloader/base/nsDocLoader.cpp:842] 21:25:05 INFO - #20: nsDocLoader::DocLoaderIsEmpty(bool) [uriloader/base/nsDocLoader.cpp:723] 21:25:05 INFO - #21: nsDocLoader::ChildDoneWithOnload(nsIDocumentLoader *) [uriloader/base/nsDocLoader.h:201] 21:25:05 INFO - #22: nsDocLoader::DocLoaderIsEmpty(bool) [uriloader/base/nsDocLoader.cpp:727] 21:25:05 INFO - #23: nsDocLoader::OnStopRequest(nsIRequest *,nsISupports *,nsresult) [uriloader/base/nsDocLoader.cpp:609] 21:25:05 INFO - #24: nsLoadGroup::RemoveRequest(nsIRequest *,nsISupports *,nsresult) [netwerk/base/nsLoadGroup.cpp:640] 21:25:05 INFO - #25: nsDocument::DoUnblockOnload() [dom/base/nsDocument.cpp:9070] 21:25:05 INFO - #26: nsDocument::UnblockOnload(bool) [dom/base/nsDocument.cpp:8997] 21:25:05 INFO - #27: nsDocument::DispatchContentLoadedEvents() [dom/base/nsDocument.cpp:5164] 21:25:05 INFO - #28: nsRunnableMethodImpl<void ( nsDocument::*)(void),1>::Run() [xpcom/glue/nsThreadUtils.h:870] 21:25:05 INFO - #29: nsThread::ProcessNextEvent(bool,bool *) [xpcom/threads/nsThread.cpp:960] 21:25:05 INFO - #30: NS_ProcessNextEvent(nsIThread *,bool) [xpcom/glue/nsThreadUtils.cpp:277] 21:25:05 INFO - #31: mozilla::ipc::MessagePump::Run(base::MessagePump::Delegate *) [ipc/glue/MessagePump.cpp:95] 21:25:05 INFO - #32: MessageLoop::RunInternal() [ipc/chromium/src/base/message_loop.cc:234] 21:25:05 INFO - #33: MessageLoop::RunHandler() [ipc/chromium/src/base/message_loop.cc:228] 21:25:05 INFO - #34: MessageLoop::Run() [ipc/chromium/src/base/message_loop.cc:202] 21:25:05 INFO - #35: nsBaseAppShell::Run() [widget/nsBaseAppShell.cpp:158] 21:25:05 INFO - #36: nsAppShell::Run() [widget/windows/nsAppShell.cpp:180] 21:25:05 INFO - #37: nsAppStartup::Run() [toolkit/components/startup/nsAppStartup.cpp:282] 21:25:05 INFO - #38: XREMain::XRE_mainRun() [toolkit/xre/nsAppRunner.cpp:4298] 21:25:05 INFO - #39: XREMain::XRE_main(int,char * * const,nsXREAppData const *) [toolkit/xre/nsAppRunner.cpp:4391] 21:25:05 INFO - #40: XRE_main [toolkit/xre/nsAppRunner.cpp:4493] 21:25:05 INFO - #41: do_main [browser/app/nsBrowserApp.cpp:212] 21:25:05 INFO - #42: NS_internal_main(int,char * *) [browser/app/nsBrowserApp.cpp:399] 21:25:05 INFO - #43: wmain [toolkit/xre/nsWindowsWMain.cpp:138] 21:25:05 INFO - #44: __tmainCRTStartup [f:/dd/vctools/crt/crtw32/startup/crt0.c:255] 21:25:05 INFO - #45: kernel32 + 0x53c45 21:25:05 INFO - #46: ntdll + 0x637f5 21:25:05 INFO - #47: ntdll + 0x637c8 21:25:05 INFO - ++DOMWINDOW == 114 (0EAC5800) [pid = 1260] [serial = 114] [outer = 0E905880] 21:25:05 INFO - MEMORY STAT | vsize 691MB | vsizeMaxContiguous 490MB | residentFast 293MB | heapAllocated 54MB 21:25:05 INFO - [1260] WARNING: getting z level of unregistered window: file c:/builds/moz2_slave/try-w32-d-00000000000000000000/build/src/xpfe/appshell/nsWindowMediator.cpp, line 585 21:25:05 INFO - [1260] WARNING: getting z level of unregistered window: file c:/builds/moz2_slave/try-w32-d-00000000000000000000/build/src/xpfe/appshell/nsWindowMediator.cpp, line 585 21:25:05 INFO - --DOCSHELL 123EFD00 == 18 [pid = 1260] [id = 30] 21:25:05 INFO - 2799 INFO TEST-OK | js/xpconnect/tests/chrome/test_bug618176.xul | took 259ms 21:25:05 INFO - ++DOMWINDOW == 115 (0E903800) [pid = 1260] [serial = 115] [outer = 1C35CD00] 21:25:05 INFO - 2800 INFO TEST-UNEXPECTED-ERROR | js/xpconnect/tests/chrome/test_bug618176.xul | Assertion count 1 is greater than expected range 0-0 assertions.

