Reviewed-by: kvn

Summary: Fix incorrect marking of methods as not compilable.
Reviewed-by: never

Summary: MethodHandle invoke call sites should not be fixed-up by SharedRuntime::fixup_callers_callsite as c2i/i2c adapters are used to implement MethodHandle actions.
Reviewed-by: kvn, never

Summary: We need to introduce a new MH deopt handler so we can easily determine if the deopt happened at a MH call site or not.
Reviewed-by: never, jrose

Reviewed-by: kvn, twisti

Summary: Speculatively unload the oldest nmethods when code cache gets full.
Reviewed-by: never, kvn
Contributed-by: eric.caspole@amd.com

Reviewed-by: mr

Summary: We need to introduce a new MH deopt handler so we can easily determine if the deopt happened at a MH call site or not.
Reviewed-by: never, jrose

Reviewed-by: kvn

Reviewed-by: kvn

Summary: Add safepoint after CompileTheWorldSafepointInterval (100) compilations and do full sweep of CodeCache each time.
Reviewed-by: never
Contributed-by: eric.caspole@amd.com

Reviewed-by: kvn, jrose

Reviewed-by: kvn, cfang

Reviewed-by: kvn, twisti

Reviewed-by: twisti, kvn

Summary: When enqueuing a completed PtrQueue buffer, cache a local pointer to the buffer and clear the field in the PtrQueue prior to unlocking the mutex referenced by the _lock field and pass the cached local value to the enqueuing routine. This will prevent the same completed buffer being enqueued multiple times, which causes the hang.
Reviewed-by: ysr

Summary: Deferred card marks are now flushed during the gc prologue. Parallel[Scavege,OldGC] and SerialGC no longer defer card marks generated by COMPILER2 as a result of ReduceInitialCardMarks. For these cases, introduced a diagnostic option to defer the card marks, only for the purposes of testing and diagnostics. CMS and G1 continue to defer card marks. Potential performance concern related to single-threaded flushing of deferred card marks in the gc prologue will be addressed in the future.
Reviewed-by: never, johnc

Summary: What the title says.
Reviewed-by: ysr, jmasa

6580131: 3/4 CompiledMethodLoad events don't produce the expected extra notifications to describe inlining
Summary: Add support for additional implementation specific info to the JVM/TI CompiledMethodLoad event via the compile_info parameter.
Reviewed-by: never, ohair, tbell, tdeneau
Contributed-by: NVasanth Venkatachalam <vasanth.venkatachalam@amd.com>

6902115: G1:assert(ignore_max_completed||thread->is_Java_thread()||SafepointSynchronize::is_at_safepoint())
Summary: Remove invalid assert and mangle filler objects in TLABs that are being retired.
Reviewed-by: ysr, jmasa

Summary: Loop predication implementation
Reviewed-by: never, kvn

Summary: Put all inline-related switches into product mode, to allow tuning by dynamic language implementors.
Reviewed-by: twisti, kvn

Summary: More informative disassembly in product mode.  Also, a more consistent CompileCommand syntax.
Reviewed-by: never

Summary: Remove AND operation only if mask is equal to shift.
Reviewed-by: never

Summary: Check RootNode before check Loop in IfNode::up_one_dom().
Reviewed-by: never
Contributed-by: kevin.brown@sun.com

Summary: Put generated MH adapter in InvokeDynamic/MethodHandle classes to be able to indentify them easily in the compiler.
Reviewed-by: kvn, never, jrose

Summary: Add missing check that value is used in memory expression in instructions with embedded load.
Reviewed-by: never, jrose

Summary: Change the flag -XX:+PrintFlagsFinal to product and add new product flag -XX:+PrintFlagsInitial.
Reviewed-by: phh, ysr
Contributed-by: gbenson@redhat.com

Summary: Block too small to split was not correctly putback to free lists.
Reviewed-by: ysr

Reviewed-by: kvn, twisti

Reviewed-by: kvn, twisti

Summary: C2 needs some additional optimizations to be able to handle MethodHandle invokes and invokedynamic instructions at the best performance.
Reviewed-by: kvn, never

Summary: C2 implementation for invokedynamic support.
Reviewed-by: kvn, never

Summary: EA incorrectly identifies escape state of an allocation passed as call argument.
Reviewed-by: never

6637203: Classunloading messages go to stdout rather than Xloggc file, causing hangs when stdout is closed
Summary: Decoupled TraceClassUnloading from verbose:gc, JVMTI_VERBOSE_GC and PrintGC[Details], making it settable in a manner identical to TraceClassLoading. Reverted an inadvertent change of TraceClassUnloading output in a previous changeset from gclog back to tty.
Reviewed-by: coleenp, dholmes, jmasa, poonam

Summary: During the work for 6829187 we have fixed a number of basic bugs which are logically grouped with 6815692 and 6858164 but which must be reviewed and pushed separately.
Reviewed-by: kvn, never

Summary: We need a way to walk chained method handles in the JVM to call the right methods and to generate required bytecode adapters for the compilers.
Reviewed-by: kvn

Summary: Zero currently won't build on zSeries or PowerPC machines with assertions turned on.
Reviewed-by: twisti
Contributed-by: NGary Benson <gbenson@redhat.com>

Summary: Smaller fixes to ensure that Zero still works with non-standard options.
Reviewed-by: twisti
Contributed-by: NGary Benson <gbenson@redhat.com>

Summary: Autonomic per-worker free block cache sizing, tunable coalition policies, fixes to per-size block statistics, retuned gain and bandwidth of some feedback loop filters to allow quicker reactivity to abrupt changes in ambient demand, and other heuristics to reduce fragmentation of the CMS old gen. Also tightened some assertions, including those related to locking.
Reviewed-by: jmasa