fast/workers/worker-cloneport.html crashed on Qt Linux Release Bot http://build.webkit.org/results/Qt%20Linux%20Release/r53236%20(5961)/fast/workers/worker-cloneport-stderr.txt Segmentation fault 0: /home/webkit/buildslave/qt-linux-release/build/WebKitBuild/Release/bin/DumpRenderTree [0x8065ab5] 1: /home/webkit/buildslave/qt-linux-release/build/WebKitBuild/Release/bin/DumpRenderTree [0x8065e38] 2: [0xb7eee400] 3: /home/webkit/buildslave/qt-linux-release/build/WebKitBuild/Release/lib/libQtWebKit.so.4 [0xb6f0186e] 4: /home/webkit/buildslave/qt-linux-release/build/WebKitBuild/Release/lib/libQtWebKit.so.4 [0xb6f01bad] 5: /home/webkit/buildslave/qt-linux-release/build/WebKitBuild/Release/lib/libQtWebKit.so.4 [0xb6f01bf5] 6: /home/webkit/buildslave/qt-linux-release/build/WebKitBuild/Release/lib/libQtWebKit.so.4 [0xb6f01c55] 7: /home/webkit/buildslave/qt-linux-release/build/WebKitBuild/Release/lib/libQtWebKit.so.4 [0xb6f01ff2] 8: /home/webkit/buildslave/qt-linux-release/build/WebKitBuild/Release/lib/libQtWebKit.so.4 [0xb6f02089] 9: /home/webkit/buildslave/qt-linux-release/build/WebKitBuild/Release/lib/libQtWebKit.so.4 [0xb6f5d6a6] 10: /home/webkit/buildslave/qt-linux-release/build/WebKitBuild/Release/lib/libQtWebKit.so.4 [0xb6ee0f30] 11: [0xb2cf22c5] 12: /home/webkit/buildslave/qt-linux-release/build/WebKitBuild/Release/lib/libQtWebKit.so.4 [0xb6ea034f] 13: /home/webkit/buildslave/qt-linux-release/build/WebKitBuild/Release/lib/libQtWebKit.so.4 [0xb6f0337c] 14: /home/webkit/buildslave/qt-linux-release/build/WebKitBuild/Release/lib/libQtWebKit.so.4 [0xb78e562b] 15: /home/webkit/buildslave/qt-linux-release/build/WebKitBuild/Release/lib/libQtWebKit.so.4 [0xb78e59d5] 16: /home/webkit/buildslave/qt-linux-release/build/WebKitBuild/Release/lib/libQtWebKit.so.4 [0xb78f2181] 17: /home/webkit/buildslave/qt-linux-release/build/WebKitBuild/Release/lib/libQtWebKit.so.4 [0xb6f8bd1f] 18: /home/webkit/buildslave/qt-linux-release/build/WebKitBuild/Release/lib/libQtWebKit.so.4 [0xb6f8a903] 19: /usr/local/Trolltech/Qt-4.6.0/lib/libQtCore.so.4 [0xb5eab8de] 20: /lib/i686/cmov/libpthread.so.0 [0xb5e284c0] 21: /lib/i686/cmov/libc.so.6(clone+0x5e) [0xb5c866de] Maybe someone could show me how I might symbolicate these backtraces? I would expect we have symbol files for these builds somewhere? Certainly we could modify the bots to symbolicate the crash traces optionally when crashing? I'll file a bug. This test is from Drew Wilson, 3 months ago: http://trac.webkit.org/browser/trunk/LayoutTests/fast/workers/worker-cloneport.html
I'll try to reproduce it in debug mode and send a backtrace.
Unfortunately I can't reproduce the crash in debug mode with "WebKitTools/Scripts/run-webkit-tests --debug fast/workers/worker-cloneport.html --iterations 500" I will run a longer test at night.
I just saw this time out on the Leopard Mac commit-queue bot.
Crashed again: http://build.webkit.org/results/Qt%20Linux%20Release/r56604%20(9132)/fast/workers/worker-cloneport-stderr.txt I recommend we skip this test to keep the bots green if we don't have a fix.
