[Webkit-unassigned] [Bug 155136] New: Speedometer benchmark test crashes on trunk.

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Mon Mar 7 14:33:28 PST 2016


https://bugs.webkit.org/show_bug.cgi?id=155136

            Bug ID: 155136
           Summary: Speedometer benchmark test crashes on trunk.
    Classification: Unclassified
           Product: WebKit
           Version: WebKit Nightly Build
          Hardware: Unspecified
                OS: Unspecified
            Status: NEW
          Severity: Normal
          Priority: P2
         Component: JavaScriptCore
          Assignee: webkit-unassigned at lists.webkit.org
          Reporter: clopez at igalia.com

Created attachment 273213
  --> https://bugs.webkit.org/attachment.cgi?id=273213&action=review
GDB backtrace for the GTK port on Linux (Debug build on r197699).

With current trunk (r197699) if you try to run the Speedometer test <http://browserbench.org/Speedometer/> it will crash.
With the release build only happens sometimes, with the debug one seems more easy to reproduce.

Is reproducible at least both in the Mac and GTK ports.

With the debug build I get the following assertion:

ASSERTION FAILED: m_structure.isClear()
/Users/clopez/webkit/Source/JavaScriptCore/dfg/DFGAbstractValue.cpp(472) : void JSC::DFG::AbstractValue::checkConsistency() const
1   0x110245ca0 WTFCrash
2   0x10f653cc1 JSC::DFG::AbstractValue::checkConsistency() const
3   0x10f654ebb JSC::DFG::AbstractValue::normalizeClarity()
4   0x10f654f98 JSC::DFG::AbstractValue::filter(unsigned int)
5   0x10f6f77d9 JSC::DFG::AbstractInterpreter<JSC::DFG::InPlaceAbstractState>::filter(JSC::DFG::AbstractValue&, unsigned int)
6   0x10f6f7603 JSC::DFG::AbstractInterpreter<JSC::DFG::InPlaceAbstractState>::filterByType(JSC::DFG::Edge&, unsigned int)
7   0x10f6f7505 JSC::DFG::AbstractInterpreter<JSC::DFG::InPlaceAbstractState>::filterEdgeByUse(JSC::DFG::Edge&)
8   0x10f6f7451 JSC::DFG::AbstractInterpreter<JSC::DFG::InPlaceAbstractState>::filterEdgeByUse(JSC::DFG::Node*, JSC::DFG::Edge&)
9   0x10f6e90fe JSC::DFG::AbstractInterpreter<JSC::DFG::InPlaceAbstractState>::executeEdges(JSC::DFG::Node*)
10  0x10f6e743b JSC::DFG::AbstractInterpreter<JSC::DFG::InPlaceAbstractState>::execute(unsigned int)
11  0x10f6e6527 JSC::DFG::CFAPhase::performBlockCFA(JSC::DFG::BasicBlock*)
12  0x10f6e5f77 JSC::DFG::CFAPhase::performForwardCFA()
13  0x10f6e5a7f JSC::DFG::CFAPhase::run()
14  0x10f6e52f5 bool JSC::DFG::runAndLog<JSC::DFG::CFAPhase>(JSC::DFG::CFAPhase&)
15  0x10f6e527e bool JSC::DFG::runPhase<JSC::DFG::CFAPhase>(JSC::DFG::Graph&)
16  0x10f6e5238 JSC::DFG::performCFA(JSC::DFG::Graph&)
17  0x10f8fec3f JSC::DFG::Plan::compileInThreadImpl(JSC::DFG::LongLivedState&)
18  0x10f8fdd7d JSC::DFG::Plan::compileInThread(JSC::DFG::LongLivedState&, JSC::DFG::ThreadData*)
19  0x10fa15ada JSC::DFG::Worklist::runThread(JSC::DFG::ThreadData*)
20  0x10fa13e34 JSC::DFG::Worklist::threadFunction(void*)
21  0x1102b0dc9 WTF::createThread(void (*)(void*), void*, char const*)::$_0::operator()() const
22  0x1102b0d9d void std::__1::__invoke_void_return_wrapper<void>::__call<WTF::createThread(void (*)(void*), void*, char const*)::$_0&>(WTF::createThread(void (*)(void*), void*, char const*)::$_0&&&)
23  0x1102b0d3c std::__1::__function::__func<WTF::createThread(void (*)(void*), void*, char const*)::$_0, std::__1::allocator<WTF::createThread(void (*)(void*), void*, char const*)::$_0>, void ()>::operator()()
24  0x10fb5e74a std::__1::function<void ()>::operator()() const
25  0x1102afa9e WTF::threadEntryPoint(void*)
26  0x1102b1331 WTF::wtfThreadEntryPoint(void*)
27  0x7fff9358d05a _pthread_body
28  0x7fff9358cfd7 _pthread_body
29  0x7fff9358a3ed thread_start


The above backtrace is from Safari running on MacOS 10.10 with the WebKit Debug build on r197699.

I'm ataching also a GDB backtrace I got for this very same crash but on the GTK port on Linux (Debug build).

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.webkit.org/pipermail/webkit-unassigned/attachments/20160307/f22b61b5/attachment.html>


More information about the webkit-unassigned mailing list