Hi, We are working to get runByDir enabled

Hi, We are working to get runByDir enabled on mochitest-chrome debug and this test has been disabled. Could you please help us in getting this fixed?

(In reply to Kaustabh Datta Choudhury from comment #0) > test_bug618176.xul fails on Windows (7, XP & 8) with runByDir enabled. > Here's a link to try : > https://treeherder.mozilla.org/#/jobs?repo=try&revision=313ffe0c5721&filter- > resultStatus=testfailed&filter-resultStatus=busted&filter- > resultStatus=exception&filter-classifiedState=unclassified&filter- > searchStr=win > > It can be reproduced by : ./mach mochitest js/xpconnect/tests/chrome > > 21:25:05 INFO - 2798 INFO TEST-START | > js/xpconnect/tests/chrome/test_bug618176.xul > 21:25:05 INFO - ++DOMWINDOW == 108 (0D243F80) [pid = 1260] [serial = > 108] [outer = 1C35CD00] > 21:25:05 INFO - ++DOCSHELL 123EFD00 == 18 [pid = 1260] [id = 30] > 21:25:05 INFO - ++DOMWINDOW == 109 (0E901F00) [pid = 1260] [serial = > 109] [outer = 00000000] > 21:25:05 INFO - ++DOMWINDOW == 110 (0E902680) [pid = 1260] [serial = > 110] [outer = 0E901F00] > 21:25:05 INFO - ++DOCSHELL 125FDA00 == 19 [pid = 1260] [id = 31] > 21:25:05 INFO - ++DOMWINDOW == 111 (0E905880) [pid = 1260] [serial = > 111] [outer = 00000000] > 21:25:05 INFO - ++DOMWINDOW == 112 (0EAC3F00) [pid = 1260] [serial = > 112] [outer = 0E905880] > 21:25:05 INFO - ++DOMWINDOW == 113 (0EAC4400) [pid = 1260] [serial = > 113] [outer = 0E905880] > 21:25:05 INFO - [1260] ###!!! ASSERTION: Compositor backend changed.: > 'mCompositorBackend == LayersBackend::LAYERS_NONE', file > c:/builds/moz2_slave/try-w32-d-00000000000000000000/build/src/gfx/thebes/ > gfxPlatform.cpp, line 2211 > 21:25:05 INFO - #01: nsBaseWidget::CreateCompositor(int,int) > [widget/nsBaseWidget.cpp:1136] > 21:25:05 INFO - #02: nsBaseWidget::CreateCompositor() > [widget/nsBaseWidget.cpp:867] > 21:25:05 INFO - #03: > nsWindow::GetLayerManager(mozilla::layers::PLayerTransactionChild > *,mozilla::layers::LayersBackend,nsIWidget::LayerManagerPersistence,bool *) > [widget/windows/nsWindow.cpp:3547] > 21:25:05 INFO - #04: nsWindow::OnPaint(HDC__ *,unsigned int) > [widget/windows/nsWindowGfx.cpp:223] > 21:25:05 INFO - #05: nsWindow::ProcessMessage(unsigned int,unsigned int > &,long &,long *) [widget/windows/nsWindow.cpp:5073] > 21:25:05 INFO - #06: nsWindow::WindowProcInternal(HWND__ *,unsigned > int,unsigned int,long) [widget/windows/nsWindow.cpp:4612] > 21:25:05 INFO - #07: CallWindowProcCrashProtected > [xpcom/base/nsCrashOnException.cpp:35] > 21:25:05 INFO - #08: nsWindow::WindowProc(HWND__ *,unsigned > int,unsigned int,long) [widget/windows/nsWindow.cpp:4564] > 21:25:05 INFO - #09: USER32 + 0x1c4e7 > 21:25:05 INFO - #10: USER32 + 0x15f9f > 21:25:05 INFO - #11: USER32 + 0x14f0e > 21:25:05 INFO - #12: USER32 + 0x14f7d > 21:25:05 INFO - #13: ntdll + 0x46fee > 21:25:05 INFO - #14: nsWindow::Show(bool) > [widget/windows/nsWindow.cpp:1266] > 21:25:05 INFO - #15: nsXULWindow::SetVisibility(bool) > [xpfe/appshell/nsXULWindow.cpp:820] > 21:25:05 INFO - #16: nsXULWindow::OnChromeLoaded() > [xpfe/appshell/nsXULWindow.cpp:1025] > 21:25:05 INFO - #17: nsWebShellWindow::OnStateChange(nsIWebProgress > *,nsIRequest *,unsigned int,nsresult) > [xpfe/appshell/nsWebShellWindow.cpp:584] > 21:25:05 INFO - #18: nsDocLoader::DoFireOnStateChange(nsIWebProgress * > const,nsIRequest * const,int &,nsresult) > [uriloader/base/nsDocLoader.cpp:1250] > 21:25:05 INFO - #19: nsDocLoader::doStopDocumentLoad(nsIRequest > *,nsresult) [uriloader/base/nsDocLoader.cpp:842] > 21:25:05 INFO - #20: nsDocLoader::DocLoaderIsEmpty(bool) > [uriloader/base/nsDocLoader.cpp:723] > 21:25:05 INFO - #21: nsDocLoader::ChildDoneWithOnload(nsIDocumentLoader > *) [uriloader/base/nsDocLoader.h:201] > 21:25:05 INFO - #22: nsDocLoader::DocLoaderIsEmpty(bool) > [uriloader/base/nsDocLoader.cpp:727] > 21:25:05 INFO - #23: nsDocLoader::OnStopRequest(nsIRequest > *,nsISupports *,nsresult) [uriloader/base/nsDocLoader.cpp:609] > 21:25:05 INFO - #24: nsLoadGroup::RemoveRequest(nsIRequest > *,nsISupports *,nsresult) [netwerk/base/nsLoadGroup.cpp:640] > 21:25:05 INFO - #25: nsDocument::DoUnblockOnload() > [dom/base/nsDocument.cpp:9070] > 21:25:05 INFO - #26: nsDocument::UnblockOnload(bool) > [dom/base/nsDocument.cpp:8997] > 21:25:05 INFO - #27: nsDocument::DispatchContentLoadedEvents() > [dom/base/nsDocument.cpp:5164] > 21:25:05 INFO - #28: nsRunnableMethodImpl<void ( > nsDocument::*)(void),1>::Run() [xpcom/glue/nsThreadUtils.h:870] > 21:25:05 INFO - #29: nsThread::ProcessNextEvent(bool,bool *) > [xpcom/threads/nsThread.cpp:960] > 21:25:05 INFO - #30: NS_ProcessNextEvent(nsIThread *,bool) > [xpcom/glue/nsThreadUtils.cpp:277] > 21:25:05 INFO - #31: > mozilla::ipc::MessagePump::Run(base::MessagePump::Delegate *) > [ipc/glue/MessagePump.cpp:95] > 21:25:05 INFO - #32: MessageLoop::RunInternal() > [ipc/chromium/src/base/message_loop.cc:234] > 21:25:05 INFO - #33: MessageLoop::RunHandler() > [ipc/chromium/src/base/message_loop.cc:228] > 21:25:05 INFO - #34: MessageLoop::Run() > [ipc/chromium/src/base/message_loop.cc:202] > 21:25:05 INFO - #35: nsBaseAppShell::Run() > [widget/nsBaseAppShell.cpp:158] > 21:25:05 INFO - #36: nsAppShell::Run() > [widget/windows/nsAppShell.cpp:180] > 21:25:05 INFO - #37: nsAppStartup::Run() > [toolkit/components/startup/nsAppStartup.cpp:282] > 21:25:05 INFO - #38: XREMain::XRE_mainRun() > [toolkit/xre/nsAppRunner.cpp:4298] > 21:25:05 INFO - #39: XREMain::XRE_main(int,char * * const,nsXREAppData > const *) [toolkit/xre/nsAppRunner.cpp:4391] > 21:25:05 INFO - #40: XRE_main [toolkit/xre/nsAppRunner.cpp:4493] > 21:25:05 INFO - #41: do_main [browser/app/nsBrowserApp.cpp:212] > 21:25:05 INFO - #42: NS_internal_main(int,char * *) > [browser/app/nsBrowserApp.cpp:399] > 21:25:05 INFO - #43: wmain [toolkit/xre/nsWindowsWMain.cpp:138] > 21:25:05 INFO - #44: __tmainCRTStartup > [f:/dd/vctools/crt/crtw32/startup/crt0.c:255] > 21:25:05 INFO - #45: kernel32 + 0x53c45 > 21:25:05 INFO - #46: ntdll + 0x637f5 > 21:25:05 INFO - #47: ntdll + 0x637c8 > 21:25:05 INFO - ++DOMWINDOW == 114 (0EAC5800) [pid = 1260] [serial = > 114] [outer = 0E905880] > 21:25:05 INFO - MEMORY STAT | vsize 691MB | vsizeMaxContiguous 490MB | > residentFast 293MB | heapAllocated 54MB > 21:25:05 INFO - [1260] WARNING: getting z level of unregistered window: > file > c:/builds/moz2_slave/try-w32-d-00000000000000000000/build/src/xpfe/appshell/ > nsWindowMediator.cpp, line 585 > 21:25:05 INFO - [1260] WARNING: getting z level of unregistered window: > file > c:/builds/moz2_slave/try-w32-d-00000000000000000000/build/src/xpfe/appshell/ > nsWindowMediator.cpp, line 585 > 21:25:05 INFO - --DOCSHELL 123EFD00 == 18 [pid = 1260] [id = 30] > 21:25:05 INFO - 2799 INFO TEST-OK | > js/xpconnect/tests/chrome/test_bug618176.xul | took 259ms > 21:25:05 INFO - ++DOMWINDOW == 115 (0E903800) [pid = 1260] [serial = > 115] [outer = 1C35CD00] > 21:25:05 INFO - 2800 INFO TEST-UNEXPECTED-ERROR | > js/xpconnect/tests/chrome/test_bug618176.xul | Assertion count 1 is greater > than expected range 0-0 assertions. This has been affecting the Windows debug platforms

This looks like a graphic layers assertion. It looks like the assertion was added by dvander in bug 1179051.

Well, that's weird. I can reproduce this, IDXGIFactory::CreateSwapChain fails with E_INVALIDARG in CompositorD3D11 and we fall back to the basic compositor. Later we get D3D11 again. A while back we decided we have to support mixing d3d11/basic compositors, so maybe now is a good time to nuke this assert and code that depends on a static compositor type. However I'm curious why this call would randomly fail.

(In reply to David Anderson [:dvander] from comment #5) > Well, that's weird. I can reproduce this, IDXGIFactory::CreateSwapChain > fails with E_INVALIDARG in CompositorD3D11 and we fall back to the basic > compositor. Later we get D3D11 again. > > A while back we decided we have to support mixing d3d11/basic compositors, > so maybe now is a good time to nuke this assert and code that depends on a > static compositor type. However I'm curious why this call would randomly > fail. We already support 'mixing' them in certain scenarios, what scenario are you suggesting we add? I'm also uncertain about why it would fail, what does 'runByDir' do?

runbydir creates a fresh profile, fresh browser, runs just the tests in the given directory (i.e. js/xpconnect/tests/chrome), then terminates the browser. common failures when doing this: * tests are run before components are initialized * we shutdown the browser before all components are cleaned up * tests depend on state/prefs/etc. from previous tests in directories that run prior but now don't why we do run-by-dir: * reduce intermittents * allow for flexible scheduling to get results out faster on try/integration branches

This test has been disabled on windows-debug to get run-by-dir enabled : https://dxr.mozilla.org/mozilla-central/source/js/xpconnect/tests/chrome/chrome.ini#32 This can be tested in the try server and turned back on, once fixed.

This test shouldn't be failing anymore...

pushed to try to find out: https://treeherder.mozilla.org/#/jobs?repo=try&revision=2e56af22487a

ok, try server looks great, lets enable

You can probably re-enable test_bug799348.xul (bug 1210876) at the same time, as it seemed to be having the same problem.

Comment on attachment 8673322 [details] [diff] [review] enable test case! Review of attachment 8673322 [details] [diff] [review]: ----------------------------------------------------------------- Sweet - there was another one with the same assert I saw get disabled, but I can't find the bug # in my inbox anymore. But the assert is gone now so any tests that were failing like this one should be fair game again.

Comment on attachment 8673322 [details] [diff] [review] enable test case! Review of attachment 8673322 [details] [diff] [review]: ----------------------------------------------------------------- Sweet - there was another one with the same assert I saw get disabled, but I can't find the bug # in my inbox anymore. But the assert is gone now so any tests that were failing like this one should be fair game again.

Not sure why Bugzilla posted that twice :( but looks like comment #12 has the other bug.