Actually, I bet this isn't even caused by this test. Here is another crashing workers test, likely related to this crash, fast/workers/use-machine-stack.html: http://build.webkit.org/results/Qt%20Linux%20Release/r56600%20(9129)/fast/workers/use-machine-stack-stderr.txt
More worker crashes: http://build.webkit.org/results/Qt%20Linux%20Release/r56598%20(9127)/fast/workers/worker-cloneport-stderr.txt http://build.webkit.org/results/Qt%20Linux%20Release/r56598%20(9127)/fast/workers/worker-gc-stderr.txt
Another crash: http://build.webkit.org/results/Qt%20Linux%20Release/r56633%20(9162)/fast/workers/worker-cloneport-stderr.txt
Crashed on Gtk Linux too! http://build.webkit.org/results/GTK%20Linux%2032-bit%20Release/r56748%20(10528)/results.html
Now that Qt is marked as a core builder, this crasher will hold up the commit-queue. :( Happened again just now: http://build.webkit.org/results/Qt%20Linux%20Release/r57568%20(10061)/fast/workers/worker-cloneport-stderr.txt
(In reply to comment #9) > Now that Qt is marked as a core builder, this crasher will hold up the > commit-queue. :( > > Happened again just now: > http://build.webkit.org/results/Qt%20Linux%20Release/r57568%20(10061)/fast/workers/worker-cloneport-stderr.txt :( We are green again, but we should fix it asap. I'm going to reproduce all crashing workers test in debug mode and send a symbolicated backtrace.
fails at the very same point as bug 33008. run-webkit-tests --qt fast/workers/worker-cloneport.html output: ASSERTION FAILED: isMainThread() /home/agostini/work/webkit/workdirs/wd5/Source/JavaScriptCore/wtf/CryptographicallyRandomNumber.cpp(146) : uint32_t WTF::<unnamed>::ARC4RandomNumberGenerator::randomNumber() Segmentation fault 0: /home/agostini/work/webkit/build/workdir/wd5/Release/bin/DumpRenderTree() [0x807abfd] 1: /home/agostini/work/webkit/build/workdir/wd5/Release/bin/DumpRenderTree() [0x807aea3] 2: [0xb7890400] 3: /home/agostini/work/webkit/build/workdir/wd5/Debug/lib/libQtWebKit.so.4(+0x20ed770) [0xb65a4770] 4: /home/agostini/work/webkit/build/workdir/wd5/Debug/lib/libQtWebKit.so.4(+0x20ed921) [0xb65a4921] 5: /home/agostini/work/webkit/build/workdir/wd5/Debug/lib/libQtWebKit.so.4(+0x20feb6b) [0xb65b5b6b] 6: /home/agostini/work/webkit/build/workdir/wd5/Debug/lib/libQtWebKit.so.4(_ZN3JSC14JSGlobalObject18JSGlobalObjectDataC2EPFvPvE+0x2c9) [0xb5a47651] 7: /home/agostini/work/webkit/build/workdir/wd5/Debug/lib/libQtWebKit.so.4(_ZN7WebCore17JSDOMGlobalObject21JSDOMGlobalObjectDataC1EPNS_15DOMWrapperWorldEPFvPvE+0x26) [0xb623a74c] 8: /home/agostini/work/webkit/build/workdir/wd5/Debug/lib/libQtWebKit.so.4(_ZN7WebCore19JSWorkerContextBaseC2EN3WTF17NonNullPassRefPtrIN3JSC9StructureEEENS1_10PassRefPtrINS_13WorkerContextEEE+0x66) [0xb623a87a] 9: /home/agostini/work/webkit/build/workdir/wd5/Debug/lib/libQtWebKit.so.4(_ZN7WebCore15JSWorkerContextC2EN3WTF17NonNullPassRefPtrIN3JSC9StructureEEENS1_10PassRefPtrINS_13WorkerContextEEE+0x51) [0xb5911f3f] 10: /home/agostini/work/webkit/build/workdir/wd5/Debug/lib/libQtWebKit.so.4(_ZN7WebCore24JSDedicatedWorkerContextC1EN3WTF17NonNullPassRefPtrIN3JSC9StructureEEENS1_10PassRefPtrINS_22DedicatedWorkerContextEEE+0x51) [0xb590c785] 11: /home/agostini/work/webkit/build/workdir/wd5/Debug/lib/libQtWebKit.so.4(_ZN7WebCore22WorkerScriptController10initScriptEv+0x430) [0xb623c87e] 12: /home/agostini/work/webkit/build/workdir/wd5/Debug/lib/libQtWebKit.so.4(_ZN7WebCore22WorkerScriptController18initScriptIfNeededEv+0x2f) [0xb5a4465f] 13: /home/agostini/work/webkit/build/workdir/wd5/Debug/lib/libQtWebKit.so.4(_ZN7WebCore22WorkerScriptController8evaluateERKNS_16ScriptSourceCodeEPNS_11ScriptValueE+0x8b) [0xb623d16d] 14: /home/agostini/work/webkit/build/workdir/wd5/Debug/lib/libQtWebKit.so.4(_ZN7WebCore22WorkerScriptController8evaluateERKNS_16ScriptSourceCodeE+0xa8) [0xb623cfd0] 15: /home/agostini/work/webkit/build/workdir/wd5/Debug/lib/libQtWebKit.so.4(_ZN7WebCore12WorkerThread12workerThreadEv+0x183) [0xb6252ec9] 16: /home/agostini/work/webkit/build/workdir/wd5/Debug/lib/libQtWebKit.so.4(_ZN7WebCore12WorkerThread17workerThreadStartEPv+0x1d) [0xb6252d3f] 17: /home/agostini/work/webkit/build/workdir/wd5/Debug/lib/libQtWebKit.so.4(+0x210126e) [0xb65b826e] 18: /home/agostini/work/webkit/build/workdir/wd5/Debug/lib/libQtWebKit.so.4(+0x20fba23) [0xb65b2a23] 19: /home/agostini/work/qt/build/lib/libQtCore.so.4(+0x7d0f7) [0xb33000f7] 20: /lib/libpthread.so.0(+0x5cc9) [0xb326dcc9] 21: /lib/libc.so.6(clone+0x5e) [0xb30ae69e] LEAK: 13 WebCoreNode LEAK: 1 Frame LEAK: 1 Page LEAK: 6 RenderObject LEAK: 109 Structure
(In reply to comment #11) > fails at the very same point as bug 33008. > > run-webkit-tests --qt fast/workers/worker-cloneport.html output: > > ASSERTION FAILED: isMainThread() > /home/agostini/work/webkit/workdirs/wd5/Source/JavaScriptCore/wtf/CryptographicallyRandomNumber.cpp(146) : uint32_t WTF::<unnamed>::ARC4RandomNumberGenerator::randomNumber() If qt is hitting that assert, then it appears that it is has not defined JSC_MULTIPLE_THREADS. I suspect this could result in a fair amount of instability when running workers.
(In reply to comment #12) > (In reply to comment #11) > > fails at the very same point as bug 33008. > > > > run-webkit-tests --qt fast/workers/worker-cloneport.html output: > > > > ASSERTION FAILED: isMainThread() > > /home/agostini/work/webkit/workdirs/wd5/Source/JavaScriptCore/wtf/CryptographicallyRandomNumber.cpp(146) : uint32_t WTF::<unnamed>::ARC4RandomNumberGenerator::randomNumber() > > > If qt is hitting that assert, then it appears that it is has not defined JSC_MULTIPLE_THREADS. I suspect this could result in a fair amount of instability when running workers. yes. Some mutex are just not created/used. JavaScriptCore/wtf/CryptographicallyRandomNumber.cpp line 69: #if ENABLE(JSC_MULTIPLE_THREADS) Mutex m_mutex; #endif
seems to be the same problem of bug 33008.
*** This bug has been marked as a duplicate of bug 33008 ***