Summary: The old function not defined properly, need a definition for export in dll. Also changes made to let new jvm work with old hsdis.
Reviewed-by: jrose, sspitsyn, kmo
Contributed-by: yumin.qi@oracle.com

Reviewed-by: jcoomes

Reviewed-by: dholmes, sspitsyn, kmo

Summary: Removes the two unused functions GenericTaskQueueSet::steal_1_random and GenericTaskQueueSet::steal_best_of_all
Reviewed-by: brutisso, stefank
Contributed-by: erik.x.helin@oracle.com

Reviewed-by: johnc, jcoomes, ysr

Reviewed-by: coleenp, dholmes

Summary: Fix the incorrect links in jvmti.html reported by the LinkCheck tool
Reviewed-by: jjh, dholmes
Contributed-by: serguei.spitsyn@oracle.com

Reviewed-by: coleenp, dholmes, twisti
Contributed-by: NRoman Kennke <rkennke@redhat.com>

Reviewed-by: jcoomes

Summary: Added missed check that Load node is in processed loop block.
Reviewed-by: twisti

Summary: In case of time or iterations limit reached C2 stops EA and continue compilation without EA as it does in product VM already.
Reviewed-by: twisti

Summary: Disable -XX:+UseLargePages for MacOS X
Reviewed-by: dholmes, coleenp, sla
Contributed-by: harold.seigel@oracle.com

Summary: Remove support for the UseVectoredExceptions flag
Reviewed-by: jcoomes, kamg
Contributed-by: harold.seigel@oracle.com

Reviewed-by: johnc, tamao

Summary: Use new x86 AES instructions for AESCrypt.
Reviewed-by: twisti, kvn, roland
Contributed-by: tom.deneau@amd.com

Summary: do vector right shift operation for small int types only after loads
Reviewed-by: jrose, dlong

Reviewed-by: kvn

Summary: perform transforms during IGVN phase when Load has a single user.
Reviewed-by: jrose, kvn, twisti

Summary: Enhanced virtual memory tracking to track committed regions as well as reserved regions, so NMT now can generate virtual memory map.
Reviewed-by: acorn, coleenp

Reviewed-by: jcoomes

Summary: Allow which vector rules are supported to be decided at runtime. Also a small change to allow vector types in Type::_type_info[] to apply to more platforms.
Reviewed-by: kvn, twisti
Contributed-by: dean.long@oracle.com

Summary: This is a mistake change in 6879063 about Bytes.swap. Java byte code order always follows big endian, but in that change, assume they follow native platform order that is not right.
Reviewed-by: coleenp, sspitsyn, dholmes
Contributed-by: yumin.qi@oracle.com