Summary: Fixed a missing null check on n->in(0) found by Parfait.
Reviewed-by: kvn

Summary: Fixed a missing null check on the return value of MachNodeGenerator() found by Parfait.
Reviewed-by: kvn

Summary: Fixed a missing null check on the return value of InitializeNode::allocation() found by Parfait.
Reviewed-by: zmajo

Reviewed-by: kvn

Reviewed-by: kvn, thartmann

Reviewed-by: kvn, thartmann

Reviewed-by: kvn, thartmann

Summary: Mismatched stores on same slice possible with Unsafe.Put*Unaligned methods
Reviewed-by: kvn, thartmann

Summary: fixed the jvms for callsite traps based on declared signature.
Reviewed-by: jcm, coleenp, vlivanov

Summary: Store additional information in PhiNodes corresponding to known instance field values to avoid incorrect reusage.
Reviewed-by: kvn, thartmann

Summary: fix Implicit Null Check optimization code.
Reviewed-by: roland, aph

Summary: don't convert loop with safepoint on the backedge to Counted loop
Reviewed-by: kvn

Reviewed-by: kevinw, vlivanov

Summary: Correct backport of 8159244.
Reviewed-by: kvn

Reviewed-by: kvn

Summary: don't convert loop with safepoint on the backedge to Counted loop
Reviewed-by: kvn

Summary: when speculation fails at a call fallback to profile data at the call site
Reviewed-by: kvn

Summary: Add support for AES intrinsics on PPC64.
Reviewed-by: kvn, mdoerr, simonis, zmajo
Contributed-by: NHiroshi H Horii <horii@jp.ibm.com>

Summary: Emit release barrier after String creation to prevent partially initialized object from escaping.
Reviewed-by: kvn

Summary: Emit release barrier after String creation to prevent partially initialized object from escaping.
Reviewed-by: kvn

Reviewed-by: kvn

8055530: assert(_exits.control()->is_top() || !_gvn.type(ret_phi)->empty()) failed: return value must be well defined
Summary: concurrent class loading causes return phi to become top
Reviewed-by: kvn

Reviewed-by: kvn, coleenp

Reviewed-by: kvn, mdoerr

Reviewed-by: kvn, mdoerr

Reviewed-by: stuefe, coleenp, roland

Summary: Add intrinsics for BigInteger squareToLen and mulAdd methods.
Reviewed-by: kvn, jrose

Summary: Emit CastII to make narrow ConvI2L dependent on the corresponding range check.
Reviewed-by: kvn, roland

Reviewed-by: coleenp, roland

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

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

Reviewed-by: roland

Reviewed-by: kvn

Summary: Add a new compiler phase, PhaseRenumberLive, that renumbers live nodes.
Reviewed-by: kvn, thartmann, vlivanov, shade

Reviewed-by: kvn

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: 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

Reviewed-by: kvn