Summary: Bad rewiring of memory edges when we split unique types during EA
Reviewed-by: kvn

Summary: Clean ICs of unloaded nmethods in sweeper to avoid impact on safepoint duration.
Reviewed-by: kvn, mdoerr

Summary: Removed nmethod transition to zombie outside of sweeper. Added cleaning of ICs of unloaded nmethods.
Reviewed-by: kvn, iveresov

Summary: Remove active ICStubs from zombie nmethods
Reviewed-by: kvn, iveresov

Reviewed-by: jwilhelm, jmasa

Reviewed-by: kvn, vlivanov
Contributed-by: vlad.ureche@gmail.com

Summary: Updated comments to refer to relevant patents.
Reviewed-by: dcubed

Reviewed-by: coleenp, dcubed

Summary: Add SO_AllCodeCache to root scanning options when not unloading classes with a CMS collection cycle
Reviewed-by: mgerdin, jwilhelm

Summary: Fix up GPL headers so that the RE script works.
Reviewed-by: stefank, dholmes, coleenp

Summary: CMSClassUnloadingEnabled and ExplicitGCInvokesConcurrentAndUnloadsClasses should be disabled when -Xnoclassgc is specified
Reviewed-by: jmasa, kbarrett

Reviewed-by: stefank, tschatzl, thartmann

Reviewed-by: kvn

Summary: metaspace/shrink_grow/CompressedClassSpaceSize fails with OOM: Compressed class space
Reviewed-by: jmasa, kbarrett

Summary: Node::depends_only_on_test() should return false for Unsafe loads
Reviewed-by: kvn, adinn

Summary: CmpU needs to be reprocessed by CCP when an AddI/SubI input's input type change
Reviewed-by: mcberg, kvn, roland
Contributed-by: andreas.eriksson@oracle.com

Summary: the cached class file structure must be deallocated instead of the cached class file bytes
Reviewed-by: coleenp, sla

8061715: gc/g1/TestShrinkAuxiliaryData15.java fails with java.lang.RuntimeException: heap decommit failed - after > before
Summary: added WhiteBox methods to count regions and exact aux data sizes
Reviewed-by: jwilhelm, brutisso

Summary: Prevent vectorization of memory operations with different invariant offsets if unaligned memory accesses are not allowed.
Reviewed-by: kvn

Summary: C2 should not try to vectorize loops with loop variant vector base address.
Reviewed-by: kvn

Summary: Don't assume SATB buffer entries are valid objects
Reviewed-by: brutisso, ecaspole

Summary: Null the InterpreterRuntime::member_name_arg_or_null return value when it is necessary
Reviewed-by: jbachorik, coleenp, twisti

Reviewed-by: kvn, iignatyev

Summary: Optimize the method pointer adjustments for prev klass versions and MNT
Reviewed-by: dcubed, coleenp

Reviewed-by: stuefe, dsamersoff, dcubed

Summary: NMT with detail option reports incorrect caller address on Linux.
Reviewed-by: dholmes, coleenp

Summary: Removed no longer valid checking of SATB buffers
Reviewed-by: jmasa, pliden

Summary: Apply closure directly, eliminating registration.
Reviewed-by: stefank, tschatzl

Summary: Removed the Solaris specific conditionalization for casting to void * within calls to HS_DTRACE_PROBE* to enable successful compilation with gcc Red Hat 4.4.6-4.
Reviewed-by: hseigel, stefank

Summary: Removed pre-filter verify, and made filtering more careful.
Reviewed-by: brutisso, tschatzl

Summary: Prevent eager reclaim of objects that might be on mark stack.
Reviewed-by: brutisso, tschatzl

Reviewed-by: jrose, ahgross, alanb, bmoloden

Reviewed-by: kvn

Summary: Eliminate _CHECK_BOTH_FINGERS_ and simplify.
Reviewed-by: brutisso, tschatzl

Summary: check that offset % mem_oper_size == 0 when alignment is verified during vectorization.
Reviewed-by: iveresov

Summary: meet of 2 constant arrays result in bottom
Reviewed-by: kvn

Summary: remove the assert and deallocate cashed class file bytes that are in collision
Reviewed-by: coleenp, dcubed

Summary: A dead scope  of the local needs to be identified
Reviewed-by: coleenp, vlivanov, mgronlun

Summary: For *store bytecodes, compare incoming, not outgoing, type state with exception handlers' stack maps.
Reviewed-by: acorn, dholmes

Summary: Instead of 'fpclass', use cast float->int and double->long to check if value is +0.0f and +0.0d, respectively.
Reviewed-by: kvn, simonis, dlong