During the nightly gc testing bigapps/Kitchensink/stability hit an assertion:
assert(ft == ttkp->cast_to_ptr_type(jtkp->ptr()) || ft->isa_narrowoop() && ft->make_ptr() == ttkp->cast_to_ptr_type(jtkp->ptr())) failed:

Stacktrace from hs_err:
Current thread (0x690d0c00): JavaThread "C2 CompilerThread0" daemon [_thread_in_native, id=5133, stack(0x69108000,0x69189000)]

Stack: [0x69108000,0x69189000], sp=0x69186040, free space=504k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)
V [libjvm.so+0xda1575] VMError::report_and_die()+0x185;; VMError::report_and_die()+0x185
V [libjvm.so+0x5c1998] report_vm_error(char const*, int, char const*, char const*)+0x68;; report_vm_error(char const*, int, char const*, char const*)+0x68
V [libjvm.so+0x44e899] PhiNode::Value(PhaseTransform*) const+0x9e9;; PhiNode::Value(PhaseTransform*) const+0x9e9
V [libjvm.so+0xbd018e] PhaseCCP::analyze()+0x2be;; PhaseCCP::analyze()+0x2be
V [libjvm.so+0x54346a] Compile::Optimize()+0x58a;; Compile::Optimize()+0x58a
V [libjvm.so+0x54548b] Compile::Compile(ciEnv*, C2Compiler*, ciMethod*, int, bool, bool, bool)+0x18eb;; Compile::Compile(ciEnv*, C2Compiler*, ciMethod*, int, bool, bool, bool)+0x18eb
V [libjvm.so+0x4310fb] C2Compiler::compile_method(ciEnv*, ciMethod*, int)+0x30b;; C2Compiler::compile_method(ciEnv*, ciMethod*, int)+0x30b
V [libjvm.so+0x553649] CompileBroker::invoke_compiler_on_method(CompileTask*)+0xbf9;; CompileBroker::invoke_compiler_on_method(CompileTask*)+0xbf9
V [libjvm.so+0x554e4d] CompileBroker::compiler_thread_loop()+0x61d;; CompileBroker::compiler_thread_loop()+0x61d
V [libjvm.so+0xd1c7bf] compiler_thread_entry(JavaThread*, Thread*)+0x4f;; compiler_thread_entry(JavaThread*, Thread*)+0x4f
V [libjvm.so+0xd2e734] JavaThread::thread_main_inner()+0x234;; JavaThread::thread_main_inner()+0x234
V [libjvm.so+0xd2ead4] JavaThread::run()+0x2e4;; JavaThread::run()+0x2e4
V [libjvm.so+0xb66ad9] java_start(Thread*)+0x119;; java_start(Thread*)+0x119
C [libpthread.so.0+0x596e] start_thread+0xbe

No core-file was generated but I've attached the tail from the log and the hs_err-file.

A quick look at ILW suggests HLH which means P